USF Libraries
USF Digital Collections

Perceived workplace discrimination as a mediator of the relationship between work environment and employee outcomes :

MISSING IMAGE

Material Information

Title:
Perceived workplace discrimination as a mediator of the relationship between work environment and employee outcomes : does minority status matter?
Physical Description:
Book
Language:
English
Creator:
Jagusztyn, Nicole
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Ethnic differences
Minority
Health
Stress
Job attitudes
Environment
Dissertations, Academic -- Psychology -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: The purpose of the current study was to explore the role of six organizational factors (Equal Employment Opportunity, minority segmentation, diversity climate, instrumental social support, emotional social support, and token status) in the perception of discrimination in the workplace by minorities and majority-group members. Five outcomes (job satisfaction, organizational commitment, intention to turnover, physical health, and psychological health) were investigated in response to perceived discrimination. Moderated mediation was used to test hypothesis where perceived discrimination mediated the relationship between organizational antecedents and outcomes; minority status served as the moderators. Support for the mediating role of perceived discrimination was found in the relationship between each organizational antecedent and outcome. In each case, poorer environmental conditions related to increased perceived discrimination which in turn related to more negative workplace attitudes and health outcomes. Implications for workplace design are discussed.
Thesis:
Dissertation (PHD)--University of South Florida, 2010.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Nicole Jagusztyn.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains X pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E14-SFE0004642
usfldc handle - e14.4642
System ID:
SFS0027957:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Perceived Workplace Discrimination as a Medi ator of the Relationship between Work Environment and Employee Outcomes : Does Minority Status Matter? by Nicole Ellis Jagusztyn A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Psychology Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Kristen Salomon, Ph.D. Jennifer Bosson, Ph.D. Edward Levine, Ph.D. Stephen Stark, Ph.D. Joseph Vandello, Ph.D. Date of Approval: July 12, 2010 Keywords: ethnic differences, minority, heal th, stress, job attitudes, environment Copyright 2010, Nicole Ellis Jagusztyn

PAGE 2

Acknowledgements I would like to thank my major prof essor, Kristen Salomon, Ph.D., for the guidance and feedback I needed in completing th is project. I would also like to thank my committee for providing the depth and rigor that this project needed to make a valuable contribution to the literature. To Dr. Thomas Bernard, thank you for agreeing to chair my defense on such short notice. Many thanks to my parents for supporting me through twenty-two long years of schooling. Pyrami ds took less time to build. To Jeremy Mackling, I bestow on you “credit” for “motivating” me.

PAGE 3

i Table of Contents List of Tables iii Abstract Chapter One Introduction 1 Overview of the paper 1 Minority Group Membership 1 An Aside: Perceived Sexism/Racism a nd Perceived Discrimination Defined 3 The Existence of Discrimination 3 The Prevalence of Perceived Discrimination 6 An Aside: Overt vs. Subtle Discrimination 7 Perceived Discrimination as a Stressor 8 The Costs of Perceived Discrimination 8 Perceived Discrimination and Psychological Health Outcomes 9 Perceived Discrimination and Physical Health Outcomes 11 Perceived Discrimination and Job Attitudes 12 Organization-Level Antecedents to Perceived Discrimination 13 Equal Employment Opportunity 14 Minority Segmentation 15 Perceived Diversity Climate 17 Instrumental and Emotional Social Support 20 Token Status 22 Summary 26 Hypothesis 1 27 Hypothesis 2 28 Hypothesis 3 29 Hypothesis 4 30 Hypothesis 5 31 Hypothesis 6 31 Chapter Two Method 33 Procedure 33 Participants 34 Measures 38 Demographics 38

PAGE 4

ii Equal Employment Opportunity 38 Perceived Minority Segmentation 39 Diversity Climate 39 Instrumental Social Support 40 Emotional Social Support 40 Token Status 41 Perceived Discrimination at work (overt) 41 Perceived Discrimination at work (subtle) 42 Every-day Perceived Discrimination 42 Physical Health 43 Psychological Health 44 Job Satisfaction 43 Organizational Commitment 43 Intention to Turnover 43 Data Analysis 44 Chapter Three Results 48 Analysis of Pilot Test Data 48 Analysis of Full Demonstration Data 68 Results for Perceived Equal Employment Opportunity/ Diversity Climate 73 Sex-Based Discrimination 73 Race-Based Discrimination 82 Results for Minority Segmentation 92 Sex-Based Discrimination 92 Race-Based Discrimination 102 Results for Social Support 112 Sex-Based Discrimination 112 Race-Based Discrimination 122 Analysis of Coworker Responses 133 Chapter Four Discussion 136 Summary of Resu lts for Equal Employment Opportunity/ Diversity Climate 136 Summary of Results for Minority Segmentation 137 Summary of Results fo r Social Support 140 Sex-Based vs. Race-Based Perceived Discrimination 142 Conclusions 143 Limitations 144 Suggestions for Future Research 146 References 149 Appendix A: Measures 177

PAGE 5

iii Appendix B: Original Version of Full Demonstration Analyses 189 About the Author End Page

PAGE 6

iv List of Tables Table 1 Comparison of Minority and Majority group Members 35 Table 2 Group Comparison on Con tinuous Demographic Variables 36 Table 3 Group Comparison on Categor ical Demographic Variables 37 Table 4 Item Analysis of Job Satisfaction Measure. 49 Table 5 Item Analysis of Organizational Commitment Measure. 50 Table 6 Item Analysis of SF-36 Physical Functioning Sub-scale 51 Table 7 Item Analysis of SF-36 P hysical Role Limitations Scale 52 Table 8 Item Analysis of SF-36 Emotional Role Limitations Scale 52 Table 9 Item Analysis of the SF-36 Energy Sub-scale 53 Table 10 Item Analysis of the SF-36 Emotional Well-being Sub-scale 53 Table 11 Item Analysis of the SF-36 Social Functioning Sub-scale 54 Table 12 Item Analysis of the SF-36 Pain Sub-scale 54 Table 13 Item Analysis of th e SF-36 General Health Sub-scale 54 Table 14 Item Analysis of the In strumental Social Support Scale 55 Table 15 Item Analysis of th e Emotional Social Support Scale 55 Table 16 Item Analysis of the Minority Segmentation Scale 56 Table 17 Item Analysis of the Diversity Climate Scale 57 Table 18 Item Analysis of the Subtle Gender Discrimination Scale 58 Table 19 Item Analysis of the Subtle Race Discrimination at Work Scale 59

PAGE 7

v Table 20 Item Analysis of the Overt Race Discrimination at Work Scale 60 Table 21 Item Analysis of the Over t Gender Discrimination at Work Scale 61 Table 22 Item Analysis of the General Every-day Race Discrimination Scale 62 Table 23 Item Analysis of the Ge neral Every-day Gender Discrimination 63 Table 24 Item Analysis of the Eq ual Employment Opportunity Scale 64 Table 25 Percentage of Tokenism within the Workplace and Work Group 65 Table 26 Correlations Among the Discri mination Measures in Pilot Study 67 Table 27 Correlations Among all Measur es used in Original Analyses 70 Table 28 Correlations Among all Meas ures used in Revised Analyses 72 Table 29 Results of Moderated Me diation (PREDICTOR= EEODIV, Med = Sex-based Discrimination, DV= Job Satisfaction, Mod = Gender) 74 Table 30 Results of Simple Mediat ion for Men and Women (PREDICTOR = EEODIV, Med = Sex-based Discrimination, DV = Job Satisfaction) 75 Table 31 Results of Moderated Me diation (PREDICTOR= EEODIV, Med = Sex-based Discrimination, DV= Organizational Commitment, Mod = Gender) 76 Table 32 Results of Simple Mediat ion for Men and Women (PREDICTOR = EEODIV, Med = Sex-based Discrimination, DV = Organizational Commitment) 76 Table 33 Results of Moderated Me diation (PREDICTOR= EEODIV, Med = Sex-based Discrimination, DV= Intent to Turnover, Mod = Gender) 77 Table 34 Results of Simple Mediat ion for Men and Women (PREDICTOR = EEODIV, Med = Sex-based Di scrimination, DV = Intent to Turnover) 78 Table 35 Results of Moderated Me diation (PREDICTOR= EEODIV, Med = Sex-based Discrimination, DV= Physical Health, Mod = Gender) 79

PAGE 8

vi Table 36 Results of Simple Mediat ion for Men and Women (PREDICTOR = EEODIV, Med = Sex-based Di scrimination, DV = Physical Health) 80 Table 37 Results of Moderated Me diation (PREDICTOR= EEODIV, Med = Sex-based Discrimination, DV= Psychological Health, Mod = Gender) 81 Table 38 Results of Simple Mediat ion for Men and Women (PREDICTOR = EEODIV, Med = Sex-based Discrimination, DV = Psychological Health) 82 Table 39 Results of Moderated Me diation (PREDICTOR= EEODIV, Med = Race-based Discrimination, DV= Job Satisfaction, Mod = Race) 83 Table 40 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = EEODIV, Med = Race-based Discrimination, DV = Job Satisfaction) 84 Table 41 Results of Moderated Me diation (PREDICTOR= EEODIV, Med = Race-based Discrimination, DV= Organizational Commitment, Mod = Race) 85 Table 42 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = EEODIV, Med = Race-based Discrimination, DV = Organizational Commitment) 86 Table 43 Results of Moderated Me diation (PREDICTOR= EEODIV, Med = Race-based Discrimination, DV= Intent to Turnover, Mod = Race) 87 Table 44 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = EEODIV, Med = Race-based Discrimination, DV = Intent to Turnover) 88 Table 45 Results of Moderated Me diation (PREDICTOR= EEODIV, Med = Race-based Discrimination, DV= Physical Health, Mod = Race) 89 Table 46 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = EEODIV, Med = Race-based Discrimination, DV = Physical Health) 90 Table 47 Results of Moderated Me diation (PREDICTOR= EEODIV, Med = Race-based Discrimination, DV= Psychological Health, Mod = Race) 91

PAGE 9

vii Table 48 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = EEODIV, Med = Race-based Discrimination, DV = Psychological Health) 92 Table 49 Results of Moderated Mediation (PREDICTOR= Minority Segmentation, Med = Sex-based Discrimination, DV= Job Satisfaction, Mod = Gender) 93 Table 50 Results of Simple Mediat ion for Men and Women (PREDICTOR = Minority Segmentation, Med = Sex-based Discrimination, DV = Job Satisfaction) 94 Table 51 Results of Moderated Mediation (PREDICTOR= Minority Segmentation, Med = Sex-based Discrimination, DV= Organizational Commitment, Mod = Gender) 95 Table 52 Results of Simple Mediat ion for Men and Women (PREDICTOR = Minority Segmentation, Med = Sex-based Discrimination, DV = Organizational Commitment) 96 Table 53 Results of Moderated Mediation (PREDICTOR= Minority Segmentation, Med = Sex-based Di scrimination, DV= Intent to Turnover, Mod = Gender) 97 Table 54 Results of Simple Mediat ion for Men and Women (PREDICTOR = Minority Segmentation, Med = Sex-based Discrimination, DV = Intent to Turnover) 98 Table 55 Results of Moderated Mediation (PREDICTOR= Minority Segmentation, Med = Sex-based Discrimination, DV= Physical Health, Mod = Gender) 99 Table 56 Results of Simple Mediat ion for Men and Women (PREDICTOR = Minority Segmentation, Med = Sex-based Discrimination, DV = Physical Health) 100 Table 57 Results of Moderated Mediation (PREDICTOR= Minority Segmentation, Med = Sex-based Discrimination, DV= Psychological Health, Mod = Gender) 101 Table 58 Results of Simple Mediat ion for Men and Women (PREDICTOR = Minority Segmentation, Med = Sex-based Discrimination, DV = Psychological Health) 102

PAGE 10

viii Table 59 Results of Moderated Mediation (PREDICTOR= Minority Segmentation, Med = Race-based Discrimination, DV= Job Satisfaction, Mod = Race) 103 Table 60 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = Minority Segm entation, Med = Race-based Discrimination, DV = Job Satisfaction) 104 Table 61 Results of Moderated Mediation (PREDICTOR= Minority Segmentation, Med = Race-based Discrimination, DV= Organizational Commitment, Mod = Race) 105 Table 62 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = Minority Segm entation, Med = Race-based Discrimination, DV = Orga nizational Commitment) 106 Table 63 Results of Moderated Mediation (PREDICTOR= Minority Segmentation, Med = Race-based Di scrimination, DV= Intent to Turnover, Mod = Race) 107 Table 64 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = Minority Segm entation, Med = Race-based Discrimination, DV = Intent to Turnover) 108 Table 65 Results of Moderated Mediation (PREDICTOR= Minority Segmentation, Med = Race-based Discrimination, DV= Physical Health, Mod = Race) 109 Table 66 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = Minority Segm entation, Med = Race-based Discrimination, DV = Physical Health) 110 Table 67 Results of Moderated Mediation (PREDICTOR= Minority Segmentation, Med = Race-based Discrimination, DV= Psychological Health, Mod = Race) 111 Table 68 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = Minority Segm entation, Med = Race-based Discrimination, DV = Psychological Health) 112 Table 69 Results of Moderated Medi ation (PREDICTOR= Social Support, Med = Sex-based Discrimination, DV= Job Satisfaction, Mod = Gender) 113

PAGE 11

ix Table 70 Results of Simple Mediat ion for Men and Women (PREDICTOR = Social Support, Med = Sexbased Discrimination, DV = Job Satisfaction) 114 Table 71 Results of Moderated Medi ation (PREDICTOR= Social Support, Med = Sex-based Discrimination, DV= Organizational Commitment, Mod = Gender) 115 Table 72 Results of Simple Mediat ion for Men and Women (PREDICTOR = Social Support, Med = Se x-based Discrimination, DV = Organizational Commitment) 116 Table 73 Results of Moderated Medi ation (PREDICTOR= Social Support, Med = Sex-based Discrimination, DV= Intent to Turnover, Mod = Gender) 117 Table 74 Results of Simple Mediat ion for Men and Women (PREDICTOR = Social Support, Med = Sex-ba sed Discrimination, DV = Intent to Turnover) 118 Table 75 Results of Moderated Medi ation (PREDICTOR= Social Support, Med = Sex-based Discrimination, DV= Physical Health, Mod = Gender) 119 Table 76 Results of Simple Mediat ion for Men and Women (PREDICTOR = Social Support, Med = Se x-based Discrimination, DV = Physical Health) 120 Table 77 Results of Moderated Medi ation (PREDICTOR= Social Support, Med = Sex-based Discriminati on, DV= Psychological Health, Mod = Gender) 121 Table 78 Results of Simple Mediat ion for Men and Women (PREDICTOR = Social Support, Med = Se x-based Discrimination, DV = Psychological Health) 121 Table 79 Results of Moderated Medi ation (PREDICTOR= Social Support, Med = Race-based Discriminati on, DV= Job Satisfaction, Mod = Race) 123 Table 80 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = Social Support, Med = Race-based Discrimination, DV = Job Satisfaction) 124

PAGE 12

x Table 81 Results of Moderated Medi ation (PREDICTOR= Social Support, Med = Race-based Discrimination, DV= Organizational Commitment, Mod = Race) 125 Table 82 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = Social Support, Med = Race-based Discrimination, DV = Orga nizational Commitment) 126 Table 83 Results of Moderated Medi ation (PREDICTOR= Social Support, Med = Race-based Discrimination, DV= Intent to Turnover, Mod = Race) 127 Table 84 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = Social Support, Med = Race-based Discrimination, DV = Intent to Turnover) 128 Table 85 Results of Moderated Medi ation (PREDICTOR= Social Support, Med = Race-based Discriminati on, DV= Physical Health, Mod = Race) 129 Table 86 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = Social Support, Med = Race-based Discrimination, DV = Physical Health) 129 Table 87 Results of Moderated Medi ation (PREDICTOR= Social Support, Med = Race-based Discrimination, DV= Psychological Health, Mod = Race) 130 Table 88 Results of Simple Me diation for Whites and Non-Whites (PREDICTOR = Social Support, Med = Race-based Discrimination, DV = Psychological Health) 131 Table 89 Summary of Significant Results for Equal Employment Opportunity/Diversity Climate 132 Table 90 Summary of Significant Results for Minority Segmentation 132 Table 91 Summary of Significan t Results for Social Support 133 Table 92 Descriptive Statistics for Organizational Antecedents Comparing Study Participants to Coworkers 134 Table 93 Results of Moderated Media tion (Predictor = EEO, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Gender) 191

PAGE 13

xi Table 94 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EEO, Med = Subtle Discrimination, DV = Job Satisfaction) 192 Table 95 Results of Moderated Media tion (Predictor = EEO, Med = Subtle Discrimination, DV = Organiza tional Commitment, Mod = Gender) 193 Table 96 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EEO, Med = Subtle Discrimination, DV = Organizational Commitment) 194 Table 97 Results of Moderated Media tion (Predictor = EEO, Med = Subtle Discrimination, DV = Intent to Tu rnover, Mod = Gender) 195 Table 98 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EEO, Med = Subtle Discrimination, DV = Intent to Turnover) 196 Table 99 Results of Moderated Media tion (Predictor = EEO, Med = Subtle Discrimination, DV = Physical Health, Mod = Gender) 197 Table 100 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EEO, Med = Subtle Discrimination, DV = Physical Health) 198 Table 101 Results of Moderated Media tion (Predictor = EEO, Med = Subtle Discrimination, DV = Psychologi cal Health, Mod = Gender) 199 Table 102 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EEO, Med = Subtle Discrimination, DV = Psychological Health) 200 Table 103 Results of Moderated Medi ation (Predictor = EEO, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Gender) 201 Table 104 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EEO, Med = Overt Discrimination, DV = Job Satisfaction) 202 Table 105 Results of Moderated Medi ation (Predictor = EEO, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Gender) 203

PAGE 14

xii Table 106 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EEO, Med = Overt Discrimination, DV = Organizational Commitment) 204 Table 107 Results of Moderated Medi ation (Predictor = EEO, Med = Overt Discrimination, DV = Intent to Turnover, Mod = Gender) 205 Table 108 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EEO, Me d = Overt Discrimination, DV = Intent to Turnover) 206 Table 109 Results of Moderated Medi ation (Predictor = EEO, Med = Overt Discrimination, DV = Physical H ealth, Mod = Gender) 207 Table 110 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EEO, Me d = Overt Discrimination, DV = Physical Health) 208 Table 111 Results of Moderated Medi ation (Predictor = EEO, Med = Overt Discrimination, DV = Psychological H ealth, Mod = Gender) 209 Table 112 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EEO, Me d = Overt Discrimination, DV = Psychological Health) 210 Table 113 Results of Moderated Media tion (Predictor = EEO, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) 211 Table 114 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EEO, Me d = Subtle Discrimination, DV = Job Satisfaction) 212 Table 115 Results of Moderated Media tion (Predictor = EEO, Med = Subtle Discrimination, DV = Organizati onal Commitment, Mod = Race) 213 Table 116 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EEO, Me d = Subtle Discrimination, DV = Organizational Commitment) 214 Table 117 Results of Moderated Media tion (Predictor = EEO, Med = Subtle Discrimination, DV = Intent to Turnover, Mod = Race) 215 Table 118 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EEO, Med = Subtle Discrimination, DV = Intent to Turnover) 216

PAGE 15

xiii Table 119 Results of Moderated Media tion (Predictor = EEO, Med = Subtle Discrimination, DV = Physical H ealth, Mod = Race) 217 Table 120 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EEO, Med = Subtle Discrimination, DV = Physical Health) 218 Table 121 Results of Moderated Media tion (Predictor = EEO, Med = Subtle Discrimination, DV = Psychological H ealth, Mod = Race) 219 Table 122 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EEO, Me d = Subtle Discrimination, DV = Psychological Health) 220 Table 123 Results of Moderated Medi ation (Predictor = EEO, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) 221 Table 124 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EEO, Med = Overt Discrimination, DV = Job Satisfaction) 222 Table 125 Results of Moderated Medi ation (Predictor = EEO, Med = Overt Discrimination, DV = Organizati onal Commitment, Mod = Race) 223 Table 126 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EEO, Med = Overt Discrimination, DV = Organizational Commitment) 224 Table 127 Results of Moderated Medi ation (Predictor = EEO, Med = Overt Discrimination, DV = Intent to Turnove r, Mod = Race) 225 Table 128 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EEO, Med = Overt Discrimination, DV = Intent to Turnover) 226 Table 129 Results of Model rated Medi ation (Predictor = EEO, Med = Overt Discrimination, DV = Physical Health, Mod = Race) 227 Table 130 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EEO, Med = Overt Discrimination, DV = Physical Health) 228 Table 131 Results of Moderated Medi ation (Predictor = EEO, Med = Overt Discrimination, DV = Psychological H ealth, Mod = Race) 229

PAGE 16

xiv Table 132 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EEO, Med = Overt Discrimination, DV = Psychological Health) 230 Table 133 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Gender) 231 Table 134 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Job Satisfaction) 232 Table 135 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Gender) 233 Table 136 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Organizational Comm itment) 234 Table 137 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Intent to Turnover, Mod = Gender) 235 Table 138 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Intent to Turnover) 236 Table 139 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Physical Health, Mod = Gender) 237 Table 140 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Physical Health) 238 Table 141 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Psychological Health, Mod = Gender) 239 Table 142 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Psychological Health) 240

PAGE 17

xv Table 143 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Gender) 241 Table 144 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Job Satisfaction) 242 Table 145 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Gender) 243 Table 146 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Organiza tional Commitment) 244 Table 147 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Intent to Turnover, Mod = Gender) 245 Table 148 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Intent to Turnover) 246 Table 149 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Physical Health, Mod = Gender) 247 Table 150 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Physical Health) 248 Table 151 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Psychological Health, Mod = Gender) 249 Table 152 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Psychological Health) 250 Table 153 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) 251

PAGE 18

xvi Table 154 Results of Simple Mediat ion Follow-up Tests for Whites and Nonwhites (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Job Satisfaction) 252 Table 155 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Race) 253 Table 156 Results of Simple Mediat ion Follow-up Tests for Whites and Nonwhites (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Organizational Commit ment) 254 Table 157 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Intent to Turnover, Mod = Race) 255 Table 158 Results of Simple Mediat ion Follow-up Tests for Whites and Nonwhites (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Intent to Turnover) 256 Table 159 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Physical Health, Mod = Race) 257 Table 160 Results of Simple Mediat ion Follow-up Tests for Whites and Nonwhites (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Physical Health) 258 Table 161 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Subtle Discrimination, DV = Psychological Health, Mod = Race) 259 Table 162 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = MI NORITY SEG, Med = Subtle Discrimination, DV = Psychological Health) 260 Table 163 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) 261 Table 164 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = MI NORITY SEG, Med = Overt Discrimination, DV = Job Satisfaction) 262 Table 165 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Race) 263

PAGE 19

xvii Table 166 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = MI NORITY SEG, Med = Overt Discrimination, DV = Organiza tional Commitment) 264 Table 167 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Intent to Turnover, Mod = Race) 265 Table 168 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Intent to Turnover) 266 Table 169 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Physical Health, Mod = Race) 267 Table 170 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = MI NORITY SEG, Med = Overt Discrimination, DV = Physical Health) 268 Table 171 Results of Moderated Medi ation (Predictor = MINORITY SEG, Med = Overt Discrimination, DV = Psychological Health, Mod = Race) 269 Table 172 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = MI NORITY SEG, Med = Overt Discrimination, DV = Psychological Health) 270 Table 173 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Subtle Discri mination, DV = Job Satisfaction, Mod = Gender) 271 Table 174 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = DIVERS ITY CLIMATE, Med = Subtle Discrimination, DV = Job Satisfaction) 272 Table 175 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Subtle Disc rimination, DV = Organizational Commitment, Mod = Gender) 273 Table 176 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = DIVERS ITY CLIMATE, Med = Subtle Discrimination, DV = Organi zational Commitment) 274

PAGE 20

xviii Table 177 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Subtle Discrimination, DV = Intent to Turnover, Mod = Gender) 275 Table 178 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = DIVERSIT Y CLIMATE, Med = Subtle Discrimination, DV = Intent to Turnover) 276 Table 179 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Subtle Discri mination, DV = Physical Health, Mod = Gender) 277 Table 180 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = DIVERS ITY CLIMATE, Med = Subtle Discrimination, DV = Physical Health) 278 Table 181 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Subtle Disc rimination, DV = Psychological Health, Mod = Gender) 279 Table 182 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = DIVERS ITY CLIMATE, Med = Subtle Discrimination, DV = Psychological Health) 280 Table 183 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Gender) 281 Table 184 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = DIVERSITY CLIMATE, Med = Overt Discrimination, DV = Job Satisfaction) 282 Table 185 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Overt Discri mination, DV = Organizational Commitment, Mod = Gender) 283 Table 186 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = DIVERSITY CLIMATE, Med = Overt Discrimination, DV = Organizational Co mmitment) 284 Table 187 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Overt Discrimination, DV = Intent to Turnover, Mod = Gender) 285

PAGE 21

xix Table 188 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = DIVERSITY CLIMATE, Med = Overt Discrimination, DV = Intent to Turnover) 286 Table 189 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Overt Discrimi nation, DV = Physical Health, Mod = Gender) 287 Table 190 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = DIVERSITY CLIMATE, Med = Overt Discrimination, DV = Physical Health) 288 Table 191 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Overt Discrimination, DV = Psychological Health, Mod = Gender) 289 Table 192 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = DIVERSITY CLIMATE, Med = Overt Discrimination, DV = Psychological Health) 290 Table 193 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) 291 Table 194 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = DIVERS ITY CLIMATE, Med = Subtle Discrimination, DV = Job Satisfaction) 292 Table 195 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Subtle Disc rimination, DV = Organizational Commitment, Mod = Race) 293 Table 196 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = DIVERS ITY CLIMATE, Med = Subtle Discrimination, DV = Organizationa l Commitment) 294 Table 197 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Subtle Discrimination, DV = Intent to Turnover, Mod = Race) 295 Table 198 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = DIVERS ITY CLIMATE, Med = Subtle Discrimination, DV = Intent to Turnover) 296

PAGE 22

xx Table 199 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Subtle Discri mination, DV = Physical Health, Mod = Race) 297 Table 200 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = DIVERS ITY CLIMATE, Med = Subtle Discrimination, DV = Physical Health) 298 Table 201 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Subtle Disc rimination, DV = Psychological Health, Mod = Race) 299 Table 202 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = DIVERS ITY CLIMATE, Med = Subtle Discrimination, DV = Psychological Health) 300 Table 203 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) 301 Table 204 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = DIVE RSITY CLIMATE, Med = Overt Discrimination, DV = Job Satisfaction) 302 Table 205 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Overt Discri mination, DV = Organizational Commitment, Mod = Race) 303 Table 206 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = DIVE RSITY CLIMATE, Med = Overt Discrimination, DV = Organizational Co mmitment) 304 Table 207 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Overt Discrimination, DV = Intent to Turnover, Mod = Race) 305 Table 208 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = DIVE RSITY CLIMATE, Med = Overt Discrimination, DV = Intent to Turnover) 306 Table 209 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Overt Discrimi nation, DV = Physical Health, Mod = Race) 307

PAGE 23

xxi Table 210 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = DIVE RSITY CLIMATE, Med = Overt Discrimination, DV = Physical Health) 308 Table 211 Results of Moderated Me diation (Predict or = DIVERSITY CLIMATE, Med = Overt Discrimination, DV = Psychological Health, Mod = Race) 309 Table 212 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = DIVE RSITY CLIMATE, Med = Overt Discrimination, DV = Psychological Health) 310 Table 213 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Gender) 311 Table 214 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Job Satisfaction) 312 Table 215 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Gender) 313 Table 216 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Organizationa l Commitment) 313 Table 217 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Intent to Turnover, Mod = Gender) 314 Table 218 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Intent to Turnover) 315 Table 219 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Physical Health, Mod = Gender) 316 Table 220 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Physical Health) 317

PAGE 24

xxii Table 221 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Psychological Health, Mod = Gender) 318 Table 222 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Psychologi cal Health) 319 Table 223 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Overt Discriminati on, DV = Job Satisfaction, Mod = Gender) 320 Table 224 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Job Satisfaction) 321 Table 225 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Gender) 322 Table 226 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Organizational Commitment) 322 Table 227 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Intent to Turnover, Mod = Gender) 323 Table 228 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Intent to Turnover) 324 Table 229 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Physical Health, Mod = Gender) 325 Table 230 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Physical Health) 326 Table 231 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Psychological Health, Mod = Gender) 327

PAGE 25

xxiii Table 232 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Psychological Health) 327 Table 233 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) 329 Table 234 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = INST RUMENTAL SS, Med = Subtle Discrimination, DV = Job Satisfaction) 330 Table 235 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Race) 331 Table 236 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = INST RUMENTAL SS, Med = Subtle Discrimination, DV = Organizational Commitme nt) 332 Table 237 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Intent to Turnover, Mod = Race) 333 Table 238 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = INST RUMENTAL SS, Med = Subtle Discrimination, DV = Intent to Turnover) 334 Table 239 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Subtle Discriminati on, DV = Physical Health, Mod = Race) 335 Table 240 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = INST RUMENTAL SS, Med = Subtle Discrimination, DV = Physical Health) 335 Table 241 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Subtle Discrimination, DV = Psychological Health, Mod = Race) 336 Table 242 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = INST RUMENTAL SS, Med = Subtle Discrimination, DV = Psychological Health) 337

PAGE 26

xxiv Table 243 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) 338 Table 244 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = INST RUMENTAL SS, Med = Overt Discrimination, DV = Job Satisfaction) 339 Table 245 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Race) 340 Table 246 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = INST RUMENTAL SS, Med = Overt Discrimination, DV = Organizationa l Commitment) 341 Table 247 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Intent to Turnover, Mod = Race) 342 Table 248 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = INST RUMENTAL SS, Med = Overt Discrimination, DV = Intent to Turnover) 343 Table 249 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Overt Discrimination, DV = Physical Health, Mod = Race) 344 Table 250 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = INST RUMENTAL SS, Med = Overt Discrimination, DV = Physical Health) 345 Table 251 Results of Moderated Medi ation (Predictor = INSTRUMENTAL SS, Med = Overt Discriminati on, DV = Psychological Health, Mod = Race) 346 Table 252 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = INST RUMENTAL SS, Med = Overt Discrimination, DV = Psychological Health) 347 Table 253 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Gender) 348

PAGE 27

xxv Table 254 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EMOT IONAL SS, Med = Subtle Discrimination, DV = Job Satisfaction) 349 Table 255 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Gender) 350 Table 256 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EMOT IONAL SS, Med = Subtle Discrimination, DV = Organizati onal Commitment) 351 Table 257 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Intent to Turnover, Mod = Gender) 352 Table 258 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Intent to Turnover) 353 Table 259 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Physical Health, Mod = Gender) 354 Table 260 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EMOT IONAL SS, Med = Subtle Discrimination, DV = Physical Health) 355 Table 261 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Psychological Health, Mod = Gender) 356 Table 262 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EMOT IONAL SS, Med = Subtle Discrimination, DV = Psychological Health) 356 Table 263 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Gender) 358 Table 264 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Job Satisfaction) 359

PAGE 28

xxvi Table 265 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Gender) 360 Table 266 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Organizational Comm itment) 361 Table 267 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Intent to Turnover, Mod = Gender) 362 Table 268 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EMOT IONAL SS, Med = Overt Discrimination, DV = Intent to Turnover) 363 Table 269 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Physical Health, Mod = Gender) 364 Table 270 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EMOT IONAL SS, Med = Overt Discrimination, DV = Physical Health) 365 Table 271 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Psychological Health, Mod = Gender) 366 Table 272 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = EMOT IONAL SS, Med = Overt Discrimination, DV = Psychological Health) 366 Table 273 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) 368 Table 274 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Job Satisfaction) 369 Table 275 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Race) 370

PAGE 29

xxvii Table 276 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EM OTIONAL SS, Med = Subtle Discrimination, DV = Organi zational Commitment) 371 Table 277 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Intent to Turnover, Mod = Race) 372 Table 278 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EM OTIONAL SS, Med = Subtle Discrimination, DV = Intent to Turnover) 373 Table 279 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Physical Health, Mod = Race) 374 Table 280 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EM OTIONAL SS, Med = Subtle Discrimination, DV = Physical Health) 375 Table 281 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Psychological Health, Mod = Race) 376 Table 282 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EMOTIONAL SS, Med = Subtle Discrimination, DV = Psychological Health) 377 Table 283 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) 378 Table 284 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EM OTIONAL SS, Med = Overt Discrimination, DV = Job Satisfaction) 379 Table 285 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Race) 380 Table 286 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EM OTIONAL SS, Med = Overt Discrimination, DV = Organizational Commit ment) 381

PAGE 30

xxviii Table 287 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Intent to Turnover, Mod = Race) 382 Table 288 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EM OTIONAL SS, Med = Overt Discrimination, DV = Intent to Turnover) 383 Table 289 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Physical Health, Mod = Race) 384 Table 290 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Physical Health) 385 Table 291 Results of Moderated Medi ation (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Psychological Health, Mod = Race) 386 Table 292 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = EMOTIONAL SS, Med = Overt Discrimination, DV = Psychological Health) 387 Table 293 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Gender) 388 Table 294 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = TOKE N STATUS, Med = Subtle Discrimination, DV = Job Satisfaction) 389 Table 295 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Gender) 390 Table 296 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = TOKE N STATUS, Med = Subtle Discrimination, DV = Organizational Commitment ) 391 Table 297 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Subtle Discrimination, DV = Intent to Turnover, Mod = Gender) 392

PAGE 31

xxix Table 298 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = TOKE N STATUS, Med = Subtle Discrimination, DV = Intent to Turnover) 393 Table 299 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Subtle Discrimination, DV = Physical Health, Mod = Gender) 394 Table 300 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = TOKE N STATUS, Med = Subtle Discrimination, DV = Physical Health) 395 Table 301 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Subtle Discrimination, DV = Psychological Health, Mod = Gender) 396 Table 302 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = TOKE N STATUS, Med = Subtle Discrimination, DV = Psychological Health) 397 Table 303 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Gender) 398 Table 304 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Job Satisfaction) 399 Table 305 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Gender) 400 Table 306 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Organizational Comm itment) 401 Table 307 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Intent to Turnover, Mod = Gender) 402 Table 308 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Intent to Turnover) 403

PAGE 32

xxx Table 309 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Physical Health, Mod = Gender) 404 Table 310 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Physical Health) 405 Table 311 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Psychological Health, Mod = Gender) 406 Table 312 Results of Simple Medi ation Follow-up Tests for Men and Women (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Psychological Health) 407 Table 313 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) 408 Table 314 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = T OKEN STATUS, Med = Subtle Discrimination, DV = Job Satisfaction) 409 Table 315 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Race) 410 Table 316 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = T OKEN STATUS, Med = Subtle Discrimination, DV = Organizational Co mmitment) 411 Table 317 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Subtle Discrimination, DV = Intent to Turnover, Mod = Race) 412 Table 318 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = T OKEN STATUS, Med = Subtle Discrimination, DV = Intent to Turnover) 413 Table 319 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Subtle Discrimination, DV = Physical Health, Mod = Race) 414

PAGE 33

xxxi Table 320 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = T OKEN STATUS, Med = Subtle Discrimination, DV = Physical Health) 415 Table 321 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Subtle Discrimination, DV = Psychological Health, Mod = Race) 416 Table 322 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = T OKEN STATUS, Med = Subtle Discrimination, DV = Psychological Health) 417 Table 323 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) 418 Table 324 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = T OKEN STATUS, Med = Overt Discrimination, DV = Job Satisfaction) 419 Table 325 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Race) 420 Table 326 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = T OKEN STATUS, Med = Overt Discrimination, DV = Organizational Commitment ) 421 Table 327 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Intent to Turnover, Mod = Race) 422 Table 328 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = T OKEN STATUS, Med = Overt Discrimination, DV = Intent to Turnover) 423 Table 329 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Physical Health, Mod = Race) 424 Table 330 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Physical Health) 425

PAGE 34

xxxii Table 331 Results of Moderated Medi ation (Predictor = TOKEN STATUS, Med = Overt Discrimination, DV = Psychological Health, Mod = Race) 426 Table 332 Results of Simple Medi ation Follow-up Tests for Whites and Non-whites (Predictor = T OKEN STATUS, Med = Overt Discrimination, DV = Psychological Health ) 427 Table 333 Summary of Si gnificant Results for Perceived Equal Employment Opportunity 428 Table 334 Summary of Significant Results for Minority Segmentation 429 Table 335 Summary of Significant Results for Diversity Climate 430 Table 336 Summary of Significant Results for Instrumental Social Support 431 Table 337 Summary of Si gnificant Results for Emotional Social Support 432 Table 338 Summary of Significan t Results for Token Status 433

PAGE 35

xxxiii Organizational Contributions to Perceived Workplace Discrimination Nicole Ellis Jagusztyn Abstract The purpose of the current study was to explor e the role of six organizational factors (Equal Employment Opportunity, minority segm entation, diversity climate, instrumental social support, emotional social support, and token status) in the perception of discrimination in the workplace by minor ities and majority-group members. Five outcomes (job satisfaction, organizational comm itment, intention to turnover, physical health, and psychological health) were in vestigated in response to perceived discrimination. Moderated mediation was us ed to test hypothesis where perceived discrimination mediated the relationship between organizational antecedents and outcomes; minority status served as the m oderators. Support for the mediating role of perceived discrimination was found in the relationship between each organizational antecedent and outcome. In each case, poorer environmental conditions related to increased perceived discrimination which in turn related to more negative workplace attitudes and health outcomes. Implica tions for workplace design are discussed.

PAGE 36

1 Chapter One Introduction Overview of the paper There is a general paucity of research on perceived discrimination at work, and the studies that do exist in this area have focused on outcomes or individual factors contributing to the perception of differe ntial treatment. However, perceived discrimination does not originate in the mind of its victim. There are likely a host of environmental and organizational antecedents to these beliefs. Understanding these antecedents is an important first step towa rds alleviating actual discrimination in the workplace. The study to follow will investigate how or ganizational factors impact perceived discrimination from the perspectives of bot h minority and majority group members. Two types of perceived discrimination will be considered: subtle and overt. Perceived discrimination will be viewed as a stressor impacting five outcomes: job satisfaction, organizational commitment, intention to tu rnover, psychological health and physical health. Minority group membership Although “minority group” te nds to conjure images of a group composed of fewer members compared to a majority group, this is not necessarily the case. For the purposes of this paper, a minority group is any group of individuals that hold a disparate amount of

PAGE 37

2 power compared to a majority group, rega rdless of size. In the United States, the predominant majority group in most work c ontexts is white indi viduals, specifically, white American men. The largest minority groups are women and non-white individuals (both men and women). The literature has shown the members of these groups to be disadvantaged both economically and socially. For example, a 2007 Census survey revealed that the female-to-male earnings ratio was 0.78. In other words, women have approximately 78% the income compared to men (DeNavas-Walt, Proctor & Smith, 2008). Further, the characteristics of a successf ul manager tend be more masculine traits compared to feminine traits. A study co mpared data collected during 1976-1977, 19841985, and 1999, and found that although the trend is decreasing over time, a good manager is still seen as being more mascul ine (Powell, Butterfiel d, & Parent, 2002). This predicament is not limited to women. For ex ample, blacks earn less than comparable whites (Beggs, 1995; Bridges & Villemez, 1994) in addition to having more negative interpersonal outcomes when interacting wi th white majority group members (Crosby, Bromly & Saxe, 1980). Given that women and nonwhite individuals compri se the largest minority groups in the United States, it is no surp rise that the bulk of research on the discrimination of “minorities” includes fema le and non-white samples, as discrimination likely has the largest impact for thes e groups in employment. Moreover, the overwhelming majority of studies involving non-whites focus on black individuals. This might be because blacks have traditionally be en the largest non-white minority group in the U.S., and because of the lengthy history of overt institutional discrimination against this group. Thus, it is important to note that while the bulk of supporting research for

PAGE 38

3 non-whites in this paper will center on black individuals, other non-white minorities will be included in this study. For the purposes of this paper, a minority is any individual who self-identifies as female (white or non-white) or non-white (mal e or female). Because two separate sets of comparative analyses will be done, a majority group member is defined as a male (white or non-white) when being compared to a female (white or non-white) or as a white individual (male or female) when being co mpared to a non-white individual (male or female). An Aside: Perceived Sexism/Racism and Perceived Discrimination Defined Throughout this paper, sexism/raci sm and discrimination are used interchangeably. There is a difference between sexism/racism and discrimination, however. Sexism or racism includes feelings, opinions, and ideas that a person’s worth is based on gender or race. Discrimination in cludes differential actions towards people based on their gender or race as a result of th e attitudes towards people belonging to that gender/race. Throughout this paper, the terms may be used interchangeably, because the perception and outcomes of these cons tructs are closely related. The existence of discrimination Before one can discuss perceived discri mination, it is important to demonstrate the existence of actual discrimination in th e workplace. A multitude of studies have demonstrated empirical evidence of discrimina tion against minorities in every aspect of employment. First, recruitment efforts may be influenced by prejudice. One study found that companies tended to focus their recrui tment efforts towards white neighborhoods and prestigious schools while avoidi ng job advertisements in news papers and public agencies

PAGE 39

4 (Neckerman & Kirschenman, 1991) where non-white s are likely to look. These strategies systematically excluded inner-city minori ties from entering consideration; employers believed their methods attracted higher-quali ty workers. Second, s ubjectivity during the hiring process may lead to discriminati on against certain gr oups. Fidell (1970) demonstrated that women are offered lower-lev el jobs and are rated as less desirable job candidates compared to men. King, Madera, Hebl, Knight & Mendoza (2006) found that resumes with stereotypically “black” names were rated lower than those with neutral names, regardless of the qualifications of the individual. Additionally, Stewart and Perlow (2001) found that eval uators have more confidence when assigning whites to higher status jobs and convers ely assigning blacks to lower status jobs, demonstrating possible discrimination in job assignments. Next, performance appraisals may be subj ect to bias. One study found that black employees received lower supervisory ratings when rated by white managers compared to black mangers (Stauffer & Buckley, 2005). Ho wever, white ratees received equivalent ratings regardless of whether they were rate d by white or black s upervisors. Greenhaus and Parasuraman (1993) investigated the attr ibutions of performance ratings and found that the supervisors associated the success of black managers to help from others more often than either ability or effort, compared to white managers. It has also been shown that black managers are rated lower on both task and contextu al job performance compared to white managers (Greenhaus Parasuraman, & Wormley 1990). Advancement is also subject to discrimination. Landau ( 1995) demonstrated that race and gender were significantly related to ratings of promotion potential, even after controlling for age, education, tenure, salary grade, and functiona l area. Here, women were rated lower than

PAGE 40

5 men, and both Asians and blacks were rated lo wer than whites. More over, several studies have demonstrated wage disparities for minorities. Women have lower wage and salary income than men (DeNavas-Walt, Proctor, & Smith, 2008). Also, blacks earn less than comparable whites (i.e. in education, e xperience, hours worked, etc.; Beggs, 1995; Bridges & Villemez, 1994). Promotions ar e also impacted by discrimination. Maume (1999) found that blacks were less likely th an whites to be promoted, and men were highly likely to be promoted to a supervis ory position if coming from an occupation dominated by women. Similarly, Baldi and McBrier (1997) found that white employees were highly likely to be promoted when in “scarce supply” among minorities. Another study found that the determinants of promoti on differ for minorities (Baldi & McBrier, 1997). Here, education was seen as a screeni ng device for the hire of white employees but was not related to advancement. In contra st, education was considered a prerequisite for the advancement of black employees. Si milarly, Mueller, Par cel & Tanaka (1989) demonstrated that black supervisors have to exhibit positive managerial qualities to a greater extent than white supervisors in order to be considered for promotion. Additionally, black men and white women waite d longer than white men for promotions. Further, women are often on career tracks that do not allow for very high advancement within an organization (Gutek, Larwood & Stromberg, 1986). The disparities in promotion for minorities are likely linked to prejudice from majority-group decision makers. For example, white managers are more likely to associate the characteristics of successful middle managers with stereotypica lly “white” qualities rather than qualities stereotypically assigned to black individua ls (Tomkiewicz, Brenner & Adeyemi-Bello, 1998). Also, Foschi, Lai and Si gerson (1994) found that gend er is a cue to competence

PAGE 41

6 for men, but not for women. When there is a lack of evidence for competence, men will use gender in making job-related decisions. A variety of studies seem to demonstrat e the existence of discrimination in all aspects of employment. It seems plausible th at there would be an equally widespread amount of perceived discri mination among minorities. The prevalence of perceived discrimination Whether or not discriminatory practices are actually in place, the notion of their presence may impact minorities. A nati onal study on socially disadvantaged people reported on the prevalence of perceived di scrimination (Kessler, Mickelson & Williams, 1999). Results indicated that 33.5% of the re spondents experienced exposure to major lifetime discrimination and 60.9% experien ced day-to-day discrimination. Black Americans and Latina/os are more likely than white Americans to say that they, and other members of their group, have personally experienced discrimination based on their ethnicity (Operario & Fiske, 2001). Women are also more likely than men to label a negative act as discrimination if it is perpetrated by a majo rity group member against a minority group member (Rodin, Pr ice, Bryson & Sanchez, 1990). Perceived discrimination is also common in the work environment. Results from several studies indicate that non-whites do pe rceive their work environments to be somehow discriminatory, while white indi viduals tend to believe there are equal opportunities for all people (Hite, 2004; J eanquart-Barone & Sekara n, 1996). Similarly, women tend to view other women as target s of discrimination more often than men (Gutek, Cohen, & Tsui, 1996) and tend to repor t experiencing discrimination personally (Frienze, Olson, & Good, 1990). While currently there is a dearth of research on

PAGE 42

7 perceived discrimination specific to stages during the job lifecycle (e.g. during selection, job assignments, layoffs) there is some evidence of minorities’ beliefs about differential treatment during career development and advancement (Burlew & Johnson, 1992; Jackson, 1994). Thus, in keeping with the evidence of the existence of actual discrimination, there is evidence of perceived discrimination on the part of minorities. These perceptions of differential treatment are likely to impact a ttitudes and behaviors on the part of those affected. An Aside: Overt vs. Subtle Perceived Discrimination Actual discrimination may take a subtle or overt form. Overt discrimination tends to be more obvious. Individuals may be deni ed resources, experience exclusion, or be publically mistreated as a resu lt of their group status. Subt le discrimination, on the other hand, tends to be more covert. Individuals re ceive interpersonal mistreatment as a result of their group status. For example, an em ployee may insult a minority coworker based on their ethnicity. Thus, perceived discriminati on can take different forms as well. When individuals perceive overt di scrimination, they are experiencing a more blatant form of discrimination whereas subtle perceived discrimination results from actions which can be more ambiguous but are still attributed to group status. There is evidence that forms of overt discrimination are have become less common in the past few decades (see Dovidi o & Gaertner, 1991, for a review) as people of all races have tended to adopt more egal itarian views. However, as aforementioned studies have highlighted, all forms of disc rimination are still in existence today. The

PAGE 43

8 frequency, nature and severity of the perceived discrimination will differentially impact individual targets. Perceived discrimination as a stressor Perceived discrimination is a stressor fo r minority employees. Several studies have explicitly named race-based discriminati on as a general stressor (Contrada et al., 2001; Clark, Anderson, Clark & Williams, 1999) and an occupational stressor for black individuals (Hughes & Dodge, 1997). Landrin e and Klonoff (1996) found that 99.4% of the subjects in their study found racial discri mination to be stressful Discrimination has been well-documented as a stressor for black Americans, and the stress associated with it has been linked to negative physical and ps ychological outcomes (Hunter & Lewis-Coles, 2004). One definition of “stress” is “a part icular relationship between the person and the environment that is appraised by the person as taxing or exceeding his or her resources and endangering his or her well-being” (Lazarus & Folkman, 1984, p.19). The abundance of negative outcomes associated with perceived discrimination alone demonstrates how it can exceed an individual’ s resources or endanger their well-being, and thus implies it as a stressor. The costs of perceived discrimination From an organizational perspective, th e financial consequences of perceived discrimination can be devastating, as perceive d discrimination is the logical precursor to discrimination lawsuits. There have been several large class-ac tion lawsuits brought against major employers in the last two decades. In 2001, Coca-Cola paid $192.5 million to plaintiffs who accused the company of race-biased promotion practices (King & Spruell, 2001). Similar cases have been br ought against Texaco, Inc., Shoney’s, Inc.,

PAGE 44

9 Winn Dixie Stores, Inc., and CSX transporta tion. The millions spent in settlements are only part of the equation; discrimination suit s also cost companies in legal fees, lost productivity, and damage to the company’s im age. In other words, a company’s bottom line can be seriously impacted by em ployee perceptions of fairness. However, the “costs” of perceived di scrimination extend beyond the financial burden of lawsuits. Victims may experien ce chronic health outcomes, short-term physiological symptoms, job stress, psychological distress, changes in job attitudes or other negative consequences that may indi rectly impact a company’s bottom line. This paper will focus on how perceived discrimi nation affects psychological health, physical health, and job attitudes. Perceived discrimination and ps ychological health outcomes Psychological health differs between me n and women. A meta analysis found that women tend to report higher stress levels co mpared to men, and women reported more depression, anxiety, and psychosomatic sympto ms compared to men (Davis, Matthews, Twamley, 1999). However, empirical research has generally failed to demonstrate an overall disadvantage in mental health for et hnic minorities. For example, the self-esteem of African Americans and Hispanics may be higher than European Americans under certain circumstances, such as when academic self-esteem is not included in the selfesteem measure (Twenge & Crocker, 2002). Thus, because non-whites are more often victims of discrimination, on the surface it ma y appear that pervasive discrimination has no impact on psychological heal th for ethnic minorities. However, there are a handful of studies that show well-being among minorities does vary when ethnic discrimination is cons idered. For example, a 13-year longitudinal

PAGE 45

10 study of the effects of every-day perceived racism and discrimination on mental health found that greater frequency of negative ra cial encounters and a general belief that “whites wanted to keep blacks down” were predictive of poorer subjective well-being (Jackson, Brown, Williams, Torres, Sellers & Brown, 1996). Additional studies have found that perceived ethnic discrimination is significantly relate d to high level of psychological distress (Brown, Sellers, Br own & Jackson, 1999; Williams & WilliamsMorris, 2000). Researchers have similar findings when inve stigating negative mental health outcomes specific to perceived disc rimination at work, and these studies have demonstrated how discrimination affects fe male targets. Women have been found to experience psychological distress in respons e to perceived workplace discrimination, even after controlling for prior emoti onal health (Pavalko, Mossakowski & Hamilton, 2003). Similarly, perceptions of discriminati on tend to harm well-being in women more so than men (Schmitt, Bramscombe, Kobrynowicz & Own, 2002). Bergman and Drasgow (2003) found a negative relations hip between sexual harassment and psychological well-being for white, bl ack, Latina, and As ian women. Everyday workplace discrimination and perceived prejudic e has also been negatively associated with well-being among African Americans (Deitch, Barsky, Butz, Chan, Brief, & Bradley, 2003; Branscombe, Schmitt & Harvey, 1999). Exclusion, which is a form of discrimination, has been linked to poor ps ychological well-being (Mor Barak & Levin, 2002). Klonoff Landrine & Campbell (2000) f ound that women who reported frequent sexist treatment had more depressive and anxious symptoms than women who reported little sexism. Thus, perceived discrimination is one stressor which may be experienced

PAGE 46

11 more frequently by minorities, and it ma y negatively impact psychological health outcomes for women and non-white minorities. Perceived discrimination and physical health outcomes Certain ethnic minority groups have a gr eater incidence of chronic illnesses (e.g. heart disease) and shorter life expectancies than white individuals (National Center for Health Statistics, 2007). Although lifestyle factors such as high-fat diets and lack of exercise are partly to blame, it has b een suggested that perceived racism and discrimination might help explain the ethnic differences in certain health outcomes for ethnic minorities (Anderson, McNeilly, & Myers, 1993; Brondolo, Rieppi, Kelly, & Gerin, 2003; Clark, et al., 1999) Studies have demonstrat ed that the perception of discrimination in general can have negative effects on health and well-being for ethnic minorities (Pavalko, Mossakowski, & Hamilton, 2003; Landrine & Klonoff, 1996). For example, one longitudinal study of the effect s of every-day percei ved discrimination on physical health found that greater frequency of reports of poor treatment due to race was positively associated with number of docto r-reported physical health problems and presence of health disability (Jackson, Br own, Williams, Torres, Sellers & Brown, 1996). There are also important gender differen ces in physical health. Pre-menopausal women have longer life expectancies and a lo wer incidence of certa in chronic illnesses compared to their male counterparts (National Center for Health Statistics, 2007). The “pre-menopausal” caveat is important because as some researcher s posit, there is a protective effect of female reproductive hormones (e.g. estrogen) against risk for coronary heart disease or other chronic illn esses. Gender differences lessen or disappear when post-menopausal women are considered. Because of the role biology plays in

PAGE 47

12 explaining sex differences in health, it may a ppear that sexism has no effect. However, Bergman and Drasgow (2003) found a negative relationship between sexual harassment and perceptions of health for white, blac k, Latina, and Asian women. Also, Klonoff, Landrine & Campbell (2000) found that women who reported frequent sexist treatment had more somatic symptoms (i.e. headache, gastrointestinal symptoms, and pain) than women who reported little sexism. It appears that perceived discriminati on negatively impacts physical health for both women and minorities. There are multiple explanations for group-level differences in physical health, and discriminatio n is likely one contributing factor. Perceived discrimination and job attitude outcomes The literature has demonstrated the ill effects of perceived discrimination on a number of job attitudes. Among them are le ss job involvement and career satisfaction, and fewer career prospects (Foley & Kidder, 2002; Perr y, Hendricks, & Broadbent, 2000; Sanchez & Brock, 1996; Shaffer, Joplin, Bell Lau & Oguz, 2000; Valentine, Silver & Twigg, 1999); greater work conflict, lower fee lings of power, and decreased job prestige (Gutek, Cohen & Tsui, 1996); and fewer orga nizational citizenship behaviors (Ensher, Grant-Vallone, & Donaldson, 2001). Taken toge ther, perceived discrimination is related to more negative job attitudes. Two of the more commonly studied j ob attitudes are job satisfaction and organizational commitment. Levitin, Quinn and Staines (1971) found an association between discrimination and job satisfac tion for women. Perceived race-based discrimination negatively impacts job satisfa ction as well (Valentine, Silver & Twigg, 1999). Hispanic employees were found to have lower job satisfaction and organizational

PAGE 48

13 commitment when perceiving discriminati on (Sanchez & Brock, 1996). Ensher et al. (2001) demonstrated that per ceived discrimination from supe rvisors, coworkers, and the organization itself impacted job satisfaction and organizational commitment in a sample of ethnically diverse blue-collar employees Also, gender discrimination has been negatively associated with job satisfaction and affective commitment (Shaffer et al., 2000). Other studies have investigated the relationship between discrimination and turnover, and have found a pos itive relationship between th e two. Shaffer et al. (2000) found a positive association between gender-based discrimination and turnover intentions. Also, a study with a sample of physicians found the perceived discrimination promotes turnover (Nunez-Smith, Pilgrim, W ynia, Desai, Bright, Krumholtz, & Bradley, 2009). Physicians belonging to a minority group were significantly more likely than majority-group physicians to have left a job at least once due to pe rceived discrimination. Situational factors at work th at signal possible discriminatio n may also promote turnover. While recruitment practices centering on divers ity may be effective at recruiting diverse candidates, they may actually increase tur nover when high expectations for a positive diversity climate are not fulfilled (McKay & Avery, 2005). Also, the paucity of diversity among upper management may increase turnover by indicating to minorities that there are no opportunities for advancement (Elvira & Cohen, 2001). Organization-level antecedents to perceived discrimination The research on perceived discrimination as whole is sparse and few studies have concentrated on the antecedents of perceived workplace discrimination. These antecedents are multi-faceted: influenced by a complex web of organizational, group, and

PAGE 49

14 individual-level factors. However, the curr ent paper will focus on only six organizational factors. These include Equal Employmen t Opportunity (EEO) practices, minority segmentation, diversity climate, instrumental social support, emotional social support, and token status. It is critical to note that only the perc eption of organization-level factors will be evaluated, not objective measures of these factors. Objectiv ely measuring these variables would be difficult even given unlimited resour ces within a single organization, let alone multiple organizations. While a comparison between objective and subject organizational antecedents to perceived discrimination would valuably contribute to the literature, it is beyond the scope of this project. Equal Employment Opportunity Today, it is more and more common th at an Equal Employment Opportunity (EEO) policy is part of an organization’s st rategy. Having EEO provisions means that an organization does not consider any protected class information (e.g. sex, race, color, religion) when making organizational decisions such as hiring or promotions (Gutman, 2000). Essentially, companies with a strong EE O policy take an identity-blind approach to organizational decisions. Under Title VII of the Civil Rights Act of 1964, no organization (with a few exceptions) should ma ke organizational decisions based on this information. However, an organization’s commitment to EEO, either formally or informally, will vary widely company to company. Research seems to indicate that the existence of EEO has a positive effect on discrimination; that is, a greater commitment to EEO is associated with less perceived discrimination. Including EEO as part of an organization’s strategic business plan

PAGE 50

15 provides a tangible indication th at discrimination is not tolerated in an organization (Morrison & Von Glinow, 1990). Adding EEO st atements to company publications (e.g. website or newsletters) that frame diversity as a source of competitive advantage promotes the belief within the organization th at diversity presents an opportunity rather than a burden (e.g. Cox & Blake, 1991). B ecause EEO policies are more group-statusblind, it is unlikely there will be large gr oup differences in how the presence of EEO impacts perceived discrimination. In other words, men, women, whites, and non-whites, will tend to perceive less discrimination when a company has a strong, visible EEO policy. Konrad and Linnehan (1995) explored this idea in a study of line managers’ attitudes towards human resource management practices. All line managers, regardless of gender or ethnicity, favored identity-blind pol icies over identity-conscious ones. To date no studies could be found on the relationship between EEO and psychological health, physical health, or job attitudes such as job satisfaction, organizational commitment, or intention to turn over. It is possible th at EEO policies have a positive impact on these outcomes given the fact that most employees tend to view these policies favorably. However, the dire ct impact of EEO on psychological and physical health, and job attit udes is likely very small. Minority Segmentation Minority segmentation occurs when minorit ies are relegated to certain jobs or areas within an organization. The occupati onal disadvantage for et hnic minorities has been well-documented (Kaufman, 2 001; Tomaskovic-Devey, 1993; Vaughn-Cooke, 1983). The marginalization and segregation of minorities to certain jobs is also well established (Collins, 1989; Kaufman, 2001; Steinberg, 1995). Usually these jobs are the

PAGE 51

16 least desirable, the lowest in terms of prestige and power, and contain the least opportunities for advancement (Kaufman, 2001), but it is also possible that these jobs have a decent income and high job securit y. However, when segmented jobs appear advantageous, they are often disadvantage ous in other ways (e.g., opportunities for advancement, location) and thus become ear marked for minorities (Steinberg, 1995). In one study among black managers, 67% reported th at they were holdi ng a racialized job (Collins, 1997). Other studies have revealed th at even higher-status positions set aside for minorities offer fewer opportunities to supervise production or non-minority subordinates, or to influence organizational policies (Cose, 1993; Feagin & Sikes, 1994; Zweigenhaft & Domoff, 1991, 1998, 2003). How minority segmentation will impact pe rceived discrimination is unclear. No studies could be located inve stigating a relationship betw een these two concepts. The bulk of the aforementioned studies are self-re port, illustrating that minorities are aware of the fact that they are relegated to certain jobs based on their group status. Thus, minorities will likely perceive more discrimination from the organization when their places of work funnel them into certain jobs or areas. However, there will likely be a negative relationship between minority segm entation and interper sonal discrimination from the viewpoint of minori ties. In other words, women and non-whites will experience more interpersonal discrimination as their workplace becomes more integrated because they will be dealing more with majority group members. Majority group members, on the other hand, will likely display a negative re lationship between minority segmentation and perceived organizational or interpersonal disc rimination, as their group will be favored in most situations as they become a numerical majority. Minority segmentation is difficult

PAGE 52

17 to observe directly, and some studies have relied on subjective measures. However, some researchers have found a relationship between objective and subjectiv e reports of this phenomenon (Gomez & Trierwe iler, 2001; Hammer & Green, 1998; Turner & Turner, 1981). The current study will use subjective measures of minority segmentation. It is possible there is a direct effect of minority segmentation on outcomes. A handful of studies suggest negative outcom es in terms of psychological health under conditions of minority segmentation. Severa l researchers have speculated that the abundance of white men in positions of aut hority may have a negative impact on women and non-white subordinates (Ely, 1994; Pfe ffer, 1989; Ridgeway, 1988). Specifically, Forman (2003) found a negative relationship between perceived raci al segmentation in the workplace and psychological well-being (i.e. self-efficacy) However, to date there have been no studies on minority segmenta tion in the workplace and physical health. The relationship between minority segm entation and job attitudes, like job satisfaction and organizational commitment, re mains largely unexplored in the literature. One exception is intention to turnover. Tur nover among minorities ma y actually increase as segmentation increases. J obs with higher concentrations of Latina/os and African Americans have less status, lower pay, le ss influence, and fewer opportunities for advancement (Chatman & O’Reilly, 2004; Tsui et al., 1992). Toma skovik-Devey (1993) found that job quality improved as white ma le incumbents increased. Thus, minority segmentation may signal turnover for minor ity individuals. However, as minority segmentation increases, majority group memb ers are funneled to the more advantageous positions. Majority group members will be prob ably be less likely to turnover as minority segmentation increases.

PAGE 53

18 Perceived Diversity Climate An organization’s diversity climate ca n be defined as employees’ shared perceptions of the polices, practices, and pro cedures that both implicitly and explicitly communicate the extent to which maintaining an inclusive environment for all employees is a priority (Dipboye & Colella 2005). In a positive diversit y climate, group identities have no weight in organizational functio ning. Here, human resource functions are consistently enacted, minorities exist at all levels of the or ganization, and prejudice is not rewarded. Discrimination should be less comm on within a positive diversity climate, compared to a company who does not value inclusion. Although it is possible to examine diversity climate at the organizational level of analysis, most studies have focused on indivi dual level perceptions. This approach is useful because if minorities believe that their organization fosters bias, then this is the “reality” that the organization should tend to, as employees behavior reflects their perceptions (Weick, 1995). Research has shown that women and ethnic minorities tend to perceive a poorer diversity climate then their majority counterparts; with lower levels of inclusion, bias in informal processes, lost opportunities because of bias, and insufficient attention paid to diversity (Kossek & Z onia, 1993; Mor Barak, Cherin, & Berkman, 1998). Organizations with positive diversity c limates tend to exhibi t lower levels of actual discrimination (Cox, 1994). This is likel y due to a heightened commitment towards maintaining a diverse workforce. However, the relationship between diversity climate and perceived discrimination has yet to be established. There is likely a negative relationship between diversity climate a nd perceived discrimination, as a positive

PAGE 54

19 diversity climate would entail organizatio nal efforts to minimize discrimination. The current study will explore this possibility. Further, diversity climate perceptions are related to job satisfaction and organizational commitment (Hicks-Clarke & Il es, 2000). As minorities are more likely to experience discrimination and thus benefit more from diversity efforts, a positive diversity climate will likely have positive eff ects on job attitudes for minority employees. Negative racial conditions within an or ganization have been shown to undermine minorities’ job attitudes (Chrobot-Mason, 2003; Foley, Kidder, & Powell 2002), which tend to be precursors to attrition (Griff eth & Hom, 2001). How diversity climate will impact job attitudes for majority group memb ers is unclear. Majority group members are generally less likely to experience discri mination, and thus the strength of an organization’s diversity climate may be inc onsequential. One study found that diversity climate was significantly and negatively rela ted to turnover intentions for all groups, although the relationship was stronger for bl ack employees (McKay, Avery, Tonidandel, Morris, Hernandez, & Hebl, 2007). Thus, dive rsity climate could have a positive impact on job attitudes for all groups. Conversely, aversive racism theory (Dovidio, Gaertner, Kawakami, & Hodson, 2002; McConahay, 1983) suggests that majority group members may have a subconscious aversion to mi nority group members and thus may respond negatively to diversity efforts. A positive di versity climate is c ounter to their selfinterests as it is benefiting other groups; job satisfaction, organizat ional commitment, and willingness to remain at an organization may suffer as a result. The direct effect of diversity climate on psychological and ph ysical outcomes has been previously unexplored in th e literature. It is possible th at a positive diversity climate

PAGE 55

20 will have a positive impact on health outcomes just as the effect is positive on job attitudes. Instrumental and Emotional Social Support One of the defining features of employm ent for women and non-white individuals seems to be the lack of similar individuals with whom to interact (Ibarra, 1993). A shortage of like others may mean that minor ities are missing an important aspect of work life: social support. There are various forms of social support. First, instrumental support provides a minority worker with information or resources. Access to informal networks or the presence of a mentor may provide inst rumental social support. Emotional social support, on the other hand, provides minor ity individuals a venue for venting on workplace issues, an opportunity to relate to co-workers on a personal level, and a chance to go beyond appearances and reveal personal qu alities to others in the workplace. Access to informal networks and a mentor may sim ilarly provide this emotional social support, but it may also originate from coworkers or a supervisor. Empirical evidence from several studies shows how minorities are shortchanged in both instrumental and emotional social support. Racial and ethnic minorities have lower participation rates in informal networks compared to white individuals (Lincoln & Miller, 1979). Some of this absence may be voluntary, but not all of it. A common complaint among women and ethnic minorities is limited access to or exclusion from informal interaction networks (Miller, 1986; Morrison & Von Glinow, 1990; O’Leary & Ickovics, 1992). Similarly, a consistent findi ng in interviews with women and ethnic minorities was the importance of informal processes in promotion opportunities in addition to the barriers they faced in gain ing access to these informal processes (Mor

PAGE 56

21 Barak et al., 1998). Apparently, th ese individuals feel that or ganizations not only tolerate but create barriers to promotion through info rmal resources and power networks. Access to these networks provides instrumental re sources such as information, which can be critical for capitalizing on promotion opport unities. Further, ther e are interpersonal resources such as friendship, social support, and “face time” with decision makers which can improve well-being and also improve promotion potential (e.g. Ibarra, 1993). Thus, an absence from these informal networks, be it volitional or not, can be detrimental to career advancement. There is also eviden ce that minorities tend to receive fewer opportunities for training and development that may prepare them for advancement (Alderfer, Alderfer, Tucker & Tucker 1980; Nixon, 1985; Jeanquart-Barone, 1996). Without access to informal networks, minoritie s may be unaware of opportunities. Black employees also tend to believe that they do not receive as mu ch critical career information as their white counterparts (Alder fer et al., 1980). Wome n also tend to have less access to organizational information co mpared to men (Alderfer, 1987). Mentors provide support themselves, but may also provide access to informal networks. A performance barrier in the workplace indicated by female securities employees was lack of a mentor (Roth, 2004). Sponsorship can be crucial for career success for any individual, and especially so for minority i ndividuals (Kanter, 1997; Yoder et al., 1985). The gender and the race of the mentor may in fluence their efficacy. White male mentors offer substantial benefits over women and non-white employees (Dreher & Ash, 1990; Dreher & Chargois, 199 8; Dreher & Cox, 1996). The absence of either instrumental or emotional social support may signal the presence of discrimination for minorities; especially when others in the work

PAGE 57

22 environment are receiving it. For example, one study found a negative relationship between social support and perceived race-b ased discrimination, where perceptions of discrimination increased as social support de creased (Prelow, Mosher, Bowman, 2006). There is also evidence of a relations hip between social support and work outcomes. For example, black employees tend to view factors relating to an individual in their work environment, such as their supervis or, as their main source of job stress more so than white individuals (Stroman & Selt zer, 1991). Another study demonstrated that lack of emotional support from supervisors was the strongest predic tor of negative work outcomes (i.e. job stress) for black employees (Ford, 1985). There is a direct link between social support and psychological well-being. Support provided by one’s social network is important for maintaining a person’s well-bei ng (Babin & Boles, 1996). Social support at work has also been associated with lower le vels of emotional exha ustion and anxiety in addition to improved mental health (Snow & Kline, 1995). Workplace support may also decrease perceived stressors, improve gene ral well-being, and redu ce job dissatisfaction (Gant et al., 1993). Taken together, there is evidence that lack of social support is related to perceived discrimination. The absence of instrumental and emotional social support is potentially problematic not only because it may serve as a signal for possible discrimination, but also because it may negatively impact health and job attitudes. Token Status Kanter (1977) defines token status as an individual who shares group status with less than 15% of fellow employees. For ex ample, a single woman in a workplace composed of seven men would be a token employee. Kanter originally formed her

PAGE 58

23 tokenism theory around women, although it has b een expanded to other minorities as the theory was designed with numerical imbalances in mind rather than physical differences. Kanter posited that there are three comm on token experiences in the workplace: heightened visibility in the work environment, which escalates performance pressures; isolation from social and professional ne tworks, which serves to exaggerate the differences between tokens and the majo rity group (i.e. contrast); and “role encapsulation”, where tokens are forced into stereotypical script s in order to make themselves understandable to the major ity group. These common experiences may increase the perception of discriminati on for women and non-white tokens. Relationships between tokenism and sexand race-based discrimination have been found. One study on forty-four female firefi ghters (all tokens with in their respective units) found that all but three reported experiencing sexist ev ents within the last year (Yoder & McDonald, 1998). Experiencing sexist discrimination was positively associated with visibility of mistakes and negatively a ssociated with colleagueship. Moreover, being the first woman in a firehouse was positivel y related to sexist discrimination and perceptions of differential treatment. Fema le lawyers in token positions have also reported higher levels of sexist behavior than non-tokens (Ros enberg, Perlstadt, & Phillips, 1993). The relationship between t okenism and discrimination has also been found among non-whites. For example, fort y-four African-American journalists identified tokenism as one of the many di scriminatory practices they face in the newsroom (Shafer, 1993). Although Kanter (1977) originally stat ed that tokenism depends more on numerical imbalances than group status, studie s have shown that group status does have

PAGE 59

24 an impact. The consequences of tokenism ma y depend on the relative status of the tokens compared to the majority group (Alexander & Thoits, 1985; Yoder, 1991). For example, male tokens in female-dominated occupations actually receive adva ntages, such as in promotions and pay (Fairhurst & Snavel y, 1983; Floge & Merrill, 1985; Williams, 1992, 1995). Similarly, Fairhurst and Snavely (1983) found that men in nursing school (i.e. tokens) did not report feeling so cially isolated or under greate r pressure to perform. This may be due to homophily preferences, or the tendency for individuals to associate with others like themselves (Berscheid & Hatf ield, 1978; McPherson, Smith-Lovin, & Cook, 2001; Smith-Lovin & McPherson, 1993). If i ndividuals in power, such as upper managers, are male, they may provide preferential treatment to men in token positions. Thus, white male tokens may not experien ce the same discrimination as women or nonwhite individuals. On the other hand, thes e same homophily preferences explain why female and non-white tokens experience di fferential treatment both socially and structurally. Roth (2004) posited this idea in her qualitative work with male and female wall-street employees. Interviewees revealed that homophily preferences were often a barrier to performance, advancement, and bonus compensation, which was based on performance appraisals. Thus, token status is necessary but not sufficient to produce perceptions of discrimination; one must also be a gender or ethnic minority. Given this information, token status will likely be pos itively associated with discrimination for minorities, whereas there will be a negative or a nonexistent relationship between token status and discrimination fo r majority group members. Tokens should experience more work st ress and psychological symptoms than non-tokens (Kanter, 1977). Jackson, Thoits and Taylor (1995) demonstrated how

PAGE 60

25 tokenism is related to psychological well-bei ng. Racial tokens experi enced high levels of depression while gender tokens experienced high levels of anxiety. Also, racial tokens experienced “token stress”, which is charac terized by loss of black identity, multiple demands of being black, a se nse of isolation, and having to show greater competence. Also, greater role overload, a work stressor, was associated with gender token status. Token female police officers experienced less job satisfaction and more jobrelated depression compared to non-token fe male officers (Krimmel & Gormley, 2003). However, male tokens may also experience negative job attitudes. One study found that token male flight attendants, a position do minated by females, experienced lower job satisfaction, less organizational commitment and higher intentions to quit (Young & James, 2001). This is consistent with othe r research demonstrating traditional majority group members experiencing less affective organizational co mmitment when they find themselves with minority status (Chattopa dhyay, 1999; Tsui et al ., 1992). Neimann and Dovidio (1998) found black, Latina/o, and Asian professors were less satisfied with their jobs when they were the solo minority in an academic setting, although the relationship between solo status and job satisfaction was mediated by distinctiveness, or the stigmatizing feelings of minorities in a token status. There is also research on the link between minority tokenism and turnover. Hom, Roberson, and Ellis (2008) analyzed attrition information from 20 American firms contributing data on over 400,000 professional a nd managerial workers from a variety of demographic groups. They found that women quit more than men overall and racial minorities quit more than white Americans. Moreover, Asian American, black, and Latina women quit more than their male counterparts and white Americans as a whole,

PAGE 61

26 although this effect was eliminated after te nure was controlled. Interestingly, being a female numerical minority in a male-dominat ed job did not result in greater turnover, although men do tend to turnover more as female incumbents increase. These results are mirrored in a study by Chatman and O’Reilly (2004). Men are less likely to leave allmale or male-dominated groups, and are mo re likely to leave female-dominated or balanced groups (i.e. groups with equal amounts of men and wo men). Women, on the other hand, were most likely to leave all-women or balanced groups and least likely to leave femaleor male-dominated groups. Thes e findings have implications for minority segmentation or tokenism, and turnover. Turnov er will likely increase for men and white individuals as they become tokens in a minority-dominated workforce. However, turnover will decrease for women and minorities as they become tokens because they will view themselves as holding a position of status along with majority group members (i.e. white men). Summary The purpose of the current study is to inve stigate the mediating role of perceived discrimination in the relationship between organizational antecedents and work outcomes. The organizational antecedents wi ll include employee perceptions of EEO policy, minority segmentation, diversity climate, instrumental social support, emotional social support, and token status. The out comes will include self-reported psychological health, physical health, and job attitudes (j ob satisfaction, organizational commitment, and intention to turnover). An alyses testing six different c onceptual hypotheses will be conducted separately for comparing women to men, and white to nonwhite participants. Additionally, two types of perceived discrimination will be analyzed: subtle and overt.

PAGE 62

27 Note that in each of the hypotheses below, pr edictions are made regarding the directions of paths in the models. For some hypotheses, the predicted directi ons are different for minority and majority group members. However, in every case, the relationship between perceived discrimination and job satisfacti on, organization commitment, physical health, and psychological health is predicted to be negative while the relationship between perceived discrimination and turnover intention should be positive. Equal Employment Opportunity policy is an identity-blind approach to company practices such as hiring and promotions. Individuals from all groups should be treated equally when a company has a strong EEO policy. Thus, a stronger EEO policy will relate to less perceived di scrimination for both minority a nd majority group members, which will in turn relate to more positive job attitudes, and psychological and physical health outcomes. Hypothesis 1 Perceived discrimination mediates the re lationship between EEO policy and work outcomes. There will be a negative rela tionship between EEO policy and perceived discrimination for both minority and major ity group members, although the relationship should be stronger for minorities. Simila rly, both groups will demonstrate a positive relationship between EEO policy and psyc hological and physica l health, and job satisfaction and organizational commitment in addition to a negative relationship between EEO policy and intention to turnover. Thus, minority status should not moderate the relationship between EEO polic y and perceived discrimination. Minority segmentation occurs when wo men and non-whites are relegated to certain areas or jobs within a company. Gr eater minority segmentation favors majority-

PAGE 63

28 group members, but may serve as a signal to possible discrimination for minorities. Further, minority segmentation may negativ ely impact health, job satisfaction, and organizational commitment for minorities, bu t will have the opposite effect for whites and men. Minority segmentation will increase in tention to turnover among minorities, but will decrease intention to turnover among ma jority-group members. When minority segmentation is prevalent with in a company, minority indivi duals tend to relegated to lower-status jobs while majority group members are placed in positions of higher prestige. Thus, as minority segmentation in creases, outcomes are better for white and men, but worse for women and non-white persons. Hypothesis 2 Perceived discrimination mediates the relationship between minority segmentation and work outcomes. Minorities will perceive more discrimination as minority segmentation increases, whereas majority group members will display a negative relationship between minority se gmentation and perceived discrimination. Minority group members will report lower job satisfaction, organiza tional commitment, psychological health, physical health, and greater intention to turnover as reported minority segmentation increases. The oppos ite will be found for majority group members. Thus, minority status should m oderate the relationship between minority segmentation and perceived discrimination. Diversity climate reflects shared percep tions as to the extent to which an organization values the inclusion of indivi duals from all groups through its policies and procedures. Minorities are the lik ely beneficiaries of such efforts. Thus, a strong diversity climate is likely to lead to less perceived discrimination and more positive outcomes in

PAGE 64

29 terms of job attitudes, psychological health, and physical health for minorities, in addition to a decreased desire to turnover. Howeve r, it is unclear how majority-group members react to diversity efforts. They are less lik ely to benefit from such policies and it is possible they would react negatively to the inclusion of more out -group individuals due to possible losses in benefits. It is possible they will have worse health and job attitudes outcomes in addition to an increase d likelihood to intend to turnover. Hypothesis 3 Perceived discrimination mediates the re lationship between diversity climate and work outcomes. Diversity climate is negativ ely related to perceived discrimination for minority group members. These individuals wi ll also experience grea ter job satisfaction, organizational commitment, psychological hea lth, physical health, a nd less intention to turnover as diversity climate improves. Ma jority group members, on the other hand, will report a positive relationship between diversity climate and perceived discrimination. There will also be a negative relationship be tween diversity climat e and job attitudes, psychological health, and physical health, in addition to a positive relationship with intention to turnover. Instrumental social support includes re sources (e.g. information) provided to workers from supervisors or co-workers. Wh en minority group members fail to receive these resources, they are more likely to report their group status as a reason. There will be a negative relationship between instrumental social support and perceived discrimination for minority group members. Further, the lack of instrumental social support will result in negative health and j ob attitude outcomes, in addition to an increased intention to turnover for minority individuals. However, beca use the lack of social support serves as

PAGE 65

30 signal for discrimination for minority but not majority group members, there will be no relationship between instrumental social s upport and perceived discrimination for this group, although majority group members will still experience negative outcomes as instrumental support decreases. When majority group members fail to receive instrumental social support, th ey are more likely to attribute this failure to reasons other than group status. Hypothesis 4 Perceived discrimination will mediate the relationship between instrumental social support and work outcomes, but only for minor ity respondents. Mino rities will perceive greater discrimination as instrumental social support decreases while men and whites will report no relationship between level of instrumental suppo rt and discrimination. All groups will report lower levels of job satisfaction, organizational commitment, psychological health, physical health, and grea ter intention to turnover as instrumental social support decreases. Emotional social support includes aff ective resources (e.g. opportunity for venting, advice) provided to workers from s upervisors or co-workers. There will be a negative relationship between emotional soci al support and perceived discrimination for minority group members. Similar to instrume ntal social support, minority individuals tend to link a lack of emotiona l social support to their group st atus. Further, the lack of emotional social support will result in negativ e health and job attitude outcomes, in addition to an increased intention to turnove r. However, because the lack of social support serves as signal for discrimina tion for minorities but not majority group members, there will be no relationship be tween emotional social support and perceived

PAGE 66

31 discrimination for this group, although majo rity group members will still experience negative outcomes. Hypothesis 5 Perceived discrimination will mediate the relationship between emotional social support and work outcomes, but only for minor ity participants. Minor ities will perceive greater discrimination as emotional social support decreases while men and whites will report no relationship between level of emotional support an d discrimination. All groups will report lower levels of job satisfacti on, organizational commitment, psychological health, physical health, and gr eater intention to turnover as emotional social support decreases. An employee is a “token” when they are pa rt of a group that comprises less than 15% of the workforce within their organization. Tokenism is associated with perceived discrimination among minorities in add ition to negative j ob satisfaction and organizational commitment, and psychologica l/physical health outcomes. Conversely, majority-group tokens are less likely to perc eive discrimination as they still hold a position of power due to their status in societ y. However, majority t okens are also likely to have lower job satisfaction and organiza tional commitment, as research has shown these negative outcomes for majority gr oup members who find themselves surrounded by minority coworkers Additionally, predictions re garding intention to turnover will likely be opposite those of the other j ob attitudes. Minority tokens will be more likely to intend to stay in a job as they are surrounded by indi viduals with a high-stat us and will perceive themselves as holding a higher-status positi on. Conversely, majority-group tokens will be

PAGE 67

32 more likely to intend to turnover because they will perceive themselves as being surrounded by lower-status co-workers and thus holding a lower-status position. Hypothesis 6 Perceived discrimination will mediate the relationship between token status and work outcomes. There will be a positive relationship between tokenism and perceived discrimination among minority individuals, but a negative relationship for majority individuals. The relationship between toke nism and job satisfaction, organizational commitment, psychological health, and physical health will be negative for minority participants, but the relations hip will be positive with inte nt to turnover. Majority individuals will also report lower levels of job satisfaction, organizational commitment, psychological health, and physical health as t okenism increases, but these individuals will report greater intention to turnover. Thus, mi nority status will moderate the relationship between token status and perceived discrimination.

PAGE 68

33 Chapter Two Method Procedure The study occurred in two phases: a pilot study and full demonstration. The purpose of the pilot study was to gather data for an item analysis. Several of the measures intended for use in the full demonstration ha d been created specifically for the current study, or had limited reliability and validity in formation. Participants were recruited for the pilot study, and only after items had been evaluated based on the item analysis were participants recruited for the full demonstration. For both the pilot study and the full dem onstration, participants completed all measures online during a single session. Some of the participants were recruited through the USF Psychology Department Participant Pool while others were participants from previous studies recruited via an email. For individuals who participated via the subject pool, the study was posted on the SONA website, visible to indivi duals who had selfreported working more than 10 hours per w eek. A qualified indivi dual could elect to participate in the study. The remaining indivi duals received an email which contained a short message explaining the scope of the study. Regardless of how participants were recruited, all were directed to a link for the website containing the measures. Directions for the survey were provided when participan ts first opened the we b page containing the survey.

PAGE 69

34 When participants completed the full surve y, the final screen ga ve participants the option of forwarding a message to a coworker Participants were directed to send the survey to a coworker who was preferably in the same position as themselves. The message contained a link to another surve y, when contained items assessing the work environment only. The purpose of this survey was to get second-source data on the environmental conditions of the participants’ workplace. Participants A minority, for the purposes of this study, is defined as any individual who selfidentifies as a women or non-white individual. White individuals and males are considered majority group members. A total of 726 individuals we re recruited to participate in this study. This number includes individuals who participated in the pilot study ( n = 240), demonstration ( n = 457), and optional coworker survey ( n = 29). Some overlap between the pilot study and demonstration participants was possible, bu t because participation was anonymous it is impossible to quantify. Additi onally, analyses may not include every individual who participated in the study as some responde nts did not complete the entire survey. Two-hundred and forty individuals partic ipated in the pilot study. Sixty-five percent of the sample was female, and 64% of the sample included white individuals. The average age was 30.15 years ( SD = 14.02). Participants wo rked 32.15 hours per week on average ( SD = 13.82), were in their job about 5.35 years ( SD = 7.67), were predominately in the service industry (32%), and ha d not yet completed college (41%). Four-hundred and fifty seven individuals participated in the full demonstration. However, all respondents were not included in analyses. Several pe ople did not indicate

PAGE 70

35 their gender ( n = 24) or ethnicity ( n = 22), or did not complete enough of the survey to be included in all of the analyses The average age was 24.50 years ( SD = 8.56). Participants worked 28.20 hours per week on average ( SD = 12.38), were in their position about 3 years ( SD = 4.49), were predominately in the service industry (34.71%), and had completed some college (51.16%). There were a total of 298 women and 135 men, as well as 236 whites and 197 non-whites. Since part of the analyses compared majority and minority group members, and these groups are defined two ways (i.e. women vs. men and whites vs. nonwhites), it is of interest to know how the balance of the sample. Table 1 displays how many individuals fall into each category. Table 1. Comparison of Minority and Majority group Members MenWomenTotal Whites 70166236 Non-whites 65132197 Total 135298433 There were some significant differences on the demographic variables between the minority and majority group me mbers. White participants ( M = 25.42) were significantly older than n on-white participants ( M = 23.39; F = 5.10, p < .05). White participants ( M = 29.45) also worked more hours pe r week than non-white respondents ( M = 26.70; F = 5.30, p < .05). There were also diffe rences between men and women. Men ( M = 32.18) similarly tended to work longe r hours than their female counterparts ( M

PAGE 71

36 = 26.40; F = 20.93, p < .01). Men (M = 3 .70 years) also held their positions significantly longer than women ( M = 2.65; F = 5.19, p < .05). A complete comparison of demographic information for the full demonstration participants can be found in Table 2 and Table 3. Table 2. Group Comparison on Continuous Demographic Variables Overall Men Women Whites Non-Whites M SD M SD M SD M SD M SD Age 24.50 8.56 25.218.4424.208.6525.429.30 23.297.36 Work hours 28.20 12.38 32.1815.1826.4010.4329.4512.58 26.7011.99Tenure 3.00 4.49 3.704.732.654.333.254.76 2.694.13

PAGE 72

37 Table 3. Group Comparison on Categoric al Demographic Variables Overall Men Women Whites NonWhites Education Some HS 0.69%0.00%1.01%0.43% 1.02% HS diploma/GED 5.56%2.99%6.76%3.40% 8.12% Some college 51.16%49.25%52.03%48.94% 53.81% AA degree 21.76%18.66%22.97%22.13% 21.32% Bachelor’s degree 10.65%14.18%9.12%13.62% 7.11% Master’s degree 6.71%9.70%5.41%7.66% 5.58% Doctorate 3.47%5.22%2.70%3.83% 3.05% Industry Business/Financial 8.74%11.85%7.38%11.81% 5.05% Education 8.97%2.96%11.74%6.75% 11.62% Healthcare 10.34%10.37%10.07%8.44% 12.63% Culture/Arts 2.99%1.48%3.69%3.38% 2.53% Service 34.71%25.19%38.93%38.82% 29.80% IT/Computer 3.91%7.41%2.35%4.22% 3.54% Blue Collar 3.68%5.93%2.68%3.80% 3.54% Other 26.67%34.81%23.15%22.78% 31.31% Twenty-nine individuals participated in the coworker survey. Eight of these individuals had to be excluded because they could not be matched with a demonstration study participant. Eighty-six pe rcent of the sample were women, and 81% of the sample included white individuals. Th e average age was 29.63 years ( SD = 10.51). Participants worked 36.33 hours per week on average ( SD = 10.49), were in their job about 4.83 years

PAGE 73

38 ( SD = 9.26), were predominately in the edu cation industry (19%), and had earned a Bachelor’s degree (76%). Participants were recruited onlin e through several mechanisms. Employed individuals were identified th rough the Psychology Participant Pool at the University of South Florida. Also, the principa l investigator has two contact lists from previous studies: one containing the email addresses of over 2,000 minority employed individuals and another containing over 1,000 email addresses of employed individuals in the Tampa Bay area. Participation was voluntary and anonymous. Remuneration was provided for individuals who participated through the Psychology department participant pool only, in the form of partial course credit. Measures Demographics Information on gender, ethnicity, age, hi ghest level of education achieved, hours worked per week, job tenure, and industry ty pe was collected. Unle ss otherwise noted, all other measures include a 5-point Likert-type response scale where 1 = Strongly disagree, 2 = Disagree, 3 = Neither agree nor disagr ee, 4 = Agree, and 5 = Strongly Agree. Measures are scored by adding the responses for each scale. Equal Employment Opportunity The extent to which the organizatio n is perceived as emphasizing equal employment opportunity was measured with 4 items. An example item is “Senior managers emphasize Equal Employment Opportunity” These items were created specifically for this study, however inspir ation was gleaned from Parker, Baltes & Christiansen (1997).

PAGE 74

39 Perceived Minority Segmentation Employee perceptions of how the orga nization allocates positions based on minority status was measured with three item s. No measure could be located to measure this concept, therefore these items were de veloped as modifications of items from two large surveys of black indi viduals. The National Survey of Black Americans (NSBA) asked two dichotomous items designed to meas ure perceived racial segmentation: “In the place where you work, do Black people tend to ge t certain kinds of jobs?” and “Is your job one that Black people tend to get more than whites?” The NSBA is a series of studies initiated in 1977, developed by the Program for Research on Black Americans at the Institute for Social Research, University of Michigan. Similarly, the 1995 Detroit Area Study (DAS) asked a single dichotomous item: “Do you think your job is one that people of your ethnic or racial group tend to get more than peopl e of other groups?” The 1995 DAS is one of a series of studies from the Su rvey Research Center and the Department of Sociology at the University of Michigan. Each DAS poses a unique set of research questions and is headed by di fferent principal investigator s each year. The 1995 DAS was headed by James Jackson and David Williams. From these items, three items were developed to be measured on a Likert-type sc ale and to be applicable to women and nonwhite minorities. These items are: “At my organization, minorities tend to get certain types of jobs”, “My job is one that tends to be given to minorities”, and “At my organization, minorities tend to be assi gned to certain areas/departments”. Diversity climate

PAGE 75

40 Climate for diversity was measured with nine items (McKay et al., 2007). Note that the wording of the items was modified sl ightly to make more sense with the rating scale of current study. Reported alpha for this measure was 0.91. Perceived instrument al social support Instrumental social support at work wa s measured with four items. These items have been developed by Caplan, Cobb, Fr ench, Van Harrison, and Pinneau (1980). Instrumental social support describes the ex tent to which resources are provided from other people at work. An example item it “M y immediate supervisor goes out of his/her way to make things easier for you at work. Participants are aske d each question twice: once for their immediate supervisor and once fo r “other people at work”. In addition to the 5-point Likert-type scale for the item, ther e is a “not applicable” option in the event that a participant does not have a supervisor or coworkers. Alpha for the supervisor scale has ranged from .86 to .91 (Lee & Ashfor th, 1993; Repeti & Cosmas, 1991). Alpha for the co-worker subscale has been reported at .79 (Repeti & Cosmos, 1991). Lim (1996) also reported an alpha of .80 for the combined supervisor/co-worker scale. Additionally, three items were asked regarding particip ant’s access to informal networks and mentoring. These items were primarily expl oratory and have been developed for use within this project. Because it was determ ined that they added little value to the measurement of instrumental soci al support, they were left out of the analyses in favor of keeping the well-validated original measure intact. Perceived emotional social support Emotional social support was measured w ith four items. These items have been developed by Caplan et al. ( 1980). Emotional social support de scribes the extent to which

PAGE 76

41 individuals have an outlet at work to vent their personal problems. An example item is “My immediate supervisor is easy to talk to”. Participants are asked each question twice: once for their immediate supervisor and once fo r “other people at work”. In addition to the 5-point Likert-type scale for the item, ther e is a “not applicable” option in the event that a participant does not have a supervisor or coworkers. Alpha for the supervisor scale has ranged from .86 to .91 (Lee & Ashfor th, 1993; Repeti & Cosmas, 1991). Alpha for the co-worker subscale has been reported at .79 (Repeti & Cosmos, 1991). Lim (1996) also reported an alpha of .80 for the combined supervisor/co-worker scale. Token Status Kanter (1977) defines a t oken as an individual whose group status (e.g. gender or ethnicity) comprises less than 15% of the total group com position. Since obtaining this information objectively within the context of this study is impossible given the anonymous participation, subjects will be asked to estimate the group composition of their workplace. Their token status both with in the organization as a whole and within their work group/department was estimated. Four items have been created for this purpose. The items direct participants to repo rt what percentage of their coworkers share their gender or ethnicity either within their workgroup or enti re physical location of their workplace. The response scale is as follows: ‘Less than 15%’, ‘At least 15%’, ‘At least 25%’, ‘At least 50%’, ‘At le ast 75%’, ‘I don’t know/Not appl icable’. Individuals who chose “Less than 15%” are classified as either an ethnic or gender t oken, while all others are non-tokens. Only token status within the group was used in analyses, however. Token status within the workplace was asked mainly for exploratory purposes. Fewer individuals identified themselves as gender or racial token within the workplace as a

PAGE 77

42 whole, compared to the workgroup. Further, tokenism may become more salient and the effects of tokenism may become more inte nse as the group becomes smaller. For these reasons, only work group tokenism was included in the mediation models. Perceived discrimination at work (overt) No single measure could be found whic h satisfactorily measured perceived discrimination within the workplace from all sources (supervisors, coworkers, and the organization itself). Thus items were taken from three previously used measures and modified to meet the needs of the current study. Four items were created based on Landrine and Klonoff’s (1996) The Schedule of Racist Events. Six items were created based on items used by Ensher, Grant-Vallone and Donaldson (2001) Also, eight items were created based on the Perceived Discri mination Scale used by Sanchez and Brock (1996). A total of eighteen ite ms measured the extent to which employees perceive racebased and sex-based discrimination in the workplace. Perceived discrimination at work (subtle) The measure of subtle perceived discrimi nation at work was a slight modification of an existing measure developed by Willia ms, Yu, Jackson, and Anderson (1997). The original measure asked participants to reco rd how often they experienced various forms of poor interpersonal treatment (e.g. treated w ith less respect). After rating ten forms of interpersonal treatment, responde nts indicated which of nine characteristics (e.g. race, gender, age, etc.) were reasons for the poor treatment. Participants could choose all or none of the characteristics. The modificat ion for the current study involved adding the phrase “because of your gender” or “because of your race/ethnicity” to the end of the first ten items. For example, “You are treated w ith less courtesy than other people” was

PAGE 78

43 changed to “You are treated with less courtesy than other people because of your gender” for the current study. Every-day Perceived Discrimination A measure of general perceived discrimina tion was used in the pilot study only to evaluate the discriminant validity of the pe rceived discrimination at work scales. It consisted of 15 items adapted from Landr ine and Klonoff’s (1996) The Schedule of Racist Events. Although this scale was developed for race -based discrimination, items were slightly modified to inquire about gender-based discrimination. Job satisfaction Global job satisfaction was measured w ith a 3-item measure developed by Cammann, Fichman, Jenkins, & Klesh (1983). An example item is “All in all, I am satisfied with my job”. Seve ral studies have provided al pha estimates from .67 to .95 (McFarlin & Rice, 1992; McLain, 1995; Pear son, 1991; Sanchez & Brock, 1996; Siegall & McDonald, 1995). Organizational Commitment Overall organizational commitment was measured with nine items (Mowday, Steers, & Porter, 1979). The scale has a dem onstrated coefficient alpha ranging from .74 to .92 (see Fields, 2002 for a review), and test-r etest reliability has been shown to be 0.74 (Vandenberg & Lance, 1992). Intention to Turnover Intention to turnover was measured with a single item, “How often have you seriously considered quitti ng your present job?”. Partic ipants respond on a 6-point

PAGE 79

44 response scale where 1 = Never, 2 = Rarel y, 3 = Sometimes, 4 = Somewhat often, 5 = Quite often, and 6 = Extremely often. Physical Health Part of the SF-36 Health Survey (Med ical Outcomes Trust, 1992) assessed physical health. The full measure contains 36 items comprising ei ght subscales which measure both physical and psychological heal th. One can assess either physical or psychological health by breaking the scale in to two composite scores. The Physical Composite Score is comprised of the follo wing subscales: physical functioning (10 items), role limitations due to physical proble ms (4 items), pain (2 items), and general health perceptions (5 items). The measur e has demonstrated internal consistency reliability coefficients between 0.60 to 0.94 a nd test-retest from .43 to .90 (see Bowling, 1997 for a review). Also, the response scale varies widely within the measure and is not consistent with the other measures in the current study. See Appendix A for more information. Psychological Health Part of the SF-36 Health Survey (Medi cal Outcomes Trust, 1992) was used to assess psychological health. The following subs cales were used to measure psychological health: social functioning (2 items), role lim itations due to emotional problems (3 items), mental health/well-being (5 items), and en ergy/vitality (4 items). The measure has demonstrated internal consistency reliabi lity coefficients between 0.60 to 0.94 and testretest from .43 to .90 (see Bowling, 1997 for a review). Also, the response scale varies widely within the measure and is not consiste nt with the other measures in the current study. See Appendix A for more information.

PAGE 80

45 Data Analysis All six hypotheses were initially tested with moderated mediation. In each case, majority status (majority vs. minority group me mber) was evaluated as a moderator in the relationship between organizational ant ecedent (e.g. EEO status) and perceived discrimination. Further, perceived discrimina tion was evaluated as a mediator in the relationship between environmental condition and outcomes (e.g. intention to turnover). The analysis was conducted via a method de scribed in Preacher, Rucker, and Hayes (2007). The paper directed readers to a website which provides a SPSS macro for conducting the analyses. This macro was downl oaded and used in the current study. The macro provides multiple regression information including a mediator model and a dependent variable model. In the mediator model, the effects of the predictor, the moderator, and their interacti on on the mediator are displayed. This is where the presence of moderation is determined. In the dependent variable model, the effects of the mediator, the moderator, the independent variable, a nd the interaction betw een the independent variable and the moderator are displayed. In addition to results of the multiple regressions, the conditional indi rect effects for the major ity and minority group members are displayed. Following the moderated mediation, tests of simple mediation were conducted separately for minority and majority group members. A SAS macro developed by Preacher and Hayes (2004) was downloaded a nd used to run the analyses. The macro provides results of the multistep approach proposed by Barron and Kenny (1986), known as the causal steps test. However, fulfillment of the multistep approach was not a requisite condition for media tion in the current study. Me thodologists have identified

PAGE 81

46 several shortcomings in the causal steps te st. Primarily, methodologists question whether it is necessary to demonstrate a direct eff ect from the independent variable to the dependent variable (MacKinnon, Krull & Lo ckwood, 2000; Shrout & Bolger, 2002). As the mediation relationship becomes more distal or complex, it will become more difficult to establish a relationship between a predicto r and outcome. This relationship is more likely to be transmitted through additional mediat ors, or affected by competing causes or random factors. Other research ers have critiqued the full cau sal-steps test on the grounds that the sample size needed to detect effect s is prohibitively large (Fritz & MacKinnon, 2007). To evaluate the presence of mediation, a variation of the causal steps test, known as the joint significance test, was used (see Mackinnon, Lockw ood, Hoffman, West, & Sheets, 2002 for a review). This test was s hown to have greater power with comparable Type I error rates to Baron and Kenny’s ( 1986) causal steps test (MacKinnon et al., 2002). In the joint signif icance test, if the relationships between the independent variable and the mediator, and the mediator and the depe ndent variable are join tly significant, then mediation is present. Although this met hod does not test the overall independentdependent variable relationship, it does provide the most dir ect test of the simultaneous null hypothesis that both paths in the media tion model are equal to zero. Men were compared to women and whites were compar ed to non-whites by examining the direction and significance of the path coefficients in their respective models. Although the tests of simple mediation give a second indication of the presence of mediation (i.e. after the indirect effects of the moderated mediati on), their main purpose was to examine the direction and significance of indi vidual paths in the model. It should also be noted that

PAGE 82

47 the joint significance approach was used to determine the presence of overall mediation in the moderated mediation models. Thus, the analyses provided four pieces of information considered in the results. First, the results of the multiple regression portion of the moderated mediation reveal if the overall mediation model is significant (regar dless of minority status). This reveals whether or not there is a relationship of the various organizational antecedents on the outcomes through perceived discrimination. The jo int significance test was used to make this determination. Second, th e interaction between the or ganizational antecedent and minority status when predicting perceived di scrimination reveals whether or not minority status is a moderator. Third, th e indirect effects reveal if th e mediation model is different for minority and majority group members. Note that it is possible to have a significant interaction term and non-significant indire ct effects and vice versa, or have both significant (or non-significant) in teraction term and indirect e ffects. Finally, the tests of simple mediation between groups allows fo r the examination of path strength and direction in the mediation model between gr oups. The results of the separate tests of mediation should also align so mewhat with the indirect e ffects, since both tests are examining the same outcome using different methods.

PAGE 83

48 Chapter Three Results Analysis of Pilot Test Data The purpose of the pilot test was to evalua te all items intended for inclusion in the full demonstration. Item characteristics, in cluding item mean, item standard deviation, item-total correlations, alpha if item deleted, a nd total alpha were evaluated for all Likerttype items. The Token Status scale was open-ended, thus the response patterns were evaluated in this measure. Tables 4 th rough 26 display item mean, item standard deviation, item-total correlation, and alpha if item deleted for each item in every measure included in the pilot study. Results of the item analysis for most measures indicated that no changes were needed for the full demonstration. This in cludes scales measuring job satisfaction, organizational commitment, all SF-36 subscal es, instrumental and emotional social support, minority segmentation, diversity climate, subtle workplace gender discrimination, subtle workplace racial discrimination, overt workplace gender discrimination, and overt workpl ace racial discrimination. The majority of these scales exhibited total coefficient alphas over 0.80. The exceptions to this were the SF-36 Emotional well-being scale ( = 0.77), SF-36 Social Functioning Scale ( = 0.79), SF-36 Pain Scale ( = 0.78), and the SF-36 General Health Scale ( = 0.74). Although the alphas for these scales are lower than the ot hers, given the well-valid ated nature of the

PAGE 84

49 SF-36 as a whole, changes to this scale are not warranted. Moreover, the majority of the scales exhibited item means close to the na tural means of the Likert-type scales. The exceptions to this were scales measuring subtle workplace gender discrimination, subtle workplace race discrimination, overt wor kplace gender discrimination, and overt workplace race discrimination. For these meas ures, item means trended towards the lower end of the scale. This is not surp rising, given the sensitive nature of these questions. Also, it is unlikely that the major ity of the sample would have experienced an abundance of discrimination. Since these kinds of response patterns are expected, no changes were needed to these scales. Finally, the item-total correlations within each scale as well as alpha-if-item-deleted did not reveal cause for any modifications to these items. Table 4. Item Analysis of Job Satisfaction Measure. M SD I-T r Deleted All in all, I am satisfied with my job. 3.84 1.07 .94 .79 In general, I don’t like my job. 3.91 1.06 .88 .90 In general, I like working here. 3.92 0.98 .91 .84 Note. Coefficient alpha for th e scale as a whole was 0.89

PAGE 85

50 Tables 5. Item Analysis of Organizational Commitment Measure M SD I-T r Deleted I am willing to put a great deal of effort beyond that normally expected in order to help this organization be successful. 3.90 1.04 .76 .93 I talk up this organization to my friends as a great organization to work for. 3.58 1.09 .83 .92 I would accept almost any type of job assignment in order to keep working for this organization. 3.04 1.23 .71 .93 I find that my values and the organization’s values are very similar. 3.51 1.14 .80 .93 I am proud to tell others that I am part of this organization. 3.78 1.07 .86 .92 This organization really inspires the very best in me in the way of job performance. 3.50 1.18 .89 .92 I am extremely glad that I chose this organization to work for over others I was considering at the time I joined. 3.71 1.08 .84 .92 I really care about the fate of this organization. 3.77 1.11 .75 .93 For me, this is the best of all possible organizations for which to work. 3.23 1.24 .83 .92 Note. Coefficient alpha for th e scale as a whole was 0.93

PAGE 86

51 Table 6. Item Analysis of SF-36 Physical Functioning Sub-scale M SD I-T r Deleted My health now limits vigorous activities, such as… 2.65 0.57 0.65 0.95 My health now limits moderate activities, such as.. 2.89 0.37 0.88 0.85 My health now limits lifting or carrying groceries. 2.88 0.39 0.81 0.76 My health now limits climbing several flights of stairs. 2.81 0.46 0.82 0.76 My health now limits climbing one flight of stairs. 2.92 0.32 0.88 0.85 My health now limits bending, kneeling, or stooping. 2.83 0.43 0.77 0.70 My health now limits walking more than one mile. 2.84 0.45 0.82 0.76 My health now limits walking several blocks. 2.87 0.39 0.91 0.88 My health now limits walking on block. 2.92 0.34 0.85 0.82 My health now limits bathing or dressing myself. 2.93 0.31 0.77 0.72 Note. Coefficient alpha for th e scale as a whole was 0.94

PAGE 87

52 Table 7. Item Analysis of SF-36 Phys ical Role Limitations Scale M SD I-T r Deleted During the past four weeks, I have cut down on the amount of time I spend on work or other activities as a result of m y p h y sical health. 1.94 0.23 0.83 0.81 During the past four weeks, I have accomplished less than I would like as a result of m y p h y sical health. 1.89 0.31 0.87 0.80 During the past four weeks, I was limited in the kind of work or other activities I could do as a result of my physical health. 1.92 0.26 0.87 0.79 During the past four weeks, I had difficulty performing work or other activities because of my physical health. 1.91 0.28 0.78 0.85 Note. Coefficient alpha for th e scale as a whole was 0.85 Table 8. Item Analysis of SF-36 Emo tional Role Limitations Scale M SD I-T r Deleted During the past four weeks, I have cut down the amount of time I spend on work or other activities as a result of emotional problems. 1.41 0.71 0.90 0.88 During the past four weeks, I accomplished less than I would like as a result of emotional problems. 1.50 0.81 0.93 0.85 During the past four weeks, I didn’t do work or other activities as carefully as usua l as a result of emotional problems. 1.41 0.73 0.92 0.86 Note. Coefficient alpha for the scale as a whole was 0.90

PAGE 88

53 Table 9. Item Analysis of the SF-36 Energy Sub-scale M SD I-T r Deleted How much of the time during the past four weeks did you feel full of p e p ? 3.75 1.30 0.74 0.72 How much of the time during the past four weeks did you have a lot of ener gy ? 4.34 1.61 0.79 0.72 How much of the time during the past four weeks did you feel worn out? 4.13 1.22 0.73 0.71 How much of the time during the past four weeks did you feel tired? 3.62 1.26 0.80 0.67 Note. Coefficient alpha for the scale as a whole was 0.76. Table 10. Item Analysis of the SF-36 Emotional Well-being Sub-scale M SD I-T r Deleted How much of the time during the past four weeks have you been a ver y nervous p erson? 4.71 1.29 0.70 0.74 How much of the time during the past four weeks have you felt so down in the dumps that nothing could cheer y ou u p ? 5.19 1.22 0.78 0.69 How much of time during the past four weeks have you felt calm and p eaceful? 3.87 1.26 0.63 0.77 How much of the time during the past four weeks have you felt downhearted and blue? 4.94 1.09 0.77 0.70 How much of the time during the past four weeks have you been a happy person? 4.35 1.15 0.74 0.71 Note. Coefficient alpha for the scale as a whole was 0.77

PAGE 89

54 Table 11. Item Analysis of the SF-3 6 Social Functioning Sub-scale M SD I-T r Deleted During the past four weeks, to what extent has your physical health or emotional problems interfered with your normal social activities with family, friends, neighbors, or groups? 4.45 0.87 0.90 n/a During the past four weeks, how much of the time has your physical health or emotional problems interfered with your social activities? 4.33 0.96 0.92 n/a Note. Coefficient alpha for the scale as a whole was 0.79 Table 12. Item Analysis of the SF-36 Pain Sub-scale M SD I-T r Deleted How much bodily pain have you had during the past four weeks? 4.93 1.12 0.94 n/a During the past four weeks, how much did pain interfere with your normal work (including both work outside the home and housework). 4.56 0.81 0.88 n/a Note. Coefficient alpha for the scale as a whole was 0.78. Table 13. Item Analysis of the SF-3 6 General Health Sub-scale M SD I-T r Deleted In general, would you say your health is: 3.97 0.85 0.75 0.66 I seem to get sick a little easier than other people. 4.12 1.11 0.63 0.75 I am healthy as anybody I know. 4.00 0.95 0.73 0.67 I expect my health to get worse. 4.02 1.12 0.65 0.73 My health is excellent. 3.92 0.98 0.77 0.65 Note. Coefficient alpha for the scale as a whole was 0.74

PAGE 90

55 Table 14. Item Analysis of the Inst rumental Social Support Scale M SD I-T r Deleted My immediate supervisor goes out of his/her way to do things to make my work life easier for me. 3.45 1.23 .70 .85 Other people at work go out of their way to make my work life easier for me. 3.27 1.14 .67 .85 My immediate supervisor can be relied on when things get tough at work. 3.93 1.09 .75 .84 Other people at work can be relied on when things get tough at work. 3.82 1.03 .71 .85 I have access to a mentor at my workplace. 3.64 1.54 .77 .84 I have access to informal social networks at my workplace. 3.84 1.41 .78 .84 I have access to informal information networks at my workplace. 3.79 1.42 .82 .83 Note. Coefficient alpha for the scale as a whole was 0.86 Table 15. Item Analysis of the Em otional Social Support Scale M SD I-T r Deleted It is easy to talk to my supervisor. 4.10 1.01 .90 .61 It is easy to talk to other people at work. 4.20 0.94 .79 .79 My immediate supervisor is willing to listen to my personal problems. 3.85 1.13 .85 .77 Note. Coefficient alpha for the scale as a whole was 0.80

PAGE 91

56 Table 16. Item Analysis of the Minority Segmentation Scale M SD I-T r Deleted In my organization, minorities tend to get certain types of jobs. 2.55 1.13 0.90 0.87 In my organization, minority and majority group members tend to get different types of jobs. 2.61 1.14 0.84 0.73 My job is one that tends to be given to minorities rather than majority group members. 2.21 0.98 0.77 0.65 In my organization, minorities tend to be assigned to certain areas or departments whereas majority group members tend to be assigned elsewhere. 2.42 1.07 0.91 0.86 In my organization, minority and majority group members tend to work at different physical locations (i.e. branch offices, buildings, neighborhoods). 2.28 1.05 0.86 0.78 Note. Coefficient alpha for the scale as a whole was 0.91

PAGE 92

57 Table 17. Item Analysis of the Diversity Climate Scale M SD I-T r Deleted My organization recruits from diverse sources. 3.64 0.95 0.73 0.93 My organization offers equal access to training. 3.83 0.90 0.77 0.93 My organization promotes open communication on diversity. 3.68 0.91 0.85 0.92 My organization publicizes diversity principles. 3.45 1.03 0.81 0.92 My organization offers training to manage a diverse population. 3.27 1.08 0.77 0.93 My organization respects the perspectives of people like me. 3.65 0.97 0.83 0.92 My organization maintains a diversity-friendly work environment. 3.79 0.90 0.83 0.92 My workgroup has a climate that values diverse perspectives. 3.62 0.92 0.82 0.92 Top leaders visibly commit to diversity. 3.490.950.84 0.92 Note. Coefficient alpha for the scale as a whole was 0.92

PAGE 93

58 Table 18. Item Analysis of the Subtl e Gender Discrimination Scale M SD I-T r Deleted You are treated with less courtesy than other people because of your gender. 1.34 0.65 0.86 0.95 You are treated with less respect that other people because of your gender. 1.34 0.66 0.87 0.95 People act as if they think you are not smart because of your gender. 1.35 0.66 0.90 0.95 People act as if th ey are afraid of you because of your gender. 1.27 0.62 0.86 0.95 People act as if they think you are dishonest because of your gender. 1.24 0.56 0.86 0.95 People act as if th ey are better than you are because of your gender. 1.39 0.69 0.86 0.95 You or your family members are called names or insulted because of your gender. 1.17 0.49 0.84 0.95 You are threatened or harassed because of your gender. 1.22 0.56 0.86 0.95 People ignore you or act as if you are not there because of your gender. 1.23 0.55 0.85 0.95 Note. Coefficient alpha for the scale as a whole was 0.96

PAGE 94

59 Table 19. Item Analysis of the Subtle Ra ce Discrimination at Work Scale M SD I-T r Deleted You are treated with less courtesy than other people because of your 1.25 0.59 0.91 0.96 You are treated with less respect that other people because of your 1.27 0.58 0.90 0.97 People act as if they think you are not smart because of race/ethnicity. 1.27 0.60 0.95 0.96 People act as if th ey are afraid of you because of your race/ethnicity. 1.23 0.59 0.91 0.96 People act as if they think you are dishonest because of your 1.23 0.59 0.90 0.96 People act as if th ey are better than you are because of your 1.29 0.67 0.89 0.97 You or your family members are called names or insulted because of 1.23 0.60 0.86 0.97 You are threatened or harassed because of your race/ethnicity. 1.18 0.53 0.85 0.97 People ignore you or act as if you are not there because of your 1.22 0.58 0.92 0.96 Note. Coefficient alpha for the scale as a whole was 0.97

PAGE 95

60 Table 20. Overt Race Discrimina tion at Work Scale M SD I-T r Deleted I have been treated unfairly by employers, bosses, or supervisors b ecause of m y race/ethnicit y 1.49 0.88 0.84 0.97 I have been treated unfairly by coworkers or colleagues because of my race/ethnicity. 1.43 0.81 0.84 0.97 My supervisor sometimes makes racist decisions against my racial/ethnic group. 1.38 0.74 0.93 0.96 My coworkers sometimes make racist statements directed at my racial/ethnic group. 1.42 0.79 0.87 0.97 I feel that some of the policies and practices of this organization are racist, and they are directed against my racial/ethnic group. 1.42 0.80 0.94 0.97 At work, I sometimes feel that my race/ethnicity is a limitation. 1.45 0.83 0.87 0.97 At work, I do not get enough recognition because of my race/ethnicit y 1.41 0.77 0.93 0.96 At work, I sometimes feel that people actively try to stop me from advancing because of my race/ethnicity. 1.38 0.73 0.94 0.96 At work, I feel that others exclude me from their activ ities because of my race/ethnicity. 1.39 0.75 0.93 0.96 Note. Coefficient alpha for the scale as a whole was 0.97

PAGE 96

61 Table 21. Overt Gender Discrimina tion at Work Scale M SD I-T r Deleted I have been treated unfairly by employers, bosses, or supervisors because of my gender. 1.50 0.91 0.89 0.97 I have been treated unfairly by coworkers or colleagues because of my gender. 1.47 0.86 0.90 0.97 My supervisor sometimes makes sexist decisions against my gender group. 1.50 0.87 0.91 0.97 My coworkers sometimes make sexist statements directed at my gender group. 1.58 0.94 0.89 0.97 I feel that some of the policies and practices of this organization are sexist against my gender group. 1.45 0.83 0.88 0.97 At work, I sometimes feel that my gender is a limitation. 1.52 0.87 0.87 0.97 At work, I do not get enough recognition because of my gender. 1.45 0.80 0.92 0.97 At work, I sometimes feel that people actively try to stop me from advancing because of my gender. 1.45 0.82 0.91 0.97 At work, I feel that others exclude me from their activ ities because of my gender. 1.49 0.87 0.91 0.97 Note. Coefficient alpha for the scale as a whole was 0.97

PAGE 97

62 Table 22. General Every-day Race Discrimination Scale M SD I-T r Deleted I have been treated unfairly by teachers and professors because of m y race/ethnicit y 1.44 0.78 0.82 0.96 I have been treated unfairly by people in service jobs (store clerks, waiters, bartenders, and others) because of my race or ethnicity. 1.54 0.89 0.86 0.96 I have been treated unfairly by strangers because of my race/ethnicity. 1.67 1.02 0.86 0.96 I have been treated unfairly by people in helping jobs (doctors, nurses, school counselors, therapists, 1.43 0.78 0.87 0.96 I have been treated unfairly by neighbors because of my race/ethnicity. 1.47 0.81 0.81 0.96 I have been treated unfairly by institutions (schools, universities, the police, the courts, and others) 1.54 0.91 0.82 0.96 I have been treated unfairly by people that I thought were my 1.45 0.84 0.78 0.96 I have been accused or suspected of doing something wrong (such as stealing, cheating, not doing your 1.45 0.84 0.81 0.96 People have misunderstood my intentions and motives because of 1.64 1.06 0.84 0.96 There were times when I wanted to tell off someone for being racist 1.81 1.22 0.84 0.96 There have been times when I have been really angry about racist 1.70 1.10 0.85 0.96 There have been times when I was forced to take drastic steps (such as filing a grievance, filing a lawsuit, 1.38 0.76 0.77 0.96 I have been called racist names. 1.67 1.08 0.81 0.96 I have gotten into an argument or a fight about somethi ng racist that was 1.66 1.08 0.71 0.96 I have been made fun of, picked on, pushed, shoved, hit or threatened 1.54 0.96 0.79 0.96 Note. Coefficient alpha for the scale as a whole was 0.96

PAGE 98

63 Table 23. General Every-day Gender Discrimination M SD I-T r Deleted I have been treated unfairly by teachers and professors because of my gender. 1.45 0.81 0.79 0.95 I have been treated unfairly by people in service jobs (store clerks, waiters, bartenders, and others) because of my gender. 1.54 0.88 0.82 0.95 I have been treated unfairly by strangers because of my gender. 1.62 0.95 0.81 0.95 I have been treated unfairly by people in helping jobs (doctors, nurses, school counselors, therapists, dh)bfd 1.46 0.79 0.83 0.95 I have been treated unfairly by neighbors because of my gender. 1.43 0.77 0.82 0.95 I have been treated unfairly by institutions (schools, universities, the police, the courts, and others) 1.47 0.83 0.80 0.95 I have been treated unfairly by people that I thought were my friends because of my gender. 1.46 0.86 0.80 0.95 I have been accused or suspected of doing something wrong (such as stealing, cheating, not doing your hfhkbki 1.39 0.75 0.77 0.95 People have misunderstood my intentions and motives because of d 1.81 1.12 0.67 0.95 There were times when I wanted to tell someone off for being sexist 1.85 1.16 0.72 0.95 There have been times when I have been really angry about something 1.85 1.16 0.82 0.95 There have been times when I was forced to take drastic steps (such as filing a grievance, filing a lawsuit, quitting my job, moving away, and 1.46 0.85 0.71 0.95 I have been called sexist names. 1.88 1.21 0.81 0.95 I have gotten into an argument or a fight about something sexist that was 1.73 1.13 0.83 0.95 I have been made fun of, picked on, pushed, shoved, hit or threatened 1.49 0.89 0.83 0.95 Note. Coefficient alpha for the scale as a whole was 0.95

PAGE 99

64 The results of the pilot test did reveal modifications necessary for two scales: perceived organizational support for Equal Em ployment Opportunity, and Token Status. In the perceived organizational support for EEO scale, one item asked “When organizational decisions are made (e.g. hiring, promotions), they are typically identityblind.” This item had a lo w item-total correlation ( r = 0.79) and alpha ( = 0.86) would have been improved if the item was deleted ( = 0.88). The wording of the item may have been confusing to respondents, as many we re likely unsure what was meant by “identityblind”. This item was borrowed from a previ ously used measure, where the meaning of item in context may have been more apparent. This item was removed from the survey used in the full demonstration. Table 24. Item Analysis of the Equal Employment Opportunity Scale M SD I-T r Deleted My organization has a strong Equal Employment Opportunity (EEO) Policy. 3.86 0.91 0.81 0.82 My organization has a visible EEO Policy. 3.66 1.02 0.88 0.79 Senior managers emphasize EEO. 3.48 1.06 0.87 0.80 When organizational decisions are made (e.g. hiring, promotions), they are typically identity-blind. 3.32 0.99 0.64 0.88 My organization displays their commitment to EEO in statements on company publications (e.g. company website, posters, etc.) 3.38 0.99 0.79 0.83 Token status was measured in the pi lot study with six open-ended items. Two items asked respondents to estimate the to tal number of people em ployed within their workgroup and within the physical location of their job. Four more questions asked

PAGE 100

65 individuals to estimate the num ber of people sharing their ge nder and ethnicity within their workgroup and entire work location. The or iginal idea was to calculate a percentage based on the numbers provided. The open-ende d nature of these items was the most problematic. Twenty-eight percent of the sample either did not respond to one or more of the open-ended items, or provided answers that were difficult to interpret. For example, one individual indicated that the number of people employed in the workgroup was larger than the number of people employed at the workplace. Many of those that did respond provided a range of people. At best these rang es could be averaged or interpreted, leading to a loss of measurement precision. The six open-ended items in the pilot were abandoned in favor of four Li kert-type items for the full demonstration. Instead of asking participants to fill-i n numbers, there were merely asked to choose a percentage of individuals who shared their ge nder or ethnicity from a sixpoint Likert scale. One of these options was “I don’t know/ not applicable.” Many particip ants in the pilot indicated not being able to estimate numbers of people, or indicated they were self-employed or worked remotely, making tokenism impossible. Table 25. Percentage of Tokenism within the Workplace and Work Group Token Status Percentage Gender token in workplace 9.7% Racial token in workplace 10.3% Gender token in work group 1.1% Racial token in work group 2.9% Of particular concern was the measurem ent of workplace discrimination. No scale could be located to meet the needs of the current study. Thus, two sc ales were developed

PAGE 101

66 for use in this study. The first was perceive d subtle workplace gender/race discrimination and the other was perceived overt workplac e gender/race discrimination. The “subtle” scale was a small modification of an exis ting scale, while the “overt” scale was a compilation and modification of various items fr om three previous studies. Not only were item characteristics examined for these scales but these scales were compared to a measure of general every-da y gender/race discrimination. The comparison was made to establish discriminant validity. Since all si x scales were measuring discrimination, a moderate correlation among measures is to be expected. However, ideally the correlations between the workplace measures should be higher than between the workplace and non-workplace measure, and th e correlations among the race-based or gender-based discrimination measures should be higher than between the race-based and gender-based discrimination measures. None of the correlations exceed 0.80. Th e subtle discrimination scales showed the best discriminant validity. Both sub tle workplace discrimination scales correlated highest with their overt workplace discrimina tion scale counterpart and lowest with the other-type every-day discrimination scal e. Overt workplace gender discrimination correlated highest with overt workplace race discrimination and lowest with subtle workplace race discrimination. Overt workplac e race discrimination correlated highest with overt workplace gender discrimination and lowest with subtle workplace gender discrimination. Moreover, each overt scale was more correlated with its same-type everyday discriminations scale than with the other-type every-day discrimination. The correlations between overt wo rkplace discrimination and ev ery-day discrimination are likely higher than subtle discrimination becau se they are both measuring more serious

PAGE 102

67 forms of mistreatment whereas the subtle di scrimination scale measures more ambiguous, less serious mistreatment. Although the ove rt workplace discrimination scales are correlated with each other, they appear to be distinguishing themse lves from other-type discrimination both in the subtle wor kplace form and in every-day life. The correlation matrix is shown in Ta ble 26. Again, the correlations are higher than desired in a discriminant validity anal ysis, but this is beca use all six scales are measuring some form of discrimination. Indi viduals who perceive discrimination in one context may be more likely to perceive di scrimination in another. The pattern of correlations is generally supportive of the c onstruct validity of the subtle and overt workplace discrimination scales. Thus, they were unchanged for use in the full demonstration. Table 26. Correlations among the di scrimination measures 1 2 3 4 5 1 Subtle gender discrimination at work 2 Subtle race discrimination at work 0.68* 3 Overt gender discrimination at work 0.69* 0.56* 4 Overt race discrimination at work 0.56* 0.71* 0.80* 5 World every-day gender discrimination 0.64* 0.54* 0.79* 0.71* 6 World every-day race discrimination 0.45* 0.63* 0.62* 0.78* 0.73* *Correlation is sign ificant at p < .0001

PAGE 103

68 Analysis of the full demonstration data Initially, analyses were conducted and hypot heses tested exactly as specified in the method of the current paper. The orig inal design specified six predictors, and intended to conduct analyses separately for ove rt and subtle discrimination, mainly as an exploratory task. These results are presented in their entirety in A ppendix B. Examination of the original findings led to a condense d version which is presented and discussed below. The condensed version reflects a numbe r of changes. First, the predictors “EEO” and “diversity climate” were combined in to a single index. Table 27 displays the correlations among all measures used in the study, as well as the means and standard deviations for each scale. The table contai ns the correlation between EEO and diversity climate, which was high ( r = .70). Additionally, the pattern of results between EEO and diversity climate was very similar. These findi ngs likely result from a conceptual overlap between perceptions of equal employment opport unity and diversity climate, in that EEO is a part of, or will contribut e to, diversity climate. For th at reason, these two predictors have been combined into a single index (hencef orth referred to as EEO/diversity climate, or EEODIV). Second, instrumental social support and emotiona l social support have also been combined into an index. The correlation between the two was high ( r = .74), and the pattern of results in the original analysis was nearly identical. As there is not a large contribution to the knowledge base by consid ering these two variab les separately, they have been combined (henceforth referred to as social support). Next, overt and subtle discrimination (both race-based and gender-based) have been combined into a single index of perceived discrimination. These tw o types of perceived discrimination were originally treated separately as an exploratory analysis. Howeve r, as the original analyses

PAGE 104

69 revealed, there was little difference between s ubtle and overt discrimi nation within either race-based or gender-based discrimination. Mu ch more information could be gleaned by comparing race-based to gender-based discrimination. Thus, the analyses to follow contain a single index of e ither race-based or gender-bas ed workplace discrimination (henceforth referred to as “race discrimina tion” or “gender discrimination”). Finally, token status is not included in the analyses to follow. There are two reasons for this decision. Primarily, token status is highly correlated with minority status in that most of the tokens in the sample are minorities. More over, there are a dearth of tokens overall. The pattern of results presente d in Appendix B is merely a reflection of this imbalance and likely not a reflection of reality. S econd, there have been problems with the measurement of token status from the outse t of the study. The items were dramatically changed from the pilot study to the full dem onstration, and the items as they currently stand have no validity information. Further, the items ask participants to make a subjective assessment of a very objective cons truct. Although the questions posed in the current study would make a va luable contribution to the literature, any interpretations made from the quality of th e data collected would not.

PAGE 105

70 Table 27. Correlations Among all Measures used in Original Analyses Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 EEO 14.43 3.88 .92 2 Minority Seg 12.03 4.47 -.14 .88 3 Diversity Climate 32.21 7.32 .70 -.31 .94 4 Instrumental Support 13.29 3.76 .32 -.11 .44 .83 5 Emotional Support 14.91 3.55 .35 -.15 .43 .74 .81 6 Subtle Race Discrim 11.65 5.24 -.20 .22 -.33 -.21 -.20 .97 7 Subtle Sex Discrim 12.37 4.93 -.26 .29 -.35 -.24 -.30 .70 .94 8 Overt Race Discrim 13.06 7.21 -.23 .25 -.39 -.26 -.24 .82 .60 .97 9 Overt Sex Discrim 13.82 7.42 -.27 .28 -.35 -.26 -. 30 .57 .76 .66 .96 10 Job Sat 10.92 2.95 .21 -.16 .33 .54 48 -.22 -.22 -.24 -.25 .89 11 Org Comm 30.31 7.98 .32 -.19 .46 .60 48 -.18 -.20 -.21 -.24 .77 .93 12 Physical Health 65.59 6.08 .18 -.12 .26 .20 .23 -.28 -.28 -.35 -.31 .24 .18 .83 13 Psych Health 50.12 7.10 .22 -.17 .27 .28 28 -.13 -.21 -.20 -.25 .33 .26 .49 0.85 14 Intent to Turnover 2.72 1.40 -.21 .17 -.31 -.48 -.38 .28 .25 .28 .28 -.71 -.65 -.19 -.29 --15 Gender Token .05 .22 .06 .03 -.02 -.08 -.11 -.05 09 -.05 .14 .01 -.06 .01 -.04 .03 --16 Racial Token .16 .36 .02 .00 .02 .03 .01 .12 -.08 .11 -.01 -. 01 -.03 .02 -.02 .00 .07 --

PAGE 106

71 In summary, the results of the full demonstr ation below present moderated mediation and simple mediation findings related to three predictors (EEO/diversity climate, minority segmentation, and social support), two mediat ors (perceived race-bas ed or gender-based discrimination), and five outcomes (job satisfaction, organizational commitment, intention to turnover, physical health, and psychological h ealth). Table 28 presents descriptive statistics and co rrelations among measures used in the revised analyses.

PAGE 107

72 Table 28. Correlations Among all Measures used in Revised Analyses Mean SD 1 2 3 4 5 6 7 8 9 10 1 EEO/Diversity Climate 46.74 10.42 .94 2 Minority Segmentation 12.03 4.47 -.27 .88 3 Social Support 28.17 6.88 .46 -.14 .89 4 Race Discrimination 24.65 11.89 -.35 .24 -.27 .97 5 Sex Discrimination 26.12 11.57 -.35 .30 -.32 .71 .96 6 Job Satisfaction 10.92 2.95 .30 -.16 .55 -.25 -.24 .89 7 Org Commitment 30.31 7.98 .44 -.19 .58 -.22 -.23 .77 .93 8 Physical Health 65.59 6.08 .25 -. 12 .23 -.35 -.32 .24 .18 .83 9 Psychological Health 50.12 7.09 .27 -.17 .29 -. 17 -.25 .33 .26 .49 .85 10 Intent to Turnover 2.71 1.40 -.30 .17 -.46 .30 .28 -.71 -.65 -.19 -.29 --

PAGE 108

73 Results for Perceived Equal Employme nt Opportunity/Diversity Climate Sex-based Discrimination Perceived Equal Employment Opportunity (EEO) and diversity climate in the workplace was investig ated as a predictor in a model where sexbased discrimination mediates the relati onship between EEODIV and five dependent variables (job satisfaction, or ganizational commitment, intention to turnover, physical health, and psychological health). Minority status was investig ated as a moderator in the relationship between the predictor and the mediator. Additionally, separate mediation models were tested for each of the co mparison groups (men, women, whites, and nonwhites). First, a model was tested where sex discrimination mediates the relationship between EEODIV and job satisfaction. The results of the moderated mediation demonstrated overall support for mediation as the path from EEODIV to subtle sex discrimination was significant ( B = -.46, t = -4.39, p < .00) in addition to the path from the mediator to job satisfaction ( B = -.05, t = -3.50, p < .00). Perceived EEODIV was related to less perceived di scrimination, which in turn was related to greater job satisfaction. No significant in teraction was found between the predictor (EEODIV) and the moderator (minority status). Thus, minority status does not moderate the relationship between EEODIV and perceived subtle sex di scrimination. The indi rect effects support the finding of overall mediation for both men an d women, as the indirect effects for both groups were significant. Results for the mode rated mediation can be found in Table 29.

PAGE 109

74 Table 29. Results of Moderated Mediation ( PREDICTOR = EEODIV, Med =Sex-based Discrimination, DV = Job Satisfaction, Mod = Gender Predictor B SE t p Sex discrimination Constant 46.395.11 9.08.00 EEODIV -.46.10 -4.39.00 Minority Status -2.945.94 -.49.62 EEODIV Minority Status .10.12 .79.43 Job satisfaction Constant 8.481.45 5.82.00 Sex discrimination -.05.01 -3.50.00 EEODIV .07.03 2.62.01 Minority Status .721.53 .47.64 EEODIV Minority Status -.01.03 -.21.84 Minority Status Indirect Effect SE z p Men .02.01 2.69.01 Women .02.01 2.95.00 The separate tests of mediation for bot h men and women support the finding that perceived subtle sex discrimination mediat ed the relationship between EEO and job satisfaction, and that there doe s not appear to be a differe nce for minority or majority group members. Also, the paths were in the hypothesized directions The results for the simple mediation tests for men and women are displayed in Table 30.

PAGE 110

75 Table 30. Results of Simple Mediation for Men and Women (PREDICTOR = EEODIV, Med = Sexbased Discrimination, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .09.033.27.00.08.02 5.07 .00 b(MX) -.46.10-4.69.00-.36.07 -5.54 .00 b(YM.X) -.08.03-3.12.00-.03.02 -2.18 .03 b(YX.M) .06.031.87.06.07.02 4.13 .00 Next, a model was tested where the de pendent variable was organizational commitment. Again, support for overall media tion was found as the path from EEODIV to sex discrimination ( B = -.44, t = -4.26, p = .00) and the path from the mediator to organization commitment ( B = -.08, t = -2.42, p < .05) was significant. Perceived EEODIV was related to less pe rceived discrimination, which in turn was related to greater organizational comm itment. However, no significant interaction was found between EEODIV and minority status. When ex amining the indirect effects, both men and women exhibited a significan t indirect effect of EEO on organizational commitment through perceived discrimination. Results of the moderated mediation are displayed in Table 31.

PAGE 111

76 Table 31. Results of Moderated Mediation ( PREDICTOR = EEODIV, Med = Sex-based Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Sex discrimination Constant 45.065.05 8.91.00 EEODIV -.44.10 -4.26.00 Minority Status -1.585.87 -.27.79 EEODIV Minority Status .08.12 .63.53 Organizational Constant 20.153.78 5.34.00 Sex discrimination -.08.04 -2.42.02 EEODIV .24.07 3.28.00 Minority Status -3.103.98 -.78.44 EEODIV Minority Status .11.08 1.32.19 Minority Status Indirect Effect SE z p Men .04.02 2.06.04 Women .03.01 2.20.03 The results of the separate mediation te sts demonstrate that the model is supported for women, but only one path was significant for men. However, all paths were in the hypothesized directions. Results for the simple mediation tests are shown in Table 32. Table 32. Results of Simple Mediation for Men and Women (PREDICTOR = EEODIV, Med = Sexbased Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) .27.073.93.00.38.04 8.71 .00 b(MX) -.44.09-4.73.00-.36.07 -5.57 .00 b(YM.X) -.07.07-.95.34-.09.04 -2.24 .03 b(YX.M) .24.083.21.00.34.05 7.56 .00

PAGE 112

77 Next, a model predicting intention to turnover was examined. Support for overall mediation is again found as the paths from EEODIV to perceived discrimination ( B = .46, t = -4.41, p < .01) and from perceived discrimination to intention to turnover ( B = .03, t = 4.41, p < .01) were significant. Perceived EEODIV was related to less perceived discrimination, which in turn was related to less intention to turnover. However, the interaction between minority status and the independent variables was not significant, indicating that relationships may not be different for men and women. The indirect effects were significant for bot h men and women participants. Results of the moderated mediation are shown in Table 33. Table 33. Results of Moderated Mediation ( PREDICTOR = EEODIV, Med = Sex-based Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Sex discrimination Constant 46.395.09 9.12.00 EEODIV -.46.10 -4.41.00 Minority Status -3.015.90 -.51.61 EEODIV Minority Status .10.12 .80.42 Turnover Constant 3.35.69 4.86.00 Sex discrimination .03.01 4.41.00 EEODIV -.03.01 -2.21.03 Minority Status -.16.72 -.22.82 EEODIV Minority Status .00.01 .15.88 Minority Status Indirect Effect SE z p Men -.01.00 -3.08.00 Women -.01.00 -3.47.00

PAGE 113

78 The separate mediation analyses demonstr ated that sex disc rimination mediates the relationship between EEODI V and intention to turnover for both men and women. All paths are in the hypothesized di rections. Table 34 displays the results for the simple mediation tests. Table 34. Results of Simple Mediation for Men and Women (PREDICTOR = EEODIV, Med = Sexbased Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) -.04.01-3.33.00-.04.01 -4.56 .00 b(MX) -.46.10-4.69.00-.36.06 -5.62 .00 b(YM.X) .04.013.08.00.03.01 3.30 .00 b(YX.M) -.03.01-1.95.05-.03.01 -3.31 .00 Next, the mediating role of perceived di scrimination in the relationship between EEODIV and physical health was examined. Th e model was supported as the paths from EEODIV to the mediator ( B = -.44, t = -4.12, p < .01) and from per ceived discrimination physical health ( B = -.13, t = -4.68, p < .01) were significant. Perceived EEODIV was related to less perceived discrimination, whic h in turn was related to better physical health. However, there was no significant inte raction between minority status and EEO. Additionally, indirect effects for both men and women were significant and similar in size. The results of the moderated me diation are displayed in Table 35.

PAGE 114

79 Table 35. Results of Moderated Mediation ( PREDICTOR = EEODIV, Med = Sex-based Discrimination, DV = Physical Health, Mod = Gender) Predictor B SE t p Sex discrimination Constant 45.675.29 8.62.00 EEODIV -.44.11 -4.12.00 Minority Status -3.146.13 -.51.61 EEODIV Minority Status .09.13 .75.46 Physical Health Constant 67.783.07 22.08.00 Sex discrimination -.13.03 -4.68.00 EEODIV .05.06 .81.42 Minority Status -3.453.22 -1.07.28 EEODIV Minority Status .05.07 .79.43 Minority Status Indirect Effect SE z p Men .06.02 3.06.00 Women .05.01 3.50.00 The results for the separate tests of mediation are shown in Table 36. Here, perceived discrimination does appear to me diate the relationship between EEODIV and physical health for both men and women. Additionally, all pa ths are in the hypothesized directions.

PAGE 115

80 Table 36. Results of Simple Mediation for Men and Women (PREDICTOR = EEODIV, Med = Sexbased Discrimination, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .11.061.90.06.15.04 4.07 .00 b(MX) -.44.10-4.45.00-.35.07 -5.22 .00 b(YM.X) -.16.05-3.12.00-.12.03 -3.60 .00 b(YX.M) .03.06.57.57.10.04 2.79 .01 Finally, psychological health was investigat ed as a dependent variable in a model where perceived discrimination mediated the relationship between EEODIV and the psychological health. There was support for an overall mediating effect as the paths between EEODIV and perceived discrimination ( B = -.42, t = -3.98, p < .01) and from the mediator to psychological health ( B = -.11, t = -3.44, p < .01) were jointly significant. Perceived EEODIV was related to less perc eived discrimination, which in turn was related to better psychological health. However, the in teraction between EEO and minority status was not significant, indicating that minority status does not moderate the relationship between EEODIV and perceived discrimination. Furt her, the indirect effects for both men and women respondents were bot h significant, but similar in size and direction. Results are shown in Table 37.

PAGE 116

81 Table 37. Results of Moderated Mediation ( PREDICTOR = EEODIV, Med = Sex-based Discrimination, DV = Psychologi cal Health, Mod = Gender) Predictor B SE t p Sex discrimination Constant 44.195.21 8.47.00 EEODIV -.42.11 -3.98.00 Minority Status -.976.06 -.16.87 EEODIV Minority Status .07.12 .54.59 Psychological Health Constant 51.363.58 14.33.00 Sex discrimination -.11.03 -3.44.00 EEODIV .05.07 .76.45 Minority Status -7.183.79 -1.89.06 EEODIV Minority Status .14.08 1.81.07 Minority Status Indirect Effect SE z p Men .05.02 2.56.01 Women .04.01 2.87.00 The simple tests of mediation reveal that the mediation model is not fully supported for men, as one of the paths is not significant. However, the model is fully supported for women. Further, all paths are in the hypothesi zed directions. Results are shown in Table 38.

PAGE 117

82 Table 38. Results of Simple Mediation for Men and Women (PREDICTOR = EEODIV, Med = Sexbased Discrimination, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) .10.071.43.16.23.04 5.76 .00 b(MX) -.43.09-4.51.00-.36.07 -5.18 .00 b(YM.X) -.10.07-1.48.14-.12.04 -3.15 .00 b(YX.M) .06.08.74.46.19.04 4.56 .00 Race-based Discrimination Perceived Equal Employment Opportunity (EEO) and diversity climate in the workplace was i nvestigated as a predictor in a model where race-based discrimination mediates the rela tionship between EEODIV and five dependent variables (job satisfaction, or ganizational commitment, intention to turnover, physical health, and psychological health). Minority status was investig ated as a moderator in the relationship between the predictor and the mediator. Additionally, separate mediation models were tested for each of the co mparison groups (men, women, whites, and nonwhites). First, a model was tested where race discrimination mediates the relationship between EEODIV and job satisfaction. The results of the moderated mediation demonstrated overall support for mediat ion as the path from EEODIV to race discrimination was significant ( B = -.29, t = -4.20, p < .01) in addition to the path from the mediator to job satisfaction ( B = -.05, t = -3.40, p < .01). Perceived EEODIV was

PAGE 118

83 related to less perceived di scrimination, which in turn was related to greater job satisfaction. There was a signi ficant interaction found between the predictor (EEODIV) and the moderator (minority status). Thus, mi nority status does moderate the relationship between EEODIV and perceived race discrimi nation. However, the indirect effects for both men and women were significant and sim ilar in size and direction. Results for the moderated mediation can be found in Table 39. Table 39. Results of Moderated Mediation ( PREDICTOR = EEODIV, Med =Race-based Discrimination, DV = Job Satisfaction, Mod = Race Predictor B SE t p Race discrimination Constant 35.713.33 10.72.00 EEODIV -.29.07 -4.20.00 Minority Status 19.605.14 3.81.00 EEODIV Minority Status -.29.11 -2.73.01 Job satisfaction Constant 8.76.99 8.77.00 Race discrimination -.05.01 -3.40.00 EEODIV .07.02 3.63.00 Minority Status .141.38 .10.92 EEODIV Minority Status .00.03 .03.97 Minority Status Indirect Effect SE z p White .01.01 2.60.01 Non-white .03.01 3.06.00 The separate tests of medi ation reveal support for the mediation model in both whites and non-whites. Also, the paths were in the hypothesized directions. The results for the simple mediation tests for whites and non-whites are displayed in Table 40. Table 40.

PAGE 119

84 Results of Simple Mediation for Whites and Non-whites (PREDICTOR = EEODIV, Med = Race-based Discrimination, DV = Job Satisfaction) White Non-white B SE t p B SE t p b(YX) .08.024.34.00.10.02 4.57 .00 b(MX) -.29.06-4.74.00-.59.09 -6.43 .00 b(YM.X) -.06.02-2.73.01-.04.02 -2.13 .04 b(YX.M) .07.023.34.00.07.02 3.18 .00 Next, a model was tested where the de pendent variable was organizational commitment. Support for overall mediation wa s not found as the path from EEODIV to race discrimination ( B = -.30, t = -4.32, p < .01) was significant but the path from the mediator to organization commitment was not. However, this path was approaching significance. However, there was a signifi cant interaction fou nd between EEODIV and minority status. When examining the indir ect effects, neither whites nor non-whites exhibited significance. Results of the modera ted mediation are displayed in Table 41.

PAGE 120

85 Table 41. Results of Moderated Mediation (PRED ICTOR = EEODIV, Med = Race-based Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Race discrimination Constant 35.533.24 10.94.00 EEODIV -.30.07 -4.32.00 Minority Status 19.175.05 3.79.00 EEODIV Minority Status -.28.11 -2.66.01 Organizational Constant 17.412.56 6.80.00 Race discrimination -.06.04 -1.76.08 EEODIV .29.05 6.07.00 Minority Status -2.503.53 -.71.48 EEODIV Minority Status .09.07 1.18.24 Minority Status Indirect Effect SE z p White .02.01 1.60.11 Non-white .04.02 1.70.09 The results of the separate mediation tests demonstrate that the model is not supported for either whites or non-whites. In both cases, the path between EEODIV and perceived race discrimination is significant, but the path between race discrimination and organizational commitment is not. Howe ver, all paths were in the hypothesized directions. Results for the simple me diation tests are shown in Table 42.

PAGE 121

86 Table 42. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = EEODIV, Med = Race-based Discrimination, DV = Organizational Commitment) White Non-white B SE t p B SE t p b(YX) .31.056.22.00.41.05 8.24 .00 b(MX) -.30.06-4.98.00-.57.09 -6.27 .00 b(YM.X) -.09.06-1.58.12-.04.04 -.97 .34 b(YX.M) .28.055.39.00.39.06 6.97 .00 Next, a model predicting intention to turnover was examined. Support for overall mediation is again found as the paths from EEODIV to perceived discrimination ( B = .31, t = -4.42, p < .01) and from perceived discrimination to intention to turnover ( B = .03, t = 4.96, p < .01) were significant. Perceived EEODIV was related to less perceived discrimination, which in turn was related to less intention to turnover. The interaction between minority status and the independent variables was significant, indicating that relationships between EEODIV and race discri mination may be different for whites and non-whites. However, the indirect effects were significant for bot h white and non-white participants. Results of the moderate d mediation are shown in Table 43.

PAGE 122

87 Table 43. Results of Moderated Mediation (PRED ICTOR = EEODIV, Med = Race-based Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Race discrimination Constant 36.463.31 11.03.00 EEODIV -.31.07 -4.42.00 Minority Status 18.395.12 3.58.00 EEODIV Minority Status -.27.11 -2.52.01 Turnover Constant 3.64.46 7.94.00 Race discrimination .03.01 4.96.00 EEODIV -.03.01 -3.93.00 Minority Status -.97.63 -1.54.12 EEODIV Minority Status .01.01 1.13.26 Minority Status Indirect Effect SE z p White -.01.00 -3.26.00 Non-white -.02.00 -4.05.00 The separate mediation analyses demonstr ated that race discrimination mediates the relationship between EEODIV and inten tion to turnover for both whites and nonwhites. All paths are in the hypot hesized directions. Table 44 displays the results for the simple mediation tests.

PAGE 123

88 Table 44. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = EEODIV, Med = Race-based Discrimination, DV = Intention to Turnover) White Non-white B SE t p B SE t p b(YX) -.04.01-5.07.00-.04.01 -3.62 .00 b(MX) -.31.06-4.91.00-.58.09 -6.41 .00 b(YM.X) .04.014.11.00.03.01 2.96 .00 b(YX.M) -.03.01-3.67.00-.02.01 -1.99 .05 Next, the mediating role of perceived di scrimination in the relationship between EEODIV and physical health was examined. Th e model was supported as the paths from EEODIV to the mediator ( B = -.29, t = -4.18, p < .01) and from per ceived discrimination physical health ( B = -.18, t = -5.94, p < .01) were jointly sign ificant. Perceived EEODIV was related to less perceived discrimination, wh ich in turn was rela ted to better physical health. There was a significant interaction between minority status and EEODIV. Thus, minority status moderated the relationship between EEODIV and race discrimination. Additionally, indirect effects for both white s and non-whites were significant and similar in size. The results of the moderated mediation are displayed in Table 45.

PAGE 124

89 Table 45. Results of Moderated Mediation (PRED ICTOR = EEODIV, Med = Race-based Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Race discrimination Constant 35.443.30 10.71.00 EEODIV -.29.07 -4.18.00 Minority Status 18.635.36 3.47.00 EEODIV Minority Status -.27.11 -2.43.02 Physical Health Constant 66.202.08 31.83.00 Race discrimination -.18.03 -5.94.00 EEODIV .08.04 2.10.04 Minority Status 2.282.96 .77.44 EEODIV Minority Status -.05.06 -.75.45 Minority Status Indirect Effect SE z p White .05.02 3.39.00 Non-white .10.02 4.36.00 The results for the separate tests of mediation are shown in Table 46. Here, perceived discrimination appears to medi ate the relationship between EEODIV and physical health for whites and non-whites. A dditionally, all paths are in the hypothesized directions.

PAGE 125

90 Table 46. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = EEODIV, Med = Race-based Discrimination, DV = Physical Health) White Non-white B SE t p B SE t p b(YX) .13.043.62.00.13.05 2.49 .01 b(MX) -.29.06-4.62.00-.56.10 -5.82 .00 b(YM.X) -.16.04-3.87.00-.19.04 -4.38 .00 b(YX.M) .09.042.34.02.03.06 .45 .66 Finally, psychological health was investigat ed as a dependent variable in a model where perceived discrimination mediated the relationship between EEODIV and the psychological health. There was not full suppor t for the overall mediation model as the path from race discrimination to psychologi cal health was marginally significant. However, all paths were in the hypothesized directions. Perceived EEODIV was related to less perceived discrimination, which in turn was related to better psychological health. However, the interaction between EEODI V and minority status was significant, indicating that minority stat us does moderate the rela tionship between EEODIV and perceived discrimination. Further, the i ndirect effects for bot h whites and non-whites respondents were both non-signi ficant, but similar in size and direction. Results are shown in Table 47.

PAGE 126

91 Table 47. Results of Moderated Mediation (PRED ICTOR = EEODIV, Med = Race-based Discrimination, DV = Psychol ogical Health, Mod = Race) Predictor B SE t p Race discrimination Constant 36.843.37 10.93.00 EEODIV -.31.07 -4.40.00 Minority Status 19.835.37 3.69.00 EEODIV Minority Status -.29.11 -2.63.01 Psychological Health Constant 41.052.43 16.88.00 Race discrimination -.06.03 -1.94.05 EEODIV .23.05 5.13.00 Minority Status 7.993.42 2.33.02 EEODIV Minority Status -.17.07 -2.44.02 Minority Status Indirect Effect SE z p White .02.01 1.74.08 Non-white .04.02 1.85.06 The tests of simple mediation show fu ll support of the meditational model in nonwhite participants, but not in white participants Only one path is significant in the white participants’ model. However, paths are in the hypothesized directi ons. Results are shown in Table 48.

PAGE 127

92 Table 48. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = EEODIV, Med = Race-based Discrimination, DV = Psychological Health) White Non-whites B SE t p B SE t p b(YX) .25.045.76.00.10.06 1.84 .07 b(MX) -.31.06-4.87.00-.61.10 -6.28 .00 b(YM.X) -.02.05-.49.63-.10.05 -2.21 .03 b(YX.M) .25.055.29.00.04.06 .64 .52 Results for Minority Segmentation Sex-based Discrimination Perceived minority segmentation in the workplace was investigated as a predictor in a model wh ere sex-based discrimination mediates the relationship between minority segmentati on and five dependent variables (job satisfaction, organizational commitment, inte ntion to turnover, physical health, and psychological health). Minority st atus was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived sex-based discrimination was investigated as a mediator in the relationship between minority segmentation and job satisfaction. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .84, t = 3.88, p < .01) and between the mediator and job satisfaction ( B = -.06, t = -4.30, p < .01) were jointly si gnificant. Minority se gmentation was related to greater perceived discrimination which in turn was related to le ss job satisfaction. No

PAGE 128

93 significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are significant for both men and women, and were similar in size. The results of the moderated mediation are shown in Table 49. Table 49. Results of Moderated Mediation (PREDIC TOR = Minority Segm entation, Med = Sexbased Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Sex discrimination Constant 14.652.71 5.39.00 MINORITY SEG .84.22 3.88.00 Minority Status 2.983.32 .90.37 MINORITY SEG Minority Status -.10.26 -.36.72 Job satisfaction Constant 12.26.73 16.72.00 Sex discrimination -.06.01 -4.30.00 MINORITY SEG -.01.06 -.14.89 Minority Status 1.24.87 1.43.15 MINORITY SEG Minority Status -.08.07 -1.10.27 Minority Status Indirect Effect SE z p Men -.05.02 -2.84.00 Women -.04.01 -3.22.00 A significant mediating effect was f ound for both men and women. Thus, sexbased discrimination mediates the relations hip between minority segmentation and job satisfaction. Paths are not all in the hypothesized directions. Path coefficients were predicted to be different in directions based on minority stat us. However, paths are in the same direction for both majority and mi nority group members. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceive d discrimination, whereas majority group

PAGE 129

94 members would perceive a nega tive relationship. However, both groups demonstrated a positive relationship. Results for the simple tests of mediation are shown in Table 50. Table 50. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Sex-based Discri mination, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) -.06.06-.90.37-.13.04 -3.26 .00 b(MX) .84.204.14.00.75.15 4.86 .00 b(YM.X) -.10.03-4.06.00-.04.02 -2.53 .01 b(YX.M) .03.06.51.61-.10.04 -2.44 .02 Next, perceived sex-based discrimination wa s investigated as a mediator in the relationship between minority segmentation an d organizational commitment. Support for overall mediation was found as the paths be tween minority segmentation and perceived discrimination ( B = .78, t = 3.62, p < .01) and between the mediator and organizational commitment ( B = -.14, t = -3.99, p < .01) were jointly signific ant. Minority segmentation was related to greater perceived discrimi nation which in turn was related to less organizational commitment. No significant interaction was found be tween the predictor (minority segmentation) and the moderator (mi nority status). Thus, mi nority status is not a moderator in the current model. Additionall y, the indirect effects are significant for both men and women, and were similar in size. Results are shown in Table 51.

PAGE 130

95 Table 51. Results of Moderated Mediation (PREDIC TOR = Minority Segm entation, Med = Sexbased Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Sex discrimination Constant 14.942.70 5.51.00 MINORITY SEG .78.22 3.62.00 Minority Status 2.503.31 .75.45 MINORITY SEG Minority -.02.26 -.08.94 Organizational Constant 32.452.00 16.19.00 Sex discrimination -.14.04 -3.99.00 MINORITY SEG .02.15 .16.88 Minority Status 6.832.36 2.89.00 MINORITY SEG Minority -.42.19 -2.22.03 Minority Status Indirect Effect SE z p Men -.11.04 -2.64.01 Women -.11.04 -3.11.00 The results for the separate tests of mediation are shown in Table 52. A significant mediating effect was found for bot h men and women. Paths are not all in the hypothesized directions. Path co efficients were predicted to be different in direction based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceived discrimination, whereas majority group member s would perceive a negative relationship. However, both groups demonstrated a positive relationship.

PAGE 131

96 Table 52. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Sex-based Discriminat ion, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) -.09.15-.59.56-.50.11 -4.55 .00 b(MX) .78.203.97.00.77.15 4.97 .00 b(YM.X) -.17.07-2.53.01-.13.04 -3.14 .00 b(YX.M) .05.16.29.77-.40.11 -3.52 .00 Next, perceived sex-based discrimination wa s investigated as a mediator in the relationship between minority segmentation and intention to turnover. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .84, t = 3.89, p < .01) and between the mediator and intention to turnover ( B = .03, t = 5.06, p < .01) were jointly significan t. Minority segmentation was associated with greater perceived discrimina tion which in turn was related to greater turnover intention. No signifi cant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Further, the indirect effects were significant for both men and women, and were similar in size Results are displayed in Table 53.

PAGE 132

97 Table 53. Results of Moderated Mediation (PREDIC TOR = Minority Segm entation, Med = Sexbased Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Sex discrimination Constant 14.642.71 5.41.00 MINORITY SEG .84.22 3.89.00 Minority Status 2.893.32 .87.38 MINORITY SEG Minority Status -.08.26 -.32.75 Turnover Constant 2.09.34 6.10.00 Sex discrimination .03.01 5.06.00 MINORITY SEG -.02.03 -.58.56 Minority Status -.78.41 -1.94.05 MINORITY SEG Minority Status .06.03 2.01.05 Minority Status Indirect Effect SE z p Men .03.01 3.05.00 Women .02.01 3.54.00 The results for the separate tests of mediation are shown in Table 54. A significant mediating effect was found for bot h men and women. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceived discrimination, whereas majority group member s would perceive a negative relationship. However, both groups demonstrated a positive relationship.

PAGE 133

98 Table 54. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Sex-based Discriminat ion, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) .01.03.39.70.07.02 3.81 .00 b(MX) .84.204.14.00.76.15 4.93 .00 b(YM.X) .05.014.14.00.03.01 3.43 .00 b(YX.M) -.03.03-1.05.29.05.02 2.74 .01 Next, perceived sex-based discrimination wa s investigated as a mediator in the relationship between minority segmentati on and physical health. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .78, t = 3.55, p < .01) and between the mediator and physical health ( B = -.16, t = -5.89, p < .01) were jointly si gnificant. Minority se gmentation was related to greater perceived discrimination which in turn was related to poorer physical health. No significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Additionally, the indir ect effects are significant for both men and women, and were similar in size. Results for the m oderated mediation are shown in Table 55.

PAGE 134

99 Table 55. Results of Moderated Mediation (PREDIC TOR = Minority Segm entation, Med = Sexbased Discrimination, DV = Phys ical Health, Mod = Gender) Predictor B SE t p Sex discrimination Constant 15.272.76 5.52.00 MINORITY SEG .78.22 3.55.00 Minority Status 2.443.41 .72.47 MINORITY SEG Minority -.06.27 -.21.83 Physical Health Constant 70.371.50 46.93.00 Sex discrimination -.16.03 -5.89.00 MINORITY SEG .04.12 .34.73 Minority Status .151.77 .08.93 MINORITY SEG Minority -.11.14 -.79.43 Minority Status Indirect Effect SE z p Men -.13.04 -3.01.00 Women -.12.03 -3.58.00 A significant mediating eff ect was found for women and men. Paths are not all in the hypothesized directions. Path coefficients we re predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceived discrimination, whereas majority group member s would perceive a negative relationship. However, both groups demonstrated a positive re lationship. Results for the simple tests of mediation are shown in Table 56.

PAGE 135

100 Table 56. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Sex-based Discri mination, DV = Physical Health) Men Women B SE T p B SE t p b(YX) -.09.11-.78.44-.19.09 -2.14 .03 b(MX) .78.213.77.00.72.16 4.46 .00 b(YM.X) -.16.05-3.16.00-.17.03 -4.95 .00 b(YX.M) .03.11.30.77-.07.09 -.79 .43 Finally, perceived sex-based discrimination was investigated as a mediator in the relationship between minority segmentation an d psychological health. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .77, t = 3.54, p < .01) and between the mediator and psychological health ( B = -.14, t = -4.27, p < .01) were jointly significan t. Minority segmentation was related to greater perceived discriminati on which in turn was related to poorer psychological health. No significant inte raction was found between the predictor (minority segmentation) and the moderator (mi nority status). Thus, mi nority status is not a moderator in the current model. The indir ect effects are significant for both men and women, and were similar in size. Re sults are displayed in Table 57.

PAGE 136

101 Table 57. Results of Moderated Mediation (PREDIC TOR = Minority Segm entation, Med = Sexbased Discrimination, DV = Psychological Health, Mod = Gender) Predictor B SE t p Sex discrimination Constant 15.002.72 5.52.00 MINORITY SEG .77.22 3.54.00 Minority Status 3.343.37 .99.32 MINORITY SEG Minority Status -.07.27 -.24.81 Psychological Health Constant 56.131.78 31.55.00 Sex discrimination -.14.03 -4.27.00 MINORITY SEG -.14.14 -1.02.31 Minority Status -.562.13 -.26.79 MINORITY SEG Minority Status -.03.17 -.18.86 Minority Status Indirect Effect SE z p Men -.11.04 -2.68.01 Women -.10.03 -3.06.00 A significant mediating effect was found for women, but the model was not fully supported for men. Thus, sex-based discrimi nation mediates the relationship between minority segmentation and psychological heal th for women, but not for men. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Speci fically, it was hypothesized that minority group members would perceive a positive re lationship between minority segmentation and perceived discrimination, whereas ma jority group members would perceive a negative relationship. However, both groups de monstrated a positive relationship. Results of the simple tests of medi ation are shown in Table 58.

PAGE 137

102 Table 58. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Sex-based Discrimi nation, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) -.25.14-1.81.07-.27.10 -2.66 .01 b(MX) .77.193.98.00.71.16 4.30 .00 b(YM.X) -.09.06-1.42.16-.16.04 -4.12 .00 b(YX.M) -.18.15-1.22.23-.16.10 -1.57 .12 Race-based Discrimination Perceived minority segmentation in the workplace was investigated as a predictor in a model where race-based discrimination mediates the relationship between minority segmentati on and five dependent variables (job satisfaction, organizational commitment, inte ntion to turnover, physical health, and psychological health). Minority st atus was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived race-based discrimination was investigated as a mediator in the relationship between minority segmentation and job satisfaction. Support for overall mediation was not found, as only one of the paths in the model, the path between the mediator and job satisfaction ( B = -.06, t = -4.70, p < .01), was significant. However, the other path was approaching significance. No significant interac tion was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in th e current model. However, the interaction effect approached significance in this case ( B = .44, t = 1.74, p = .08). The indirect effects indicate a

PAGE 138

103 difference between groups. Wh ile the indirect effect fo r non-whites is significant (Indirect effect = -.05, z = -3.12, p < .01), the indirect effect for white participants is approaching significance. Results for the m oderated mediation are shown in Table 59. Table 59. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Racebased Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Race discrimination Constant 18.222.12 8.57.00 MINORITY SEG .34.17 1.98.05 Minority Status -.853.25 -.26.79 MINORITY SEG Minority .44.25 1.74.08 Job Satisfaction Constant 13.49.59 22.74.00 Race discrimination -.06.01 -4.70.00 MINORITY SEG -.11.04 -2.42.02 Minority Status -.44.83 -.53.60 MINORITY SEG Minority .07.07 1.08.28 Minority Status Indirect Effect SE z p Whites -.02.01 -1.79.07 Non-whites -.05.02 -3.12.00 The results for the separate tests of medi ation indicated support for the mediation model for both white and non-white participan ts. Paths are not all in the hypothesized directions. Path coefficients were predicted to be different in dire ctions based on minority status. However, paths are in the same di rection for both majority and minority group members. Specifically, it wa s hypothesized that minority gr oup members would perceive a positive relationship between minority se gmentation and perceived discrimination, whereas majority group members would perceive a negative relations hip. However, both

PAGE 139

104 groups demonstrated a positive relationship. Ta ble 60 displays the results for the simple tests of mediation. Table 60. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Race-based Discri mination, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) -.13.05-2.84.01-.09.05 -1.78 .08 b(MX) .35.152.26.02.79.21 3.77 .00 b(YM.X) -.07.02-3.52.00-.05.02 -3.17 .00 b(YX.M) -.11.05-2.34.02-.04.05 -.89 .37 Next, perceived race-based discrimination was investigated as a mediator in the relationship between minority segmentation an d organizational commitment. Support for overall mediation was found as the paths be tween minority segmentation and perceived discrimination ( B = .35, t = 2.02, p < .05) and between the mediator and organizational commitment ( B = -.15, t = -4.19, p < .01) were jointly signific ant. Minority segmentation was related to greater perceived discrimi nation which in turn was related to less organizational commitment. No significant interaction was found be tween the predictor (minority segmentation) and the moderator (mi nority status). Thus, mi nority status is not a moderator in the current model. However, similar to the findings for job satisfaction, the interaction effect a pproached significance ( B = .43, t = 1.69, p = .09). The indirect effects indicate a difference between groups. The indirect effect is significant for non-

PAGE 140

105 white participants (Indirect effect = -.12, z = -2.93, p < .01), but not so for white participants. Results are displayed in Table 61. Table 61. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Racebased Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Race discrimination Constant 18.062.10 8.57.00 MINORITY SEG .35.17 2.02.04 Minority Status -.653.24 -.20.84 MINORITY SEG Minority .43.25 1.69.09 Organizational Constant 35.971.61 22.34.00 Race discrimination -.15.04 -4.19.00 MINORITY SEG -.28.12 -2.31.02 Minority Status 3.242.27 1.42.16 MINORITY SEG Minority -.04.18 -.23.82 Minority Status Indirect Effect SE z p Whites -.05.03 -1.79.07 Non-whites -.12.04 -2.93.00 The results for the separate tests of mediation indicated full support for the models for both white and non-white particip ants. Paths are not al l in the hypothesized directions. Path coefficients were predicted to be different in dire ctions based on minority status. However, paths are in the same di rection for both majority and minority group members. Specifically, it wa s hypothesized that minority gr oup members would perceive a positive relationship between minority se gmentation and perceived discrimination, whereas majority group members would perceive a negative relations hip. However, both groups demonstrated a positive relationship. Results are shown in Table 62.

PAGE 141

106 Table 62. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Race-based Discriminat ion, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) -.33.13-2.62.01-.44.13 -3.44 .00 b(MX) .35.152.35.02.78.21 3.66 .00 b(YM.X) -.18.06-3.15.00-.13.04 -2.86 .00 b(YX.M) -.27.13-2.14.03-.34.13 -2.61 .01 Next, perceived race-based discrimination was investigated as a mediator in the relationship between minority segmentation and intention to turnover. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .38, t = 2.18, p < .05) and between the mediator and intention to turnover ( B = .04, t = 5.98, p < .01). Moreover, all paths were in the predicted direction. No significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. However, the interaction e ffect was approaching significance ( B = .43, t = 1.70, p = .09). The indirect effects were both significan t and similar in size and direction. Results are shown in Table 63.

PAGE 142

107 Table 63. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Racebased Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Race discrimination Constant 17.982.13 8.42.00 MINORITY SEG .38.17 2.18.03 Minority Status -.963.27 -.29.77 MINORITY SEG Minority .43.26 1.70.09 Turnover Constant 1.43.27 5.23.00 Race discrimination .04.01 5.98.00 MINORITY SEG .05.02 2.36.02 Minority Status .11.38 .29.77 MINORITY SEG Minority -.04.03 -1.26.21 Minority Status Indirect Effect SE z p Whites .01.01 2.02.04 Non-whites .03.01 3.51.00 A significant mediating effect wa s found for both white and non-white participants. Paths are not al l in the hypothesized directions. Path coefficients were predicted to be different in directions based on minority stat us. However, paths are in the same direction for both majority and mi nority group members. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceive d discrimination, whereas majority group members would perceive a nega tive relationship. However, both groups demonstrated a positive relationship. Results for the simple tests of mediation are shown in Table 64.

PAGE 143

108 Table 64. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Race-based Discriminat ion, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) .06.022.92.00.04.02 1.73 .08 b(MX) .38.162.45.02.82.21 3.92 .00 b(YM.X) .04.014.75.00.03.01 3.79 .00 b(YX.M) .05.022.24.03.02.02 .65 .51 Next, perceived race-based discrimination was investigated as a mediator in the relationship between minority segmentati on and physical health. Support for overall mediation was not found as the paths betw een minority segmentation and perceived discrimination was approaching significance while the path between the mediator and physical health ( B = -.20, t = -6.92, p < .01) was significant. Minority segmentation trended toward a relationship with greater perceived discrimination which in turn was related to poorer physical health. There wa s a significant inter action was found between the predictor (minority segmentation) and the moderator ( B = .72, t = 2.75, p < .05). Thus, minority status is a moderator in th e current model. The indirect effects are supportive of the interaction effect. The indi rect effects are significant for non-white participants, but not so for white part icipants. Results are shown in Table 65.

PAGE 144

109 Table 65. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Racebased Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Race discrimination Constant 18.712.10 8.89.00 MINORITY SEG .30.17 1.71.09 Minority Status -4.553.34 -1.36.17 MINORITY SEG Minority .72.26 2.75.01 Physical Health Constant 71.931.23 58.45.00 Race discrimination -.20.03 -6.92.00 MINORITY SEG -.13.09 -1.44.15 Minority Status -2.411.77 -1.36.17 MINORITY SEG Minority .23.14 1.62.11 Minority Status Indirect Effect SE z p Whites -.06.04 -1.64.10 Non-whites -.20.05 -4.13.00 The results for the separate tests of mediation indicated full support of the mediation model for non-white participants, bu t marginal support for white participants as one of paths was significant while the ot her was approaching si gnificance. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Speci fically, it was hypothesized that minority group members would perceive a positive re lationship between minority segmentation and perceived discrimination, whereas ma jority group members would perceive a negative relationship. However, both groups de monstrated a positive relationship. Results are shown in Table 66.

PAGE 145

110 Table 66. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Race-based Discri mination, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) -.23.12-1.97.05-.11.12 -.89 .37 b(MX) .38.162.38.021.02.21 4.75 .00 b(YM.X) -.10.05-1.98.05-.22.04 -5.40 .00 b(YX.M) -.19.12-1.64.10.12.12 1.04 .30 Finally, perceived race-based discrimination was investigated as a mediator in the relationship between minority segmentation an d psychological health. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .38, t = 2.11, p < .05) and between the mediator and psychological health ( B = -.09, t = -2.89, p < .01) were jointly significan t. Minority segmentation was related to greater perceived discriminati on which in turn was related to poorer psychological health. No significant inte raction was found between the predictor (minority segmentation) and the moderator (mi nority status). Thus, mi nority status is not a moderator in the current model. The indir ect effects were different for white and nonwhite participants, however. Non-white particip ants exhibited a significant indirect effect (Indirect effect = -.07, z = -2.24, p < .05), whereas white participants did not. Results for the moderated mediation are shown in Table 67.

PAGE 146

111 Table 67. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Racebased Discrimination, DV = Psychological Health, Mod = Race) Predictor B SE t p Race discrimination Constant 18.072.21 8.19.00 MINORITY SEG .38.18 2.11.03 Minority Status .283.44 .08.93 MINORITY SEG Minority .35.27 1.31.19 Psychological Health Constant 54.511.46 37.33.00 Race discrimination -.09.03 -2.89.00 MINORITY SEG -.19.11 -1.73.09 Minority Status 1.112.09 .53.60 MINORITY SEG Minority -.05.16 -.32.75 Minority Status Indirect Effect SE z p Whites -.04.02 -1.64.10 Non-whites -.07.03 -2.24.03 Non-white participants exhibited suppor t for the mediation model in the simple test of mediation. However, only one of path s in the model of white participants was significant, while the other was approachi ng significance. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceived discrimination, whereas majority group member s would perceive a negative relationship. However, both groups demonstrated a positiv e relationship. Results are displayed in Table 68.

PAGE 147

112 Table 68. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Race-based Discrimi nation, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) -.23.12-1.97.05-.31.12 -2.69 .01 b(MX) .38.162.38.02.73.22 3.32 .00 b(YM.X) -.10.05-1.98.05-.09.04 -2.13 .03 b(YX.M) -.19.12-1.64.10-.25.12 -2.09 .04 Results for Social Support Sex-based Discrimination Social support in the wor kplace was investigated as a predictor in a model where sex-based discri mination mediates the relationship between social support and five dependent variable s (job satisfaction, orga nizational commitment, intention to turnover, physica l health, and psychological h ealth). Minority status was investigated as a moderator in the relations hip between the predictor and the mediator. First, perceived sex-based discrimination was investigated as a mediator in the relationship between social support and j ob satisfaction. Support for overall mediation was not found as the paths between social support and perceived discrimination ( B = -.59, t = -3.71, p < .01) was significant, but the path be tween the mediator and job satisfaction was not significant. No signi ficant interaction wa s found between the predictor (social support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are similar for men and women, in that they are

PAGE 148

113 both non-significant, and nearly identical in size. Results fo r the moderated mediation are shown in Table 69. Table 69. Results of Moderated Mediation (PREDIC TOR = Social Support, Med = Sex-based Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Sex discrimination Constant 41.614.62 9.01.00 SOCIAL SUPPORT -.59.16 -3.71.00 Minority Status 1.055.48 .19.84 SUPPORT* Minority Status .04.19 .22.83 Job Satisfaction Constant 3.171.14 2.78.01 Sex discrimination -.02.01 -1.79.07 SOCIAL SUPPORT .28.04 7.89.00 Minority Status 2.131.22 1.75.08 SUPPORT* Minority Status -.07.04 -1.64.10 Minority Status Indirect Effect SE z p Men .01.01 1.57.12 Women .01.01 1.68.09 The results for the separate tests of me diation indicate that the model was not fully supported in either men or women. He re, the paths between social support and perceived discrimination are significant, but the other paths are not. However, all paths were in the hypothesized directions. Results are shown in Table 70.

PAGE 149

114 Table 70. Results of Simple Mediation for Men and Women (PREDICTOR = Social Support, Med = Sex-based Discrimination, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .30.048.41.00.23.02 10.16 .00 b(MX) -.59.16-3.78.00-.55.10 -5.45 .00 b(YM.X) -.03.02-1.45.15-.02.01 -1.20 .23 b(YX.M) -.28.047.45.00.22.02 9.23 .00 Next, perceived sex-based discrimination wa s investigated as a mediator in the relationship between social support and or ganizational commitment. Support for overall mediation was not found as the path between social support and perceived discrimination ( B = -.58, t = -3.66, p < .01) was significant, but the path between the mediator and organizational commitment wa s not significant. No signifi cant interaction was found between the predictor (social support) and the moderator (minority status). Thus, minority status is not a moderator in the curren t model. The indirect effects are similar for men and women, as they are both non-significan t and similar in size. Results are shown in Table 71.

PAGE 150

115 Table 71. Results of Moderated Mediation (PREDIC TOR = Social Support, Med = Sex-based Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Sex discrimination Constant 40.694.55 8.93.00 SOCIAL SUPPORT -.58.16 -3.66.00 Minority Status 1.735.38 .32.75 SUPPORT Minority Status .03.19 .17.87 Organizational Constant 14.083.05 4.61.00 Sex discrimination -.05.03 -1.71.09 SOCIAL SUPPORT .59.10 6.10.00 Minority Status -1.673.26 -.51.61 SUPPORT Minority Status .10.11 .87.39 Minority Status Indirect Effect SE z p Men .03.02 1.50.13 Women .03.02 1.61.11 The results for the separate tests of mediation are shown in Table 72. The mediating effect was not fully supported for either men or women, as one of the two paths was not significant in bath cases. However, the path between perceived discrimination and organizati onal commitment was approa ching significance for the female sample.

PAGE 151

116 Table 72. Results of Simple Mediation for Men and Women (PREDICTOR = Social Support, Med = Sex-based Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) .62.096.99.00.72.06 11.77 .00 b(MX) -.58.15-3.85.00-.54.10 -5.44 .00 b(YM.X) -.03.06-.47.64-.06.04 -1.68 .09 b(YX.M) .61.106.36.00.68.06 10.62 .00 Next, perceived sex-based discrimination wa s investigated as a mediator in the relationship between social support and intention to tu rnover. Support for overall mediation was found as the paths between so cial support and percei ved discrimination ( B = -.59, t = -3.74, p < .01) and between the mediator and intention to turnover ( B = .02, t = 3.20, p < .01) were jointly significant. Social support was related to less perceived discrimination which in turn was related to less turnover intention. No significant interaction was found between th e predictor (social support) and the mo derator (minority status). Thus, minority status is not a moderator in the curren t model. The indirect effects were significant for both men and women, and we re similar in size and direction. Results of the moderated mediation are shown in Table 73.

PAGE 152

117 Table 73. Results of Moderated Mediation (PREDIC TOR = Social Support, Med = Sex-based Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Sex discrimination Constant 41.614.58 9.08.00 SOCIAL SUPPORT -.59.16 -3.74.00 Minority Status 1.675.41 .31.76 SUPPORT Minority Status .02.19 .09.93 Turnover Constant 4.88.58 8.47.00 Sex discrimination .02.00 3.20.00 SOCIAL SUPPORT -.10.02 -5.20.00 Minority Status -.33.61 -.54.59 SUPPORT Minority Status .01.02 .67.50 Minority Status Indirect Effect SE z p Men -.01.00 -2.39.02 Women -.01.00 -2.78.01 The results for the separate tests of me diation are shown in Table 74. The model was fully supported for women, but one of the two paths was approaching significance in the male sample ( B = .02, t = 1.90, p = .06). All paths were in the hypothesized directions.

PAGE 153

118 Table 74. Results of Simple Mediation for Men and Women (PREDICTOR = Social Support, Med = Sex-based Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) -.11.02-6.07.00-.09.01 -8.02 .00 b(MX) -.59.16-3.78.00-.57.10 -5.82 .00 b(YM.X) .02.011.90.06.02.01 2.60 .01 b(YX.M) -.09.02-5.13.00-.08.01 -6.73 .00 Next, perceived sex-based discrimination wa s investigated as a mediator in the relationship between social support and phys ical health. Support for overall mediation was found as the paths between social support and perceived discrimination ( B = -.49, t = -2.95, p < .01) and between the medi ator and physical health ( B = -.14, t = -5.02, p < .01) were jointly significant. Social support was related to less perceive d discrimination which in turn was related to better physical healt h. No significant intera ction was found between the predictor (social support) and the mode rator (minority status). Additionally, the indirect effects are significant for both men and women, and were similar in size. Results are displayed in Table 75.

PAGE 154

119 Table 75. Results of Moderated Mediation (PREDIC TOR = Social Support, Med = Sex-based Discrimination, DV = Physical Health, Mod = Gender) Predictor B SE t p Sex discrimination Constant 38.664.83 7.99.00 SOCIAL SUPPORT -.49.17 -2.95.00 Minority Status 2.405.69 .42.67 SUPPORT Minority Status -.02.19 -.08.93 Physical Health Constant 67.132.75 24.37.00 Sex discrimination -.14.03 -5.02.00 SOCIAL SUPPORT .12.08 1.38.17 Minority Status -2.312.96 -.78.44 SUPPORT Minority Status .02.10 .24.81 Minority Status Indirect Effect SE z p Men .07.03 2.50.01 Women .07.02 3.49.00 The results for the separate tests of mediation are shown in Table 76. A significant mediating effect was found for bot h men and women. All paths were in the hypothesized directions.

PAGE 155

120 Table 76. Results of Simple Mediation for Men and Women (PREDICTOR = Social Support, Med = Sex-based Discrimination, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .19.082.31.02.22.06 3.86 .00 b(MX) -.49.16-2.98.00-.51.10 -4.95 .00 b(YM.X) -.14.05-2.90.00-.15.04 -4.11 .00 b(YX.M) .12.081.46.15.15.06 2.52 .01 Finally, perceived sex-based discrimination was investigated as a mediator in the relationship between social support and psychological health. Support for overall mediation was found as the paths between so cial support and percei ved discrimination ( B = -.55, t = -3.40, p < .01) and between the mediator and psychological health ( B = -.09, t = -2.93, p < .01) were jointly signif icant. Social support was related to less perceived discrimination which in turn was related to better psychological he alth. No significant interaction was found between th e predictor (social support) and the mo derator (minority status). Thus, minority status is not a mode rator in the current model. Additionally, the indirect effects were significant for both men and women, and were similar in size. Results are displayed in Table 77.

PAGE 156

121 Table 77. Results of Moderated Mediation (PREDIC TOR = Social Support, Med = Sex-based Discrimination, DV = Psychologi cal Health, Mod = Gender) Predictor B SE t p Sex discrimination Constant 39.804.68 8.51.00 SOCIAL SUPPORT -.55.16 -3.40.00 Minority Status 2.395.56 .43.67 SUPPORT Minority Status .02.19 .08.93 Psychological Health Constant 45.983.16 14.56.00 Sex discrimination -.09.03 -2.93.00 SOCIAL SUPPORT .26.10 2.59.01 Minority Status -1.433.41 -.42.68 SUPPORT Minority Status .02.12 .14.88 Minority Status Indirect Effect SE z p Men .05.02 2.16.03 Women .05.02 2.51.01 The results for the separate tests of me diation indicated a difference between men and women. The model was fully supported for women, but not supported for men. All paths were in the hypothesized directions Results are displayed in Table 78. Table 78. Results of Simple Mediation for Men and Women (PREDICTOR = Social Support, Med = Sex-based Discrimination, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) .31.103.11.00.33.06 5.16 .00 b(MX) -.55.15-3.67.00-.53.11 -4.97 .00 b(YM.X) -.04.07-.53.60-.12.04 -3.12 .00 b(YX.M) .29.112.74.01.27.07 4.03 .00

PAGE 157

122 Race-based Discrimination Social support in the wo rkplace was investigated as a predictor in a model where race-based disc rimination mediates the relationship between social support and five dependent variable s (job satisfaction, orga nizational commitment, intention to turnover, physica l health, and psychological h ealth). Minority status was investigated as a moderator in the relations hip between the predictor and the mediator. First, perceived race-based discrimination was investigated as a mediator in the relationship between social support and j ob satisfaction. Support for overall mediation was found as the paths between social support and perceived discrimination ( B = -.33, t = -2.82, p < .05) and between the medi ator and job satisfaction ( B = -.03, t = -2.41, p < .05) were jointly significant. Social support was related to less perceive d discrimination which in turn was related to grea ter job satisfaction. A significan t interaction was found between the predictor (social supp ort) and the moderator ( B = -.36, t = -2.04, p < .05). Thus, minority status is a moderator in the current model. The indirect effects were also different for white and non-white participants. The indirect effect was significant for nonwhite participants (Indirect effect = .02, z = 2.14, p < .05), but non-significant for white participants. Results are shown in Table 79.

PAGE 158

123 Table 79. Results of Moderated Mediation (PREDIC TOR = Social Support, Med = Race-based Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Race discrimination Constant 31.853.33 9.54.00 SOCIAL SUPPORT -.33.12 -2.82.01 Minority Status 15.755.18 3.03.00 SUPPORT Minority Status -.36.18 -2.04.04 Job Satisfaction Constant 4.54.80 5.68.00 Race discrimination -.03.01 -2.41.02 SOCIAL SUPPORT .25.03 9.84.00 Minority Status .421.12 .38.71 SUPPORT Minority Status -.02.04 -.51.61 Minority Status Indirect Effect SE z p Whites .01.01 1.76.08 Non-whites .02.01 2.14.03 The results for the separate tests of mediation are shown in Table 80. A significant mediating effect was found for white participants, but the model was not fully supported for non-white participants. Howeve r, all paths were in the hypothesized directions.

PAGE 159

124 Table 80. Results of Simple Mediation for Whites and Non-whites (PREDIC TOR = Social Support, Med = Race-based Discrimina tion, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) .26.0210.55.00.25.03 8.31 .00 b(MX) -.33.11-3.12.00-.69.15 -4.70 .00 b(YM.X) -.03.02-2.08.04-.02.02 -1.38 .17 b(YX.M) .25.029.92.00.23.03 7.36 .00 Next, perceived race-based discrimination was investigated as a mediator in the relationship between social support and or ganizational commitment. Support for overall mediation was found as the paths between so cial support and percei ved discrimination ( B = -.35, t = -3.03, p < .01) and between the mediator and organizational commitment ( B = -.07, t = -2.34, p < .05) were jointly significant. So cial support was related to less perceived discrimination which in turn was re lated to greater organizational commitment. The interaction between the predictor (social support) and the moderator was marginally significant ( B = -.32, t = -1.81, p = .07). Additionally, the indirect effects are different for white and non-white participants, as the indirect effect is significant for non-white participants, but not for white particip ants. Results are shown in Table 81.

PAGE 160

125 Table 81. Results of Moderated Mediation (PREDIC TOR = Social Support, Med = Race-based Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Race discrimination Constant 32.133.27 9.82.00 SOCIAL SUPPORT -.35.11 -3.03.00 Minority Status 14.615.11 2.86.00 SUPPORT Minority Status -.32.18 -1.81.07 Organizational Constant 11.832.10 5.61.00 Race discrimination -.07.03 -2.34.02 SOCIAL SUPPORT .70.07 10.49.00 Minority Status 3.102.94 1.05.29 SUPPORT Minority Status -.07.10 -.75.46 Minority Status Indirect Effect SE z p Whites .02.01 1.80.07 Non-whites .05.02 2.09.04 The results for the separate tests of me diation indicate that the model was not fully supported for either white or non-white pa rticipants. However, al l paths were in the hypothesized directions. Results of the simple tests of mediation are shown in Table 82.

PAGE 161

126 Table 82. Results of Simple Mediation for Whites and Non-whites (PREDIC TOR = Social Support, Med = Race-based Discrimination, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) .72.0710.97.00.67.08 8.77 .00 b(MX) -.35.10-3.43.00-.67.15 -4.48 .00 b(YM.X) -.07.05-1.46.15-.07.04 -1.84 .07 b(YX.M) .70.0710.32.00.62.08 7.72 .00 Next, perceived race-based discrimination was investigated as a mediator in the relationship between social support and intention to tu rnover. Support for overall mediation was found as the paths between so cial support and percei ved discrimination ( B = -.37, t = -3.26, p < .01) and between the mediator and intention to turnover ( B = .02, t = 4.09, p < .01) were jointly significant. Social support was related to less perceived discrimination which in turn was related to less intention to turnover. A significant interaction was found between the predictor (social s upport) and the moderator ( B = -.38, t = -2.15, p < .05). Thus, minority status is a modera tor in the current mo del. The indirect effects are similar for white and non-white pa rticipants, as they are both significant and similar in size and direction. Re sults are shown in Table 83.

PAGE 162

127 Table 83. Results of Moderated Mediation (PREDIC TOR = Social Support, Med = Race-based Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Race discrimination Constant 33.253.27 10.16.00 SOCIAL SUPPORT -.37.11 -3.26.00 Minority Status 15.975.12 3.12.00 SUPPORT Minority Status -.38.18 -2.15.03 Turnover Constant 5.16.40 13.04.00 Race discrimination .02.01 4.09.00 SOCIAL SUPPORT -.10.01 -8.38.00 Minority Status -1.32.55 -2.39.02 SUPPORT Minority Status .04.02 2.23.03 Minority Status Indirect Effect SE z p Whites -.01.00 -2.50.01 Non-whites -.02.01 -3.27.00 The results for the separate tests of mediation indicate full support for the mediation model for both white and non-wh ite participants. A ll paths were in hypothesized directions. Table 84 displays the re sults for the simple tests of mediation.

PAGE 163

128 Table 84. Results of Simple Mediation for Whites and Non-whites (PREDIC TOR = Social Support, Med = Race-based Discrimination, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) -.11.01-9.54.00-.08.02 -5.13 .00 b(MX) -.37.11-3.54.00-.75.14 -5.19 .00 b(YM.X) .03.013.47.00.02.01 2.39 .02 b(YX.M) -.10.01-8.64.00-.06.02 -3.93 .00 Next, perceived race-based discrimination was investigated as a mediator in the relationship between social support and phys ical health. Support was found for overall mediation as the path between social support and perceived discrimination ( B = -.32, t = 2.71, p < .05) and the path between the mediator and physical health ( B = -.18, t = -6.07, p < .01) was jointly significant. No si gnificant interaction was found between the predictor (social support) and th e moderator (minority status). Thus, minority status is not a moderator in the current model. Additionally, the indirect effects were similar for white and non-white participants, as they were both si gnificant and similar in size and direction. Results are displayed in Table 85.

PAGE 164

129 Table 85. Results of Moderated Mediation (PREDIC TOR = Social Support, Med = Race-based Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Race discrimination Constant 31.343.36 9.32.00 SOCIAL SUPPORT -.32.12 -2.71.01 Minority Status 12.895.34 2.42.02 SUPPORT Minority Status -.27.18 -1.49.14 Physical Health Constant 66.171.98 33.49.00 Race discrimination -.18.03 -6.07.00 SOCIAL SUPPORT .13.06 2.19.03 Minority Status 1.642.80 .59.56 SUPPORT Minority Status -.06.10 -.60.55 Minority Status Indirect Effect SE z p Whites .06.02 2.45.01 Non-whites .10.03 3.42.00 The results for the separate tests of medi ation indicate that a significant mediating effect was found for both white and non-white participants. All paths were in the hypothesized directions. Tabl e 86 displays the results. Table 86. Results of Simple Mediation for Whites and Non-whites (PREDIC TOR = Social Support, Med = Race-based Discrimination, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) .19.063.20.00.18.08 2.20 .03 b(MX) -.32.11-2.93.00-.59.15 -3.85 .00 b(YM.X) -.15.04-3.78.00-.20.04 -4.67 .00 b(YX.M) .14.062.42.02.06.08 .80 .43

PAGE 165

130 Finally, perceived race-based discrimination was investigated as a mediator in the relationship between social support and ps ychological health. Support was not found for overall mediation as the path between soci al support and perceived discrimination was significant ( B = -.41, t = -3.38, p < .01) but the path be tween the mediator and psychological health was not. No significant interaction was found be tween the predictor (social support) and the mode rator (minority status). Thus minority status is not a moderator in the current model. The indirect effects were not significant for either white or non-white participants. Results are displayed in Table 87. Table 87. Results of Moderated Mediation (PREDIC TOR = Social Support, Med = Race-based Discrimination, DV = Psychol ogical Health, Mod = Race) Predictor B SE t p Race discrimination Constant 34.343.45 9.97.00 SOCIAL SUPPORT -.41.12 -3.38.00 Minority Status 12.215.42 2.26.02 SUPPORT Minority Status -.24.19 -1.31.19 Psychological Health Constant 42.042.29 18.31.00 Race discrimination -.05.03 -1.58.11 SOCIAL SUPPORT .34.07 4.67.00 Minority Status 2.623.19 .82.41 SUPPORT Minority Status -.10.11 -.93.35 Minority Status Indirect Effect SE z p Whites .02.01 1.38.17 Non-whites .03.02 1.46.14 The results for the separate tests of mediation are shown in Table 88. A significant mediating effect wa s not found for either white or non-white participants, as

PAGE 166

131 the path between perceived discrimination and psychological health was non-significant in both cases. However, all paths we re in the hypothesized directions. Table 88. Results of Simple Mediation for Whites and Non-whites (PREDIC TOR = Social Support, Med = Race-based Discriminatio n, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) .36.075.01.00.27.08 3.19 .00 b(MX) -.41.11-3.72.00-.65.16 -4.15 .00 b(YM.X) -.03.05-.67.50-.07.04 -1.54 .13 b(YX.M) .34.074.65.00.22.09 2.55 .01 Summary of Results for the Full Demonstration Because of the large amount of findings three tables have been created to summarize the significant results related to the three predictors. Tables 89, 90, and 91 highlight the significant findings from each of the mediation models tested.

PAGE 167

132 Table 89. Summary of Significant Resu lts for Equal Employment O pportunity/Diversity Climate Path Overall model Interaction Majority Indirect Effect Minority Indirect Effect Majority Simple Mediation Minority Simple Mediation Sex-based Job Satisfaction X X X X X Sex-based Org Commit X X X X Sex-based Turnover X X X X X Sex-based Physical health X X X X X Sex-based Psych health X X X X Race-based JobSatisfaction X X X X X X Race-based Org Commit X X Race-based Turnover X X X X X X Race-based Physical health X X X X X X Race-based Psych health X X X indicates a significant finding at p < .05 Table 90. Summary of Significant Resu lts for Minority Segmentation Path Overall model Interaction Majority Indirect Effect Minority Indirect Effect Majority Simple Mediation Minority Simple Mediation Sex-based Job Satisfaction X X X X X Sex-based Org Commit X X X X X Sex-based Turnover X X X X X Sex-based Physical health X X X X X Sex-based Psych health X X X X Race-based JobSatisfaction X X X Race-based Org Commit X X X X Race-based Turnover X X X X X Race-based Physical health X X X X Race-based Psych health X X X X indicates a significant finding at p < .05

PAGE 168

133 Table 91. Summary of Significant Re sults for Social Support Path Overall model Interaction Majority Indirect Effect Minority Indirect Effect Majority Simple Mediation Minority Simple Mediation Sex-based Job Satisfaction Sex-based Org Commit Sex-based Turnover X X X X Sex-based Physical health X X X X X Sex-based Psych health X X X X Race-based JobSatisfaction X X X X Race-based Org Commit X X Race-based Turnover X X X X X X Race-based Physical health X X X X X Race-based Psych health X indicates a significant finding at p < .05 Analysis of Coworker Data Data on the organizational anteceden ts only (i.e. EEO, minority segmentation, diversity climate, instrumental social support, emotional social suppor t, and token status) was collected from coworkers of the partic ipants in the sample. The purpose of the coworker survey was to analyze secondary s ource data about the work environment. In total, twenty-nine coworkers re sponded to the survey. Particip ants were asked to provide a three-word code that would be used to match them with their coworker. Twenty-one coworkers could be matched based on the matc hing criteria. Table 92 displays descriptive statistics for each of the organizational antecedents, comparing coworkers to study participants.

PAGE 169

134 Table 92. Descriptive Statistics for Organizational An tecedents Comparing Study Participants to Coworkers Mean SD r^ EEO (sample) 15.153.59 .38 EEO (coworker) 13.674.32 Minority Segmentation (sample) 11.484.34.50* Minority Segmentation (coworker) 12.484.64 Diversity Climate (sample) 33.386.16.53* Diversity Climate (coworker) 30.788.27 Instrumental Support (sample) 13.603.39.53* Instrumental Support (coworker) 14.243.27 Emotional Support (sample) 14.953.01 .40 Emotional Support (coworker) 15.443.54 Gender Tokenism (sample) 4.051.36 .37 Gender Tokenism (coworker) 4.151.14 Racial Tokenism (sample) 3.901.51.67* Racial Tokenism (coworker) 4.151.31 ^ r reflects the correlation between study participants and coworkers Indicates that r is significant at p < .05 Study participants were instructed to provi de the coworker surv ey to an individual in the workplace holding a similar position, and sharing similar demographic characteristics. Ninety-five percent of the co worker respondents indicated the survey was sent to them by a coworker, rather than a s upervisor or subordinate Seventy-one percent

PAGE 170

135 of the coworker respondents reported shar ing the same position with their matched study participant, 76% reported shar ing the same supervisor, 86% reported working in the same department, and 86% reported working at the same physical location. Although the majority of the matched pairs were similar in job characteristics, there were fewer similarities in demographic ch aracteristics. Seventy-six of the pairs shared the same gender and 57% shared the same race, but only 43% shared both gender and race. Dependent sample t -tests were used to compare differences on the organizational antecedents between respondent s and their coworkers. There were no significant differences between the two groups on perceived equal employment opportunity, minority segmentation, diversity climate, inst rumental social support, emotional social support, gender-based tokenism or race-based tokenism. However, the difference between study participants and coworkers on di versity climate was marginally significant ( t = 1.98, p = .06).

PAGE 171

136 Chapter Four Discussion Summary of Results for Equal Employ ment Opportunity/Diversity Climate Taken together, there is support for th e overall mediating ro le of perceived discrimination in the relationship be tween perceived equal employment opportunity/diversity climate and workplace out comes. Eight of the ten models examined exhibited joint significance in the paths between EEODIV an d perceived discrimination, and between perceived discrimination and outco mes. The remaining three models lacked significance in the path between perceive d discrimination and outcome only. The directions of paths revealed that a more pos itive EEO/diversity climate was related to less perceived discrimination, which was in turn related to improved outcomes. Minority status does appear to modera te the relationship between EEODIV and perceived race-based discrimination. However, none of the intera ctions in models examining sex-based discrimination were signi ficant. Thus, there are differences between white and non-white employees on the streng th of the relationship between EEODIV and perceived discrimination. However, the indirect effects and test s of simple mediation did not support a difference between minority and majority gr oup members in the overall mediation model. In nearly every instance, both majority a nd minority group members exhibited significant

PAGE 172

137 (or non-significant) findings. It is possibl e to find support for minority status as a moderator in the relationship between the predictor and mediator, and yet fail to find group differences in indirect e ffects or simple mediation mode ls. The interaction effect is considering slope differences in one path, while indirect effects are considering the multiplicative effect of two paths. Finally, each of ten models examined exhi bited the same pattern in the direction of paths. EEODIV was negatively related to perceived workplace discrimination. Thus, as the climate improved, less unfair tr eatment was reported. Also, perceived discrimination was negatively related to j ob satisfaction, organi zational commitment, physical health and psychological health, and positively related to intent to turnover. Last, EEODIV was positively related to j ob satisfaction, organizational commitment, physical health and psychological health, and ne gatively related to inte nt to turnover. Summary of Results for Minority Segmentation Hypothesis 2 predicted that percei ved discrimination would mediate the relationship between minority segmentati on and outcomes. Minority status would moderate the relationship between minority se gmentation and perceived discrimination. It was predicted that minority individuals w ould report a positive relationship between minority segmentation and perceived discri mination, a negative relationship between minority segmentation and job satisfaction, or ganizational commitment, physical health and psychological health, and a positive rela tionship with turnover intention. Majority individuals, on the other ha nd, would report a negative relationship between minority segmentation and perceived discrimination, a positive relationship between minority

PAGE 173

138 segmentation and job satisfaction, organiza tional commitment, physical health and psychological health, and a negative re lationship with turnover intention. Taken together, the results for minority segmentation indicate that there is an overall effect of minority segmentation on outcomes through perceived discrimination. All of the models for sex discrimination and three of the five models for race discrimination demonstrated support fo r mediation, suggesting that minority segmentation is related to increased perceive d discrimination, which in turn leads to poorer outcomes. Two of the models for r ace discrimination (i.e. those testing job satisfaction and physical health as outcomes) did not fully support the model. However, in all cases all of the paths were signifi cant except one that approached significance. Thus, minority segmentation does relate to perceived discrimination across groups of employees, and differentially grouping the workforce by gender or ethnicity can negatively impact outcomes. However, ther e were differences when comparing the gender groups to the ethnic groups. The results for men and women seemed to indicate support for the overall mediation model, but there was no support for mi nority status as a moderator, the indirect effects were significant for both groups in ev ery case, and the path coefficients support mediation for both groups in nearly every case. Thus, the rela tionship of minority segmentation on outcomes through perceived di scrimination does not appear to differ based on gender. The results for white and non-white particip ants were less straightforward. In one of the five models investigated, minority status was found to be a moderator in the relationship between minority segmentation and perceived discrimination. In three of the

PAGE 174

139 remaining models, this interaction effect was marginally significant. Moreover, the indirect effects support a possible difference between white and non-white individuals. In four of the five models investigated, tests of non-white individuals exhibited a significant indirect effect, whereas tests of white indivi duals did not. However, the simple tests of mediation reveal a different scenario as bot h groups displayed support for the model in most cases. Based on these findings it is difficult to conclude that the effect of minority segmentation on outcomes through perceived di scrimination differs between white and non-white individuals. However, the pattern of results suggests there may be ethnic differences. It is possible that a larger sample of employees would better reveal the group differences. Finally, the directions of the path coeffici ents are different than expected. It was hypothesized that minority group members woul d exhibit a positive relationship between minority segmentation and perceived discrimina tion, as minorities are more likely to be treated unfairly as they are gr ouped together within an orga nization. Conversely, majority group members were expected to exhibit a negative relationship between minority segmentation and perceived discrimination, as they are likely to benefit from ethnic grouping. Next, it was predicted that minority group members would report a poorer job satisfaction, organizational commitment, physi cal health, and psychological health with increasing minority segmentation. Also, they would experience a positive relationship between minority segmentation and intention to turnover. The rationale was that minority individuals will realize they are being grouped into an area with less prestige and influence, and this will negatively impact work attitudes and health. The opposite directions were predicted for majority group members as they will realize that they are

PAGE 175

140 surrounded by others with power and influen ce, and this will positively affect work attitudes and health. However, path directions were identi cal across models and between men/women and whites/non-whites. In each case, greater minority segmentation was associated with more perceived discrimi nation. Thus, as a workplace became more segmented by group status, individuals felt mo re unfair treatment, regardless of which group, minority or majority, to which they belonged. It seems that majority group members may not receive more benefits as the workplace becomes segmented; people from all groups are at a disadvantage. Also, across models there was a positive relationship between minority segmentation an d intention to turnover, and a negative relationship with the other work attitude and health outcomes. Thus, as a workplace became more segmented, individuals reporte d less job satisfaction, organizational commitment, physical health, and psychologi cal health, and a gr eater intention to turnover. This is somewhat puzzling as one would expect a worker to experience more positive outcomes as they are surrounded by more similar others. However, it is possible that an organization which funnels minority group members into certain areas has other negative environmental characteristics which make it an undesirable place to work. Employees may not experience unfair treatment based on race or sex (because they are surrounded by similar others) but they may expe rience unfair treatment for other reasons. For example, management may unfairly treat all employees below a certain level or in certain types of positions. Or perhaps a co mpany which archaically sorts employees based on physical characteristics also has archaic technology, ancient buildings, and outdated policies. Summary of Results for Social Support

PAGE 176

141 Hypothesis 4 and 5 predicted that per ceived discrimination would mediate the relationship between social support and outco mes for minority indivi duals, but not for majority individuals. Minority individuals w ould report a negative relationship between social support and perceived discriminati on, a positive relationship between social support and job satisfaction, organizati onal commitment, physical health and psychological health, and negative relationship with intent to turnover. Majority group members, on the other hand, would report no relationship between social support and perceived discrimination. However, they would report relationships between social support and outcomes in the same dir ections and minority group members. Taken together, there was some support for the overall model. In seven of the ten models examined, perceived discrimination mediated the relationship between social support and the outcomes. In the other models one path was signifi cant while the other path approached significance. As a whole, there seems to be support that social support plays a role in affecting perceived discri mination in the workplace, across gender or ethnic groups. However, there were differe nces between men and women and between white and non-white respondents. First, there was no support for the mode rating role of minority status in the relationship between social support a nd perceived sex-based discrimination. Further, for both the indirect effects a nd simple mediation models, both men and women displayed either significant or non-signi ficant findings in most cases. Thus, there does not appear to be a difference between men and women in how discrimination mediates the relationship between social support and outcomes.

PAGE 177

142 For race-based discrimination, on the ot her hand, there was some evidence of the moderating role of minority status in th e relationship betwee n social support and perceived discrimination. The interaction eff ect was significant in models pertaining to job satisfaction and turnover intent, and marg inal in one additional model. This lends some support for ethnic differences in the relationship between social support and perceived discrimination. However, the patter n of indirect effects and tests of simple mediation were the same between white and non -white individuals in most cases (i.e. all except three), implying that there may not be group differences in the overall model. Across the social support models, the path s conformed to hypothesized directions. It was predicted that there would be a negative relations hip between social support and perceived discrimination, a negative relati onship between perceived discrimination and outcomes, and a positive relationship betw een social support and outcomes. Path directions were supported for both major ity and minority group members. Although a lack of mediation was expected for majority group members, one would still expect to see the individual paths in the hypothesized directions. Sex-based vs. Race-based Perceived Discrimination An examination of models pertaining to each of the three organizational antecedents does not reveal outstanding diffe rences between race-based and sex-based discrimination. Patterns of differences seemed to lie within the models for an individual antecedent. Specifically, the models for EE ODIV revealed differences by type of minority. The sex-based discrimination mode ls all supported the overall model where discrimination mediated th e relationship between EE ODIV and outcomes. Minority status did not moderate the relationship be tween predictor and di scrimination for these

PAGE 178

143 models. The race-based discrimination models on the hand, demonstrated that minority status moderated the relationship between EEODIV and discrimination in each of the models, and the overall model was not supporte d in each case. It seems that there are ethnic differences in how EEODI V related to discrimination, but not gender differences. Taken together, the type of di scrimination (race-based or sexbased) does not seem to be as important as the antecedents and other elements entered into the mediation models. Conclusions A total of 30 moderated mediation models were investigated in the current paper. The majority of the models demonstrated that perceived workplace discrimination mediates the relationship between workplac e conditions and outcomes. Taken together, poor workplace conditions related to greater perceived discrimination which related to negative consequences for individuals from a ll groups. While not consistent with a priori hypotheses, these findings are novel and suggest that a lack of commitment to workplace diversity adversely affects all employees, minority or not. However, patterns of moderation and indirect effect s reveal that relationships may be stronger for minority group members in some cases, suggesting that these workplace factor s adversely affect minority individuals to a greater degree. The bulk of previous research has focu sed on individual differences which relate to the perception of discrimination. The current study is novel in that it demonstrates the importance of the environment in precipit ating feelings of unf air treatment. The perception of discrimination is the logical precu rsor to discrimination claims and as such organizations may want to place importan ce on creating a fair workplace for all individuals. Emphasizing a commitment to EEO policy, strengtheni ng diversity climate,

PAGE 179

144 reducing minority segmentation, and creating opportunities for soci al support may help enhance the fairness of an organization and improve workplace attitudes and health for all employees. Limitations The main limitation of the full demonstrati on is likely the untes ted nature of many of the measures. Measures of perceived di scrimination, the perceived Equal Employment Opportunity (EEO) measure, the minority se gmentation measure, and the token status measure are relatively new and do not have exte nsive reliability and validity information. Although a pilot test was conducted to evalua te the quality of all measures and many analyses resulted in many si gnificant findings, it would be wo rthwhile to gather more data using the newer scales. The measure of minority segmentation has some specific limitations. Each item in this scale refers to “minorities” or “maj ority group members”. However, unlike other scales used in the study, it does not define wh at a minority or majority member is. Thus, we do not know specifically to whom responde nts were referring when they answered questions about these groups. It could be problematic if some respondents defined a “minority” as a woman, while other intended a “minority” to be a non-white individual, and still others referred to any group who happened to comprise a small percentage of their workplace. A layer of complexity is added since the current study compared a priori-defined minority to majority gr oup members based on their perceptions of undefined minority groups. Althoug h it is currently difficult to tell, the wording of the minority segmentation scale may have had so me influence on the difference in findings

PAGE 180

145 between men and women, and whites and non-white s. This is an issue that should be addressed in future studies. Another consideration is the number of te sts conducted. Thirty tests of moderated mediation and sixty tests of simple mediati on were conducted. Descri ptive statistics such as correlations among measur es and significance testing be tween groups on demographic characteristics were also conducted. This may raise concerns about the Type I error rate. However, the number of significant findings far exceeds 5%. The consistency in findings between similar tests and within each type of organizational an tecedent lessens fears about spurious results. It is also important to note the cros s-sectional nature of the study. Although tests of mediation are designed to demonstrate possible causal relations hips, causation can only be demonstrated through proper research design. All measures were collected at one time in a survey format. The actual work e nvironment was not measured or manipulated. Thus, although one can draw conclusions about relationships between variables in the current study, one cannot draw conclusions a bout causal relationships among variables in the current study. A follow-up study would be n eeded where different work environments are manipulated to varying levels of the or ganizational antecedents, and differences in perceived discrimination and then outcomes are compared. Finally, the nature of the sample was a limitation. There were many instances of marginal significance. In several of these cas es, there was evidence from related analyses that more power might result in significan t findings. Although the analyses in the full demonstration included over 400 participants, give n the small effect sizes of the some of the relationships, the study could have bene fited from a larger sample. Moreover, the

PAGE 181

146 bulk of the sample was drawn from the Psychology participant pool. Although every participant was employed, most participants we re also students. It is possible that there are some specific ramifications of being a work ing student, or it may also be possible that a certain type of person gravitates toward s being a working student in Psychology willing to fill out an internet survey. Perhaps people in this sample were more likely to report certain relationship due to their student status or shared personality characteristics, and findings are not representative of the working population as a whole. Questions such as these need to be answered in future st udies, which will hopefully include a broader sample. Limitations of the coworker study shoul d also be mentioned. Most importantly, the sample size was very small. Less than 10% of the participants in the full demonstration had a matched coworker. It becomes difficult to conduct analyses, let alone draw many conclusions from the results. Moreover, participants were not always similar to coworkers in terms of gender, et hnicity, or position with in the organization. Participants were asked to send the coworker survey to a similar other in terms of demographic and employment characteristics. This was not the case in many instances. Coworkers were asked to report on the envir onment, and when these individuals held a different position or perhaps were physica lly located in a different area from the coworker, perceptions may differ. Further, perceptions are likely to differ even more when the coworker is not similar in gender or ethnicity. Taken together, the results of the coworker survey should be interpreted with caution. Future Directions

PAGE 182

147 A natural follow-up to the current study would be either a laboratory or field study designed to test causal relationships between variable s discussed in this paper. While the findings of the current study dem onstrated possible directional relationships, only a controlled experiment designed to provide temporal precedence and exclusion of the influence of secondary va riables can strengthen the argu ments made in the current paper. Next, the study investigated group diffe rences two ways: comparing men to women and whites to non-whites. Every pers on appears in both analyses but some individuals shift minority status between the tw o. A natural next step is to compare true majority group members (i.e. white men) to single-minority group members (i.e. white women and non-white men) and double-mi nority group members (i.e. non-white women). It would also be interesting to br eak the non-white minority group into specific ethnic groups, such as black and Hispanic individuals. The perceptions of black individuals may very well differ from other ethnic minorities, as some groups can “hide” their ethnicity and are less likely to re ceive differential treatment based on group membership. For example, many Hispanic indivi duals have light skin and an absence of an accent, and therefore may not be gene rally perceived as an ethnic minority. Finally, the nature of perceived discrimina tion at work needs more attention in the literature. Perceptions of differential treatment can cost companies millions when they lead to a lawsuit. Yet, we understand little about what leads to pe rceived discrimination, as well as the composition of perceived discri mination itself. More st udies are needed to investigate the difference between subtle and overt discrimination. Additionally, the source of discrimination at wo rk is important. The measures of perceived discrimination

PAGE 183

148 in the current study addressed discriminati on from interpersonal sources as well as institutional sources. Interper sonal sources of discriminati on could include coworkers, supervisors, or other people at work. Institutional sources of discrimination come from policies and practices enacted by the organization as a w hole. Depending on the source, feelings of differential treatment may vary by group. This study was unable to tease apart differences among source of discrimination, but fu ture studies should pay attention to this issue.

PAGE 184

149 References Alderfer, C.P. (1987). An intergroup pers pective on group dynamics. In J.W. Lorsch (Ed.), Handbook of organizational behavi or (pp. 190-222). Englewood Cliffs, NJ: Prentice Hall. Alderfer, C.P., Alderfer, C.J., Tucker, L ., & Tucker, R.C. (1980). Diagnosing race relations in management. Journal of Applied Behavioral Science, 16, 135-166. Alexander, V., & Thoits, P. (1985). Token achievement: An examin ation of proportional representation and performance outcomes. Social Forces 64 (2), 332-340. Anderson, N., McNeilly, M., & Myers, H. (1993). A biopsychosocial model of race differences in vascular reactivity. Cardiovascular reactivity to psychological stress & disease (pp. 83-108). Washington, DC US: Amer ican Psychological Association. Babin, B., & Boles, J. (1996). The effects of perceived co-worker involvement and supervisor support on service provider role stress, performance and job satisfaction. Journal of Retailing 72 (1), 57-75. Baldi, S., & McBrier, D. B. (1997). Do the determinants of promotion differ for blacks and whites? Work and Occupations, 24 (4), 478-497.

PAGE 185

150 Baron, R.M. & Kenny, D.A. (1986). The moderator-mediator distinction in social psychological research: Con ceptual, strategic, and st atistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Beggs, J.L. (1995). The institutional environment: Implications for race and gender inequality in the U.S. Labor Market. American Sociological Review, 60, 612-633. Berscheid, E. & Hatfield, E. (1978). Interpersonal Attraction. 2nd Ed. New York: Random House. Bergman, M., & Drasgow, F. (2003). Race as a moderator in a model of sexual harassment: An empirical test. Journal of Occupational Health Psychology 8 (2), 131-145. Bowling, A. (1997). Measuring health: A review of qual ity of life measurements scales. 2nd Ed. Philadelphia: Open University Press. Branscombe, N. R., Schmitt, M. T., & Ha rvey, R. D. (1999). Perceiving pervasive discrimination among african americans: Im plications for group identification and well-being. Journal of Personality and Social Psychology, 77 (1), 135-149. Bridges, W.P. & Villemez, W.J. (1994). The employment relationship: Causes and consequences of modern personnel admi nistration. New York: Plenum.

PAGE 186

151 Brondolo, E., Rieppi, R., Kelly, K., & Geri n, W. (2003). Perceived racism and blood pressure: A review of the literature and conceptual and methodological critique. Annals of Behavioral Medicine 25 (1), 55-65. Brown, T., Sellers, S., Brown, K., & Jackson, J. (1999). Race, ethnicity, and culture in the sociology of mental health. Handbook of sociology of mental health (pp. 167182). Dordrecht Netherlands: Kluwer Academic Publishers. Burlew, A. K., & Johnson, J. L. (1992). Ro le conflict and career advancement among african american women in nontraditional professions. The Career Development Quarterly, 40 (4), 302-312. Cammann, C., Fichman, M., Jenkins, D., & Kl esh, J. (1983). Assessing the attitudes and perceptions of organizational members. In S. Seashore, E. Lawler, P. Mirvis, & C. Cammann (Eds.), Assessing organizational change: A guide to methods, measures and practices. (pp.71-138). New York: John Wiley. Caplan, R.D., Cobb, S., French, J.R.P., Van Harrison, R., & Pinneau, S.R. (1980). Job demands and worker health Ann Arbor: University of Michigan, Institute for Social Research. Chatman, J., & O'Reilly, C. (2004). Asymmetric Reactions to Work Group Sex Diversity among Men and Women. Academy of Management Journal 47 (2), 193-208.

PAGE 187

152 Chattopadhyay, P. (1999). Beyond direct and symmetrical effects: The influence of demographic dissimilarity on organi zational citizenship behaviors. Academy of Management Journal, 42, 273-287. Chrobot-Mason, D. (2003). Keeping the promise: Psychological contra ct violations for minority employees. Journal of Managerial Psychology 18 (1), 22-45. Clark, R., Anderson, N. B., Clark, V. R., & W illiams, D. R. (1999). Racism as a stressor for african americans: A biopsychosocial model. American Psychologist, 54 (10), 805-816. Collins, S.M. (1989). The marginalization of black executives. Social Problems, 36(4), 429-447. Contrada, R. J. Ashmore, R. D., Gary, M. L ., Coups, E., Egeth, J.D., Sewell, A., Ewell, K., Goyal T.M., chasse, V. (2001). M easures of ethnic ity-related stress: Psychometric properties, ethnic group di fferences, and associations with wellbeing. Journal of Applied Psychology, 31, 1775-1820. Cose, E. (1993). The rage of the privileged class New York, NY: Harper Collins. Cox, T.H. & Blake, S. (1991). Managing cultural diversity: Implications for organizational competitiveness. Academy of Management Executive, 5 45-56. Crosby, F., Bromley, S., & Saxe, L. (1980). Recent unobtrusive studies of black and white discrimination and prejudice: A literature review. Psychological Bulletin, 87 (3), 546-563.

PAGE 188

153 Davis, M., Matthews, K., & Tw amley, E. (1999). Is life mo re difficult on Mars or Venus? A meta-analytic review of sex differe nces in major and minor life events. Annals of Behavioral Medicine 21 (1), 83-97. DeNavas-Walt, C., Proctor, B.D., & Smith, J.C. (2008). U.S. Census Bureau, Current Population Reports, P60-235. Income, Poverty, and Health Insurance Coverage in the United States: 2007 U.S. Government Printin g Office: Washington, D.C. Deitch, E. A., Barsky, A., Butz, R. M., Chan, S., Brief, A. P., & Bradley, J. C. (2003). Subtle yet significant: The existence and im pact of everyday racial discrimination in the workplace. Human Relations, 56 (11), 1299-1324. Dipboye, R.L. & Colella, A. (2005). Discrimination at work: The psychological and organizational bases Mahwah, NJ: Lawrence Erlbaum Associates. Dovidio, J.F., & Gaertner, S.L. (1991). Cha nges in the expression and assessment of racial predujice. In H.J. Knopke, R. J. Norrell, & R.W. Rogers (Eds.), Opening Doors: Perspectives on race rela tions in contemporary America (pp. 119-148). Tuscaloosa: The University of Alabama Press. Dovidio, J.F., Gaertner, S.L., Kawakami, K ., Hodson, G. (2002). Why can’t we just get along? Interpersonal biases and interracial distrust. Cultural diversity and Ethnic minority psychology, 8, 88-102.

PAGE 189

154 Dreher, G., & Ash, R. (1990). A comparativ e study of mentoring among men and women in managerial, professional and technical positions. Journal of Applied Psychology 75 (5), 539-546. Dreher, G., & Chargois, J. (1998). Gender, me ntoring experiences, and salary attainment among graduates of an historically Black university. Journal of Vocational Behavior 53 (3), 401-416. Dreher, G., & Cox, T. (1996). Race, gender, and opportunity: A study of compensation attainment and the establishmen t of mentoring relationships. Journal of Applied Psychology 81 (3), 297-308. Elvira, M., & Cohen, L. (2001). Location matters: A cross-level analysis of the effects of organizational sex composition on turnover. Academy of Management Journal 44 (3), 591-605. Ely, R. (1994). The effects of organizationa l demographics and social identity on relationships among professional women. Administrative Science Quarterly, 39 203-238. Ensher, E. A., Grant-Vallone, E. J., & Dona ldson, S. I. (2001). Effects of perceived discrimination on job satisfaction, orga nizational commitment, organizational citizenship behavior, and grievances. Human Resource Development Quarterly, 12 (1), 53-72.

PAGE 190

155 Fairhurst, G., & Snavely, B. (1983). Major ity and token minority group relationships: Power acquisition and communication. Academy of Management Review 8 (2), 292300. Fairhurst, G., & Snavely, B. (1983). A test of the social isolation of male tokens. Academy of Management Journal 26 (2), 353-361. Feagin, J. R., & Sikes, M. P. (1994). Living with racism: The black middle-class experience Boston, MA, US: Beacon Press. Fidell, L. (1970). Empirical ve rification of sex discrimina tion in hiring practices in psychology. American Psychologist 25 (12), 1094-1098. Fields, D.L. (2002). Taking the measure of work: A guide to validated scales for organizational research and diagnosis Thousand Oaks, CA: Sage Publications. Floge, L., & Merrill, D. (1986). Tokenism reconsidered: Male nurses and female physicians in a hospital setting. Social Forces 64 (4), 925-947. Foley, S., & Kidder, D. L. (2002). Hispanic la w students' perceptions of discrimination, justice, and career prospects. Hispanic Journal of Behavioral Sciences, 24 (1), 23-37. Foley, S., Kidder, D.L., & Powell, G.N. ( 2002). The perceived glass ceiling and justice perceptions: An investigation of Hispanic law associates. Journal of Management, 28 471-496.

PAGE 191

156 Ford, D. L. (1985). Facets of work support a nd employee work outcomes: An exploratory analysis. Journal of Management, 11 (3), 5. Forman, T. A. (2003). The social psychologi cal costs of racial segmentation in the workplace: A study of African Americans' well-being. Journal of Health and Social Behavior, 44 (3), 332-352. Foschi, M., Lai, L., & Sigerson, K. ( 1994). Gender and double standards in the assessment of job applicants. Social Psychology Quarterly 57 (4), 326-339. Frieze, I. H., Olson, J. E., & Good, D. C. (1990). Perceived and actual discrimination in the salaries of male and female managers. Journal of Applied Social Psychology, 20 (1), 46-67. Fritz, M.S., & MacKinnon, D.P. (2007). Require d sample size to detect the mediated effect. Psychological Science, 18(3), 233-239. Gant, L. M., Nagda, B. A., Brabson, H. V., & Jayaratne, S. (1993). Effects of social support and undermining on african american workers' perceptions of coworker and supervisor relationships and psychological well-being. Social Work, 38 (2), 158-164. Gomez, J. P., & Trierweiler, S. J. (20 01). Does discrimination terminology create response bias in questionnaire studies of discrimination? Personality and Social Psychology Bulletin, 27 (5), 630-638.

PAGE 192

157 Greenhaus, J. H., & Parasuraman, S. (1993). Job performance attributions and career advancement prospects: An examina tion of gender and race effects. Organizational Behavior and Human Decision Processes, 55 (2), 273-297. Greenhaus, J. H., Parasuraman, S., & Wormley, W. M. (1990). Effects of race on organizational experiences, job perfor mance evaluations, and career outcomes. Academy of Manage ment Journal, 33 (1), 64-86. Griffeth, R.W. & Hom, P.W. (2001). Retaining valued employees Thousand Oaks, CA: Sage. Gutek, B. A., Cohen, A. G., & Tsui, A. (1996). Reactions to perceived sex discrimination. Human Relations, 49 (6), 791-813. Gutek, B.A., Larwood, L, & Stromberg, A.H. (1986). Women at work. In C. Cooper & I. Robertson (Eds.), Review of Industria l-Organizational Psychology (pp.217-234). London: Wiley. Gutman, A. (1993). EEO law and personnel practices Thousand Oaks, CA, US: Sage Publications, Inc. Hammer, R. & Green, G. (1998). Perceptions of Racial Disc rimination in Employment among African Americans: Assessing the Ef fect of Occupational Exclusion/Social Closure. Presented at the Annual Meeting of th e American Sociolog ical Association, August 21-25, San Francisco, CA.

PAGE 193

158 Hicks-Clarke, D., & Iles, P. (2000). Climate for diversity and its effects on career and organisational attit udes and perceptions. Personnel Review, 29 (3), 324-345. Hite, L.M. (2004). Black and White wo men managers: Access to opportunity. Human Resource Development Quarterly, 15(2), 131-146. Hom, P., Roberson, L., & Elli s, A. (2008). Challenging conventional wisdom about who quits: Revelations from corporate America. Journal of Applied Psychology 93 (1), 134. Hughes, D., & Dodge, M. A. (1997). African american women in the workplace: Relationships between job conditions, racial bias at work, and perceived job quality. American Journal of Community Psychology, 25 (5), 581-599. Hunter, C. D., & Lewis-Coles, M. E. (2004). Coping with racism: A spirit-based psychological perspective. In J. L. Chin (Ed.), The psychology of prejudice and discrimination: Racism in america, vol. 1. (pp. 207-222). Westport, CT, US: Praeger Publishers/Greenwood Publishing Group. Ibarra, H. (1993). Personal networks of women and minorities in management: A conceptual framework. Academy of Management Review, 18 (1), 56-87. Jackson, J., Brown, T., Williams, D., Torres, M., Sellers, S., & Brown, K. (1996). Racism and the Physical and Mental Health Status of African Americans: A Thirteen-year National Panel Study. Ethnicity and Disease, 6(1-2), 132-147.

PAGE 194

159 Jackson, P., Thoits, P., & Taylor, H. (1995). Composition of the workplace and psychological well-being: The effects of tokenism on America's Black elite. Social Forces 74 (2), 543-557. Jeanquart-Barone, S. (1996). Implications of ra cial diversity in the supervisor-subordinate relationship. Journal of Applied Social Psychology, 26 (11), 935-944. Jeanquart-Barone, S., & Sekaran, U. (1996). Institutional racism: An empirical study. Journal of Social Psychology, 136 (4), 477-482. Kanter, R.M. (1977). Men and Women of the corporation New York: Basic Books. Kaufman, R. (2001). “Race and Labor market segmentation” in Sourcebook of Labor Markets 645-668 New York, NY: Kluwer Academic. Kessler, R. C., Mickelson, K. D., & Williams, D. R. (1999). The prevalence, distribution, and mental health correlates of percie ved discrimination in the united states. Journal of Health and Social Behavior, 40 (3), 208-230. King, E. B., Madera, J. M., Hebl, M. R., Kni ght, J. L., & Mendoza, S. A. (2006). What's in a name? A multiracial investigation of the role of occupational stereotypes in selection decisions. Journal of Applied Social Psychology, 36 (5), 1145-1159. King, A.G., & Spruell, S.P. (2001). Coca-Cola takes the high road. Black Enterprise, 31(7), 29.

PAGE 195

160 Konrad, A., & Linnehan, F. (1995). Race and se x differences in line managers' reactions to equal employment opportunity and affirmative action interventions. Group & Organization Management 20 (4), 409-439. Kossek, E. E., & Zonia, S. C. (1993). Asse ssing diversity climat e: A field study of reactions to employer effo rts to promote diversity. Journal of Organizational Behavior, 14 (1), 61-81. Krimmel, J., & Gormley, P. (2003). Tokenism and Job Satisfaction for Policewomen. American Journal of Criminal Justice 28 (1), 73-88. Landau, J. (1995). The relationship of race and gender to managers’ ratings of promotion potential. Journal of Organizational Behavior, 16 (4), 391-400. Landrine, H., & Klonoff, E. A. (1996). The schedu le of racist events: A measure of racial discrimination and a study of its negative physical and mental health consequences. Journal of Black Psychology, 22 (2), 144-168. Lazarus, R.S. & Folkman, S. (1984). Stress, appraisal, and coping New York: Springer. Lee. R.T., & Ashforth, B.E. (1993). A fu rther examination of managerial burnout: Toward an integrated model. Journal of Organizational Behavior, 14, 3-20. Levitin, T., Quinn, R. P., & Staines, G. L. (1971). Sex discrimination against the american working woman. American Behavioral Scientist, 15 (2), 237-254.

PAGE 196

161 Lim, V.K.G. (1996). Job insecu rity and its outcomes: Modera ting effects of work-based and non-work-based social support. Human Relations, 49(2), 171. Lincoln, J.R. & Miller, J. (1979). Work a nd friendship ties in organizations: A comparative analysis of relational networks. Administrative Science Quarterly, 14 61-81. MacKinnon, D.P., Krull, J.L., & Lockwood, C. M. (2000). Equivalence of the mediation, confounding, and suppression effects. Prevention Science, 1, 173-181. MacKinnon, D.P., Lockwood, C.M., Hoffman, J. M., West, S.G., & Sheets, V. (2002). A comparison of methods to test mediation and other intervening variable effects. Psychological Methods, 7, 83-104. Maume, D. J. J. (1999). Glass ceilings and gl ass escalators: Occupa tional segregation and race and sex differences in managerial promotions. Work and Occupations, 26 (4), 483-509. McConahay, J.G. (1983). Modern racism and mo dern discrimination: The effects of race, racial attitudes, and context on simulated hiring decisions. Personality and Social Psychology Bulletin, 9, 551-558. McFarlin, D.B. & Rice, R.W. (1992). The role of facet importance as a moderator in job satisfaction processes, Journal of Organizational Behavior, 13 41-54.

PAGE 197

162 McKay, P.F. & Avery, D.R. (2005). Warning! Diversity recruitment could backfire. Journal of Management Inquiry, 14 330-336. McKay, P.F., Avery, D.R., Tonidandel, S., Morris, M.A., Hernandez, M., Hebl, M.R. (2007). Racial differences in employee rete ntion: Are diversity climate perceptions the key? Personnel Psychology, 60, 35-62. McLain, D.L. (1995). Responses to health a nd safety risk in the work environment. Academy of Manageme nt Journal, 38(6) 1726-1746. McPherson, M., Smith-Lovin, L., & Cook, J. ( 2001). Birds of a feather: Homophily in social networks. Annual Review of Sociology 27 415-444. Medical Outcomes Trust. (1993). How to score the SF-36 Health Survey Boston, MA: Medical Outcomes Trust. Miller, J. (1986). Pathways in the workplace. Cambridge, UK: Cambridge University Press. Mor Barak, M.E., Cherin, D.A., & Berkma n, S. (1998). Organizational and personal dimensions in diversity climate: Ethn ic and gender differences in employee perceptions. Journal of Applied Behavioral Science, 34 82-104.

PAGE 198

163 Mor Barak, M., & Levin, A. (2002). Outside of the corporate mainstream and excluded from the work community: A study of di versity, job satisfaction and well-being. Community, Work & Family, 5 (2), 133-157. Morrison, A. M., & von Glinow, M. A. ( 1990). Women and minorities in management. American Psychologist, 45 (2), 200-208. Mowday, R.T., Steers, R.M., & Porter, L.W. (1979). The measurement of organizational commitment. Journal of Vocational Behavior, 14, 224-247. Mueller, C. W., Parcel, T. L., & Tanaka, K. (1989). Particularism in authority outcomes of black and white supervisors. Social Science Research, 18 (1), 1-20. National Center for Health Statistics (2007). Health, United States 2007, With Chartbook on Trends in the Health of Americans Hyattsville, MD. Neckerman, K. M., & Kirschenman, J. (1991). Hiring strategies, racial bias, and innercity workers. Social Problems, 38 (4), 433-447. Neimann, Y.F. & Dovidio, J.F. (1998). Relati onship of solo status, academic rank, and perceived distinctiveness on job satisfaction of racial/ethnic minorities. Journal of Applied Psychology, 83(1) 55-71. Nixon, R. (1985). Black managers in corporate Americ a: Alienation or integration Washington, DC: National Urban League.

PAGE 199

164 Nunez-Smith, M., Pilgrim, N., Wynia, M., Desai, M.M., Bright, C., Krumholtz, H., & Bradley, E.H. (2009). Healthcare workplac e discrimination and physician turnover. Journal of the National Me dical Association, 101(12) 1274-1282. O'Leary, V. E., & Ickovics, J. R. (1992) Cracking the glass ceiling: Overcoming isolation and alienation. In U. Sekaran, & F. T. L. Leong (Eds.), Womanpower: Managing in times of demographic turbulence. (pp. 7-30). Thousand Oaks, CA, US: Sage Publications, Inc. Operario, D., & Fiske, S. T. (2001). Ethnic id entity moderates perceptions of prejudice: Judgments of personal versus group discrimi nation and subtle versus blatant bias. Personality and Social Psychology Bulletin, 27 (5), 550-561. Parker, C., Baltes, B., & Christiansen, N. (1997). Support for affirmative action, justice perceptions, and work attitudes: A st udy of gender and racial-ethnic group differences. Journal of Applied Psychology 82 (3), 376-389. Pavalko, E. K., Mossakowski, K. N., & Hamilton, V. J. (2003). Does perceived discrimination affect health ? longitudinal relationships between work discrimination and women's physical and emotional health. Journal of Health and Social Behavior, 44 (1), 18-33.

PAGE 200

165 Pearson, C.A.L. (1991). An assessment of ex trinsic feedback on participation, role perceptions, motivation, and job satisfaction in a se lf-managed system for monitoring group achievement. Human Relations, 44(5), 517-537. Perry, E. L., Hendricks, W., & Broadbent, E. (2000). An exploration of access and treatment discrimination and job satisf action among college graduates with and without physical disabilities. Human Relations, 53 (7), 923-955. Pfeffer, J. (1989). A political perspectiv e on careers: Interests, networks, and environments. In M.B. Arthur, D. T. Hall, & B.S. Lawrence (Eds.), Handbook of career theory (pp. 380-396). Cambridge, UK: Ca mbridge University Press. Powell, G. N., Butterfield, D. A., & Pare nt, J. D. (2002). Gender and managerial stereotypes: Have the times changed? Journal of Management, 28 (2), 177-193. Preacher, K.J., & Hayes, A.F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, and Computers, 36, 717-731. Preacher, K.J., Rucker, D.D., & Hayes, A. F. (2007). Addressing Moderated Mediation Hypotheses: Theory, Methods, and Prescriptions. Multivariate Behavioral Research, 42(1), 185-227.

PAGE 201

166 Prelow, H. M., Mosher, C. E., & Bowman, M. A. (2006). Perceived racial discrimination, social support, and psychological adjust ment among african american college students. Journal of Black Psychology, 32 (4), 442-454. Repeti, R.L. & Cosmas, K.A. (1991). The quality of the social environment at work and job satisfaction Journal of Applied So cial Psychology, 21(10), 840-854. Ridgeway, C.L. (1988). Gender differences in task groups: A status and legitimacy account. In M. Webster & M. Foschi (Eds.) Status generalization: New Theory and research (pp.188-206). Stanford, CA: Stanford University Press. Rodin, M. J., Price, J. M., Bryson, J. B., & Sa nchez, F. J. (1990). Asymmetry in prejudice attribution. Journal of Experimental Social Psychology, 26 (6), 481-504. Rosenberg, J., Perlstadt, H., & Phillips, W. (1993). Now that we are here: Discrimination, disparagement, and harassment at work and the experience of women lawyers. Gender & Society 7 (3), 415-433. Roth, L.M. (2004). The social psychology of tokenism: Status and homophily processes on Wall Street. Sociological Perspectives, 47(2), 189-214. Sanchez, J. I., & Brock, P. (1996). Outcomes of perceived discrimination among hispanic employees: Is diversity management a luxury or a necessity? Academy of Management Journal, 39 (3), 704-719.

PAGE 202

167 Schmitt, M. T., Branscombe, N. R., Kobrynow icz, D., & Owen, S. (2002). Perceiving discrimination against one's gender group has different implications for well-being in women and men. Personality and Social Psychology Bulletin, 28 (2), 197-210. Seigall, M., & McDonald, T. (1995). Focus of attention and employ ee reactions to job change. Journal of Applied Social Psychology, 25(13), 1121-1141. Shafer, R. (1993). What minority journalists identify as constraints to full newsroom equality. Howard Journal of Communications 4 (3), 195-208. Shaffer, M. A., Joplin, J. R. W., Bell, M. P., Lau, T., & Oguz, C. (2000). Gender discrimination and job-related outcomes: A cross-cultural comparison of working women in the united states and china. Journal of Voca tional Behavior, 57 (3), 395427. Shrout, P.E., & Bolger, N. (2002). Mediati on in experimental and non-experimental studies: New procedures and recommendations. Psychological Methods, 7, 422-445. Smith-Lovin, L. & McPherson, J.M. (1993). “You are who you know: A network approach to gender”. Pp. 223-254. in Theory on Gender/Feminism on Theory edited by P. England, Hawthorne, NY, Aldine de Gruyter. Snow, D., & Kline, M. (1995). Preventive in terventions in the workplace to reduce negative psychiatric consequences of work and family stress. Does stress cause

PAGE 203

168 psychiatric illness? (pp. 221-270). Washington, DC US: American Psychiatric Association. Stauffer, J. M., & Buckley, M. R. (2005). The existence and nature of racial bias in supervisory ratings. Journal of Applied Psychology, 90 (3), 586-591. Steinberg, S. (1995). Turning back: The retreat from ra cial justice in American thought and policy Boston, MA: Beacon Press. Stewart, L. D., & Perlow, R. (2001). Applican t race, job status, and racial attitude as predictors of employment discrimination. Journal of Business and Psychology, 16 (2), 259-275. Stroman, C. A., & Seltzer, R. (1991). Racial differences in coping with job stress: A research note. Journal of Social Behavior & Personality, 6 (7), 309-318. Tomaskovic-Devey, D. (1993). Gender and racial inequality at work: The sources and consequences of job segregation Ithaca: NY, ILR Press. Tomkiewicz, J., Brenner, O. C., & AdeyemiBello, T. (1998). The impact of perceptions and stereotypes on the managerial mobility of african americans. Journal of Social Psychology, 138 (1), 88-92. Tsui, A., Egan, T., & O'Reilly, C. (1992). Being different: Relational demography and organizational attachment. Administrative Science Quarterly 37 (4), 549-579.

PAGE 204

169 Turner, C.B. & Turner, B.F. (1981). Racial discrimination in occ upations: Perceived and actual. Phylon, 42(4), 322-334. Twenge, J.M. & Crocker, J. (2002). Race and self-esteem: Meta-analyses comparing whites, blacks, Hispanics, Asians, and American Indians, and comment on GrayLittle and Hafdahl (2000). Psychological Bulletin, 128(3), 371-408. Valentine, S., Silver, L., & Twigg, N. (1999). Locus of control, job satisfaction and job complexity: The role of pe rceived race discrimination. Psychological Reports, 84 (3), 1267-1273. Vandenberg, R.J., & Lance, C.E. (1992). Examin ing the causal order of job satisfaction and organizational commitment. Journal of Management, 18(1), 153-167. Vaugn-Cooke, D. (1983). Blacks in labor markets: A historical assessment. Urban League Review, 7(2) 8-18. Weick, K.E. (1995). Sensemaking in organizations Thousand Oaks, CA: Sage. Williams, C. (1992). The glass escalator: Hidd en advantages for men in the ‘female’ professions. Social Problems, 39(3), 253-267. Williams, D., & Williams-Morris, R. (2000). Racism and mental health: The African American experience. Ethnicity & Health 5 (3), 243-268.

PAGE 205

170 Williams, D. R., Yu, Y., Jackson, J., & Ande rson, N. (1997). Racial differences in physical and mental health: Socioecono mic status, stress, and discrimination. Journal of Health Psychology, 2, 335–351. Yoder, J.D. (1991). Opening and Cl osing Doors to Women's Careers. PsycCRITIQUES 36 (6), 521-522. Yoder, J.D., Adams, J., Grove, S., Priest, R. F. (1985). To teach is to learn: Overcoming tokenism with mentors. Psychology of women quarterly, 8(1), 178-192. Yoder, J.D., & McDonald, T. (1998). Measuri ng sexist discrimination in the workplace: Support for the validity of the Schedule of Sexist Events. Psychology of Women Quarterly 22 (3), 487-491. Young, J.L. & James, E.H. (2001). Token majori ty: The work attitudes of male flight attendants. Sex Roles, 45 (5/6), 299-319. Zweigenhaft, R. L., & Domhoff, G. W. (1991). Blacks in the white establishment? A study of race and class in america New Haven, CT, US: Yale University Press. Zweigenhaft, R. L., & Domhoff, G. W. (1998). Diversity in the power elite: Have women and minorities reached the top? New Haven, CT, US: Yale University Press. Zweigenhaft, R. L., & Domhoff, G. W. (2003). Blacks in the white elite: Will the progress continue? Lanham, MD, US: Rowman & Littlefield.

PAGE 206

171 Bibliography Alderfer, C. P., & Smith, K. K. (1982). Studying intergroup relations embedded in organizations. Administrative Science Quarterly, 27 (1), 35-65. Barak, M. E. M., Cherin, D. A., & Berk man, S. (1998). Organizational and personal dimensions in diversity climate: Ethn ic and gender differences in employee perceptions. Journal of Applied Behavioral Science, 34 (1), 82-104. Bigham, J.C. (1974). Views of black and wh ite children concerning the distribution of personality characteristics. Journal of Personality, 42 144-158. Braddock, J.H. & McPartland, J.M. (1987). How minorities continue to be excluded from equal employment opportunities: Research on labor market and institutional barriers. Journal of Social Issues, 43(1), 5-39. Broverman, I.K., Vogel, S.R., Broverman, D.M., Clarkson, F.E. & Rosenkrantz, P.S. (1972). Sex stereotypes: A current appraisal. Journal of Social Issues, 28 59-79. Catalyst. (1993). Successful initiatives for breaking th e glass ceiling to upward mobility for minorities and women New York: Author. Cleveland, J. N., Stockdale, M., & Murphy, K. R. (2000). Women and men in organizations: Sex and gender issues at work Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers.

PAGE 207

172 Collins, S.M. (1997). Black corporate executives: The making and breaking of the black middle class. Philadelphia, PA: University of Pittsburgh Press. Cox, T., Jr. (1994). Cultural diversity in organizations : Theory, research, and practice San Francisco: Berrett-Koehler. Folger, R., & Cropanzano, R. (1998). Organizational justice and human resource management Thousand Oaks, CA, US: Sage Publications, Inc. Harris, M. M., Lievens, F., & Van Hoye, G. (2004). 'I think they discriminated against me': Using prototype theory and organi zational justice theo ry for understanding perceived discrimination in sel ection and promotion situations. International Journal of Selection and Assessment, 12 (1), 54-65. Heilbrun, A.B., Jr. (1976). Measurement of masculine and feminine sex role identification as independent dimensions. Journal of Consulting and Clinical Psychology, 44, 183-190. Heilman, M. E. (1980). The impact of s ituational factors on personnel decisions concerning women: Varying the sex composition of the applicant pool. Organizational Behavior & Human Performance, 26 (3), 386-395. Heilman, M. E., Battle, W. S., Keller, C. E., & Lee, R. A. (1998). Type of affirmative action policy: A determinant of reactions to sex-based preferential selection? Journal of Applied Psychology, 83 (2), 190-205.

PAGE 208

173 Heilman, M. E., & Blader, S. L. (2001). Assuming preferential selection when the admissions policy is unknown: Th e effects of gender rarity. Journal of Applied Psychology, 86 (2), 188-193. Heilman, M. E., & Welle, B. (2006). Disadvantag ed by diversity? the e ffects of diversity goals on competence perceptions. Journal of Applied Social Psychology, 36 (5), 1291-1319. Jackson, B. M. (1994). African-American women in the workplace: A personal perspective from AfricanAmerican female EAPs. Employee Assistance Quarterly, 9 (3), 11-19. Kahn, R. L., & Byosiere, P. (1992). Stress in or ganizations. In M. D. Dunnette, & L. M. Hough (Eds.), Handbook of industrial and organizational psychology, vol. 3 (2nd ed.). (pp. 571-650). Palo Alto, CA, US: Consulting Psychologists Press. King, K. R. (2005). Why is discrimination st ressful? the mediating role of cognitive appraisal. Cultural Diversity and Ethnic Minority Psychology, 11 (3), 202-212. Klonoff, E., Landrine, H., & Campbell, R. (2000). Sexist discrimination may account for well-known gender differences in psychiatric symptoms. Psychology of Women Quarterly 24 (1), 93-99.

PAGE 209

174 Kravitz, D. A., & Platania, J. (1993). Attitudes and beliefs about affirmative action: Effects of target and of respondent sex and ethnicity. Journal of Applied Psychology, 78 (6), 928-938. Major, B., Quinton, W.J. & McCoy, S.K. (2002). Antecedents and consequences of attributions to discrimination: Theoretic al and empirical advances. In Zanna, Mark P. (Ed). Advances in experimental and social psychology Vol 34. Academic Press; San Diego, CA. Mor Barak, M. E., Findler, L., & Wind, L. H. (2003). Cross-cultural as pects of diversity and well-being in the workplace: An international perspective. Journal of Social Work Research and Evaluation, 4 (2), 145-169. Northcraft, G.B. & Martin, J. (1982). Double jeopardy: Resistance to affirmative action from potential beneficiaries. In B. Gutek (Ed.), Sex role stereotyping and affirmative action policy Los Angeles: Institute of Industrial Relations, Univ ersity of California. Parker, C. P., Dipboye, R. L., & Jackson, S. L. (1995). Perceptions of organizational politics: An investigation of antecedents and consequences. Journal of Management, 21 (5), 891-912. Pines, A., & Aronson, E. (1988). Career burnout: Causes and cures New York, NY, US: Free Press.

PAGE 210

175 Sackett, P. R., DuBois, C. L., & Noe, A. W. (1991). Tokenism in performance evaluation: The effects of work group representation on male-female and whiteblack differences in performance ratings. Journal of Applied Psychology, 76 (2), 263267. Samuels, F. (1973). Group Images New Haven, CT: College and University Press. Snchez, J. I., & Fernndez, D. M. (1993) Acculturative stress among hispanics: A bidimensional model of ethnic identification. Journal of Applied Social Psychology, 23 (8), 654-668. Sweeney, P. D., & McFarlin, D. B. (1997). Process and outcome: Gender differences in the assessment of justice. Journal of Organizational Behavior, 18 (1), 83-98. Taylor, S. E., Fiske, S. T., Etcoff, N. L ., & Ruderman, A. J. (1978). Categorical and contextual bases of pers on memory and stereotyping. Journal of Personality and Social Psychology, 36 (7), 778-793. Triandis, H. C., Kurowski, L. L., Tecktiel, A., & Chan, D. K. (1993). Extracting the emics of diversity. International Journal of Intercultural Relations, 17 (2), 217-234. U.S. Department of Labor, Office of the Federal Contract Compliance Programs (1995). OFCCP glass ceiling initiative: Ar e there cracks in the ceiling? Washington, DC: Office of the Federal Contract Compliance Programs, U.S. Department of Labor.

PAGE 211

176 Williams, C. (1995). Still a man’s world: Men who do” women’s work” Berkley: University of California Press. Wright, R., King, S. W., & Berg, W. E. (1985) Job satisfaction in the workplace: A study of black females in management positions. Journal of Social Service Research, 8 (3), 65-79.

PAGE 212

177 Appendix A Measures

PAGE 213

178 Demographics Please choose your gender: Male Female What is your age (in years)? _______________ Are you Hispanic/Latino? YES NO What race do you most identify with? ____White/Non-Hispanic ____African American ____Asian/Pacific Islander ____American Indian ____ Multiracial ____Other____________________ What race do others see you as? ____White/Non-Hispanic ____African American ____Asian/Pacific Islander ____American Indian ____ Multiracial ____Other____________________ What race is your mother? ____White/Non-Hispanic ____African American

PAGE 214

179 ____Asian/Pacific Islander ____American Indian ____ Multiracial ____Other____________________ What race/ethnicity is your father? ____White/Non-Hispanic ____African American ____Asian/Pacific Islander ____American Indian ____ Multiracial ____Other____________________ What is your highest level of education? ______ Some high school ______ High school diploma or GED ______ Some college ______ Technical diploma/Associate’s degree ______ Bachelor’s degree ______ Graduate degree How many hours per week do you work? _____________ At your primary job, how long have you been working there? _________ Less than 1 year _________ At least 1 year

PAGE 215

180 _________ At least 5 years _________ At least 10 years _________ 15 + years Please check the industry th at your job falls into: ________Business/ Financial (including real estate or insurance) ________Education ________Healthcare ________Culture, Arts or Recreation ________Service (including sales and law enforcement) ________Agribusiness ________Manufacturing Production Trans portation or Construction ________Computer Science or Information Technology ________Other Support for Equal Employment Opportunity 1. My organization has an Equal Empl oyment Opportunity (EEO) policy. 2. My organization has a strong Equal Em ployment Opportunity (EEO) policy. 3. My organization has a visi ble Equal Employment O pportunity (EEO) policy. 4. Senior managers emphasize Equal Employment Opportunity. 5. When organizational decision are made (e.g. hiring, promotions), they are typically identity-blind. Perceived Minority Segmentation 1. At my organization, minorities te nd to get certain types of jobs. 2. My job is one that tends to be given to minorities. 3. At my organization, minorities tend to be assigned to certain areas/departments. Token Status (pilot study version) 1. Approximately how many people are empl oyed at the physical location where you are employed? 2. Approximately how many people work within your work group/department? 3. Approximately how many people shari ng your gender are employed at the physical location of your job? (Percentages are ok)

PAGE 216

181 4. Approximately how many people sharing your race/ethnicity are employed at the physical location of your job? (Percentages are ok) 5. Approximately how many people sharing your gender are employed within your work group/department? (Percentages are ok) Approximately how many people sharing your race/ethnicity are employed within your work group/department? (Percentages are ok) Token Status (full demonstration version) Think about all of the people who work at the same physical location (e.g. business, office building) that you do. You may or may not work directly with these people on a regular basis, but they work at that same location. Answer the next two questions in regards to these people. 1. Using your best guess, what percentage of people employed at the physical location of your job is the same gender as you? 2. Using your best guess, what percentage of people employed at the physical location of your job is the sa me race or ethnicity as you? Think about all of the people who work w ithin your same work group. This includes people you work with directly on a regular basis (e.g. coworkers and supervisors). Answer the next two questions in regards to these people. 3. Using you best guess, what percenta ge of people employed within your immediate work group is the same gender as you? 4. Using your best guess, what percenta ge of people employed within your immediate work group is the same race/ethnicity as you? a. Less than 15% b. At least 15% c. At least 25% d. At least 50% e. At least 75% f. I don’t know/Not applicable

PAGE 217

182 Diversity Climate Scale 1. My company recruits from diverse sources. 2. My company offers equal access to training. 3. My company promotes open communication on diversity. 4. My company publicizes diversity principles. 5. My company offers training to manage a diverse population. 6. My company respects the pe rspectives of people like me. 7. My company maintains a divers ity-friendly work environment. 8. My workgroup has a climate th at values diverse perspectives. 9. Top leaders visibly commit to diversity. Social Support Scale 1. My immediate supervisor goes out of hi s/her way to do things to make my work life easier for me. 2. Other people at work go out of their way to make my work life easier for me. 3. It is easy to talk to my immediate supervisor. 4. It is easy to talk to other people at work. 5. My immediate supervisor can be reli ed on when things get tough at work. 6. Other people at work can be relied on when things get tough at work. 7. My immediate supervisor is willing to listen to my personal problems. 8. Other people at work are willing to listen to my personal problems. 9. I have access to a mentor at my workplace. 10. I have access to informal social networks at my workplace. 11. I have access to informal information networks at my workplace. Perceived discrimination at work 1. I have been treated unfairly by empl oyers, bosses, or supervisors because of my race/ethnicity. 2. I have been treated unfairly by empl oyers, bosses, or s upervisors because of my gender. 3. I have been treated unfairly by co workers or colleagues because of my race/ethnicity. 4. I have been treated unfairly by co workers or colleagues because of my gender. 5. My supervisor sometimes makes racist decisions. 6. My supervisor sometimes makes sexist decisions. 7. My coworkers sometimes make racist statements. 8. My coworkers sometimes make sexist statements. 9. I feel that some of the policies and practices of this organization are racist. 10. I feel that some of the policies and practices of this organization are sexist. 11. At work, I sometimes feel that my race/ethnicity is a limitation.

PAGE 218

183 12. At work, I sometimes feel that my gender is a limitation. 13. At work, I do not get enough recogn ition because of my race/ethnicity. 14. At work, I do not get enough recognition because of my gender. 15. At work, I sometimes feel that people actively try to stop me from advancing because of my race/ethnicity. 16. At work, I sometimes feel that people actively try to stop me from advancing because of my gender. 17. At work, I feel that others exclude me from their activit ies because of my race/ethnicity. 18. At work, I feel that others exclude me from their activit ies because of my gender. Every-Day Perceived Discrimination 1. I have been treated unfairly by teach ers and professors because of my race/ethnicity. 2. I have been treated unfairl y by people in service jobs (store clerks, waiters, bartenders, and others) because of my race/ethnicity. 3. I have been treated unfairly by stra ngers because of my race/ethnicity. 4. I have been treated unfairly by people in helping jobs (doctors, nurses, school counselors, therapists, and others) because of my ethnicity. 5. I have been treated unfairly by neigh bors because of my race/ethnicity. 6. I have been treated unfairly by instit utions (schools, universities, the police, the courts, and others) because of my race/ethnicity. 7. I have been treated unfairly by people that I thought were my friends because of my race/ethnicity. 8. I have been accused or suspected of doing something wrong (such as stealing, cheating, not doing your share of the work, or breaking the law) because of my race/ethnicity. 9. People have misunderstood my intent ions and motives because of my race/ethnicity. 10. There were times when I wanted to tell off someone for being racist against my racial/ethnic gr oup but I didn’t say anything. 11. There have been times when I have been really angry about something racist that was done to me. 12. There have been times when I was forced to take drastic steps (such as filing a grievance, filing a lawsuit, quitting my job, moving away, and other actions) to deal with some racist thing that was done to me. 13. II have been called racist names. 14. I have gotten into an argument or a fight about somethi ng racist that was done to me or somebody else. 15. I have been made fun of, picked on, pushed, shoved, hit, or threatened with harm because of my race or ethnicity. 16. I have been treated unfairly by teach ers and professors because of my gender.

PAGE 219

184 17. I have been treated unfairl y by people in service jobs (store clerks, waiters, bartenders, and others) because of my gender. 18. I have been treated unfairly by st rangers because of my gender. 19. I have been treated unfairly by people in helping jobs (doctors, nurses, school counselors, therapists, and others) because of my gender. 20. I have been treated unfairly by ne ighbors because of my gender. 21. I have been treated unfairly by instit utions (schools, universities, the police, the courts, and others) because of my gender. 22. I have been treated unfairly by people that I thought were my friends because of my gender. 23. I have been accused or suspected of doing something wrong (such as stealing, cheating, not doing your share of the work, or breaking the law) because of my gender. 24. People have misunderstood my intent ions and motives because of my gender. 25. There were times when I wanted to tell off someone for being sexist against my gender but I didn’t say anything. 26. There have been times when I have been really angry about something sexist that was done to me. 27. There have been times when I was forced to take drastic steps (such as filing a grievance, filing a lawsuit, quitting my job, moving away, and other actions) to deal with some sexist thing that was done to me. 28. II have been called sexist names. 29. I have gotten into an argument or a fight about somethi ng sexist that was done to me or somebody else. 30. I have been made fun of, picked on, pushed, shoved, hit, or threatened with harm because of my gender. SF-36 1. In general, would you say your health is: 1 Excellent 2 Very Good 3 Good 4 Fair 5 Poor 2. Compared to one year ago, how wo uld you rate your health in general now? 1 Much better now than one year ago 2 Somewhat better not than one year ago 3 About the same

PAGE 220

185 4 Somewhat worse now than one year ago 5 Much worse now than one year ago The following items are about activities you might do during a typical day. Does your health now limit you in these activities ? If so, how much? 3. Vigorous activities, such as running li fting heavy objects, participating in strenuous sports. 4. Moderate activities, such as movi ng a table, pushing a vacuum cleaner, bowling, or playing golf. 5. Lifting or carrying groceries. 6. Climbing several flights of stairs 7. Claiming one flight of stairs. 8. Bending, kneeling, or stooping. 9. Walking more than one mile. 10. Walking several blocks 11. Walking one block. 12. Bathing or dressing yourself 1 Yes, limited a lot 2 Yes, limited a little 3 No, not at all limited During the past 4 weeks have you had any of the following problems with your work or other regular daily activities as a result of your physical health ? 13. Cut down the amount of time you spen t on work or other activities. 14. Accomplished less than you would like 15. Were limited in the kind of work or other activities. 16. Had difficulty performing the work or other activities (for example, it took extra effort) 1 Yes 2 No During the past 4 weeks, have you had any of the following problems with your work or other regular daily activities as a result of any emotional problems (such as feeling depr essed or anxious)? 17. Cut down the amount of time you spen t on work or other activities. 18. Accomplished less than you would like 19. Didn’t do work or other activities as carefully as usual 1 Yes 2 No 20. During the past 4 weeks, to what extent has your physical health or emotional problems interfered with your normal social activities with family, friends, neighbors, or groups?

PAGE 221

186 1 Not at all 2 Slightly 3 Moderately 4 Quite a bit 5 Extremely 21. How much bodily pain have you had during the past 4 weeks ? 1 None 2 Very mild 3 Mild 4 Moderate 5 Severe 6 Very severe 22. During the past 4 weeks, how much did pain interfere with your normal work (including both work outside the home and housework)? 1 Not at all 2 A little bit 3 Moderately 4 Quite a bit 5 Extremely These questions are about how you feel and how things have been with you during the past 4 weeks For each question, please give the one answer that comes closest to the way you have been feeling. 1 All of the time 2 Most of the time 3 A good bit of the time 4 Some of the time 5 A little of the time 6 None of the time How much of the time during the past 4 weeks . 23. Did you feel full of pep? 24. Have you been a very nervous person? 25. Have you felt so down in the dumps that nothing could cheer you up? 26. Have you felt calm and peaceful? 27. Did you have a lot of energy? 28. Have you felt downhearted and blue? 29. Did you feel worn out? 30. Have you been a happy person? 31. Did you feel tired?

PAGE 222

187 32. During the past 4 weeks, how much of the time has your physical health or emotional problems interfered with your social activities (like visiting with friends, relatives, etc.)? 1 All of the time 2 Most of the time 3 Some of the time 4 A little of the time 5 None of the time How TRUE or FALSE is each of the following statements for you. 1 Definitely True 2 Mostly True 3 Don’t know 4 Mostly False 5 Definitely False 33. I seem to get sick a litt le easier than other people 34. I am as healthy as anybody I know 35. I expect my health to get worse 36. My health is excellent. Job satisfaction 1. All in all, I am satisfied with my job. 2. In general, I don’t like my job. 3. In general, I like working here. Organizational Commitment Scale 1. I am willing to put a great deal of ef fort beyond that normally expected in order to help this orga nization be successful. 2. I talk up this organization to my frie nds as a great organization to work for. 3. I would accept almost any types of job assignment in order to keep working for this organization.

PAGE 223

188 4. I find that my values and the organization’s values are very similar 5. I am proud to tell others that I am part of this organization. 6. This organization really inspires the very best in me in the way of job performance. 7. I am extremely glad that I chose this organization to work for over others I was considering at the time I joined. 8. I really care about the fa te of this organization. For me, this is the best of all possible organizations for which to work.

PAGE 224

189 Appendix B Original Version of Full Demonstration Analyses

PAGE 225

190 Results Results for Perceived Equal Employment Opportunity Subtle Sex-based Discrimination Perceived Equal Employment Opportunity (EEO) in the workplace was investigated as a predictor in a model where subtle sexbased discrimination mediates the relati onship between EEO and five dependent variables (job satisfaction, or ganizational commitment, intention to turnover, physical health, and psychological health). Minority status was investig ated as a moderator in the relationship between the predictor and the mediator. Additionally, separate mediation models were tested for each of the co mparison groups (men, women, whites, and nonwhites). First, a model was tested where s ubtle sex discrimination mediates the relationship between EEO and job satisfaction. The results of the moderated mediation demonstrated overall support for mediation as the path from EEO to subtle sex discrimination was significant ( B = -.35, t = -2.99, p < .01) in addition to the path from the mediator to job satisfaction ( B = -.11, t = -3.69, p < .01). Perceived EEO was related to less perceived discrimination, which in turn was related to greater job satisfaction. No significant interaction was found between the predictor (EEO) and the moderator (minority status). Thus, minority status doe s not moderate the relationship between EEO and perceived subtle sex discrimination. Th e indirect effects support the finding of overall mediation for both men and women, as the indirect effects for both groups were significant. Results for the moderated mediation can be found in Table 93

PAGE 226

191 Table 93. Results of Moderated Mediation (PREDIC TOR = EEO, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Gender Predictor B SE t p Subtle sex Constant 16.591.809.22 .00 EEO -.35.12-2.99 .00 Minority Status .572.09.27 .79 EEO Minority Status .04.14.26 .80 Job satisfaction Constant 10.051.228.21 .00 Subtle sex -.11.03-3.69 .00 EEO .13.071.86 .06 Minority Status .421.29.32 .75 EEO Minority Status .00.08.01 .99 Minority Status Indirect Effect SE z p Men .04.022.27 .02 Women .04.012.75 .01 The separate tests of mediation for bot h men and women support the finding that perceived subtle sex discrimination mediat ed the relationship between EEO and job satisfaction, and that there doe s not appear to be a differe nce for minority or majority group members. Also, the paths were in the hypothesized directions The results for the simple mediation tests for men and women are displayed in Table 94.

PAGE 227

192 Table 94. Results of Simple Mediation for Men and Women (PREDICTOR = EEO, Med = Subtle Discrimination, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .17.082.30.02.17.04 3.82 .00 b(MX) -.35.11-3.23.00-.31.07 -4.15 .00 b(YM.X) -.19.06-3.02.00-.09.04 -2.48 .01 b(YX.M) .11.081.44.15.14.05 3.14 .00 Next, a model was tested where the dependent variable was organizational commitment. Again, support for overall mediation was found as the path from EEO to subtle sex discrimination ( B = -.32, t = -2.80, p < .05) and the path from the mediator to organization commitment ( B = -.24, t = -2.91, p < .01) was significant. Perceived EEO was related to less perceived discriminati on, which in turn was related to greater organizational commitment. However, no signi ficant interaction was found between EEO and minority status. When examining the indire ct effects, women exhibited a significant indirect effect of EEO on organizationa l commitment through perceived discrimination (Indirect effect = .07, z = 2.37, p < .05), while men did not. However, the men’s’ results are approaching significance. Results of the moderated mediation are displayed in Table 95.

PAGE 228

193 Table 95. Results of Moderated Mediation (PREDIC TOR = EEO, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 15.991.78 8.99.00 EEO -.32.11 -2.80.01 Minority Status 1.192.07 .58.57 EEO Minority Status .01.13 .06.96 Organizational Constant 25.153.20 7.85.00 Subtle sex discrimination -.24.08 -2.91.00 EEO .45.19 2.38.02 Minority Status -1.093.39 -.32.75 EEO Minority Status .23.22 1.06.29 Minority Status Indirect Effect SE z p Men .08.04 1.96.05 Women .07.03 2.37.02 The results of the separate mediation te sts demonstrate that the model is supported for women, but all paths are not significant for men. However, all paths were in the hypothesized directions. Results for the simple mediation tests are shown in Table 96.

PAGE 229

194 Table 96. Results of Simple Mediation for Men and Women (PREDICTOR = EEO, Med = Subtle Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) .53.182.92.00.76.12 6.26 .00 b(MX) -.32.10-3.19.00-.31.08 -4.14 .00 b(YM.X) -.23.16-1.44.15-.24.10 -2.52 .01 b(YX.M) .45.192.42.02.68.12 5.52 .00 Next, a model predicting intention to turnover was examined. Support for overall mediation is again found as the paths fr om EEO to perceived discrimination ( B = -.35, t = -2.99, p < .01) and from perceived discrimi nation to intention to turnover ( B = .06, t = 4.36 p < .01) were significant. Perceived EEO was related to less perceived discrimination, which in turn was related to less intention to turnover. However, the interaction between minority status and the independent variables was not significant, indicating that relationships may not be different for men and women. The indirect effects were significant for bot h men and women participants. Results of the moderated mediation are shown in Table 97.

PAGE 230

195 Table 97. Results of Moderated Mediation (PREDIC TOR = EEO, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 16.591.80 9.23.00 EEO -.35.12 -2.99.00 Minority Status .572.09 .27.78 EEO Minority Status .03.14 .24.81 Turnover Constant 2.68.57 4.70.00 Subtle sex discrimination .06.01 4.36.00 EEO -.05.03 -1.40.16 Minority Status .10.60 .17.87 EEO Minority Status -.01.04 -.33.74 Minority Status Indirect Effect SE z p Men -.02.01 -2.42.02 Women -.02.01 -3.02.00 The separate mediation analyses demons trated that subtle sex discrimination mediates the relationship between EEO and intention to turnover for both men and women. All paths are in the hypot hesized directions. Table 98 displays the results for the simple mediation tests.

PAGE 231

196 Table 98. Results of Simple Mediation for Men and Women (PREDICTOR = EEO, Med = Subtle Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) -.07.03-2.09.04-.08.02 -3.68 .00 b(MX) -.35.11-3.21.00-.31.07 -4.20 .00 b(YM.X) .09.033.53.00.05.02 3.01 .00 b(YX.M) -.04.03-1.12.26-.06.02 -2.89 .00 Next, the mediating role of perceived di scrimination in the relationship between EEO and physical health was examined. The model was supported as the paths from EEO to the mediator ( B = -.35, t = -2.95, p < .01) and from perceived discrimination physical health ( B = -.29, t = -4.54, p < .01) were significant. Per ceived EEO was related to less perceived discrimination, which in turn was re lated to better physical health. However, there was no significant interaction between minority status and EEO. Additionally, indirect effects for both men and women were significant and similar in size. The results of the moderated mediation are displayed in Table 99.

PAGE 232

197 Table 99. Results of Moderated Mediation (PREDIC TOR = EEO, Med = Subtle Discrimination, DV = Physical Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 16.741.89 8.88.00 EEO -.35.12 -2.95.00 Minority Status .122.18 .06.95 EEO Minority Status .05.14 .36.72 Physical Health Constant 69.632.57 27.14.00 Subtle sex discrimination -.29.06 -4.54.00 EEO .04.15 .24.81 Minority Status -3.322.68 -1.24.22 EEO Minority Status .16.17 .93.35 Minority Status Indirect Effect SE z p Men .10.04 2.43.01 Women .09.03 2.99.00 The results for the separate tests of mediation are shown in Table 100. Here, perceived discrimination does not appear to mediate the relationship between EEO and physical health for men, although one of the path s is marginally significant. However, for the minority group (i.e. women), there is a re lationship. Additionally, all paths are in the hypothesized directions.

PAGE 233

198 Table 100. Results of Simple Mediation for Men and Women (PREDICTOR = EEO, Med = Subtle Discrimination, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .14.14.97.34.29.10 2.98 .00 b(MX) -.35.11-3.11.00-.30.08 -3.98 .00 b(YM.X) -.23.12-1.90.06-.32.08 -4.14 .00 b(YX.M) .06.15.40.69.19.10 1.97 .05 Finally, psychological health was investigat ed as a dependent variable in a model where perceived discrimination mediated the relationship between EEO and the psychological health. There was support for an overall mediating effect as the paths between EEO and perceived discrimination ( B = -.31, t = -2.69, p < .05) and from the mediator to psychological health ( B = -.25, t = -3.22, p < .01) were jointly significant. Perceived EEO was related to less perceived discrimination, wh ich in turn was related to better psychological health. However, the in teraction between EEO and minority status was not significant, indicati ng that minority status does not moderate the relationship between EEO and perceived discrimination. Furt her, the indirect effects for both men and women respondents were both significant, but similar in size and direction. Results are shown in Table 101.

PAGE 234

199 Table 101. Results of Moderated Mediation (PREDIC TOR = EEO, Med = Subtle Discrimination, DV = Psychological Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 15.841.81 8.76.00 EEO -.31.12 -2.69.01 Minority Status 1.232.11 .58.56 EEO Minority Status .01.14 .06.95 Psychological Health Constant 52.312.95 17.74.00 Subtle sex discrimination -.25.08 -3.22.00 EEO .09.17 .51.61 Minority Status -5.683.14 -1.80.07 EEO Minority Status .36.20 1.76.08 Minority Status Indirect Effect SE z p Men .08.04 2.01.04 Women .07.03 2.47.01 Similar to the tests of mediation for phys ical health, there is no support for the mediation model for men, but the model is fully supported for women. Results are shown in Table 102.

PAGE 235

200 Table 102. Results of Simple Mediation for Men and Women (PREDICTOR = EEO, Med = Subtle Discrimination, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) .16.18.93.36.52.11 4.68 .00 b(MX) -.31.10-3.15.00-.30.08 -3.80 .00 b(YM.X) .17.17-.97.33-.27.09 -3.17 .00 b(YX.M) .11.18.62.54.44.11 3.89 .00 Overt Sex-based Discrimination Perceived Equal Employment Opportunity (EEO) in the workplace was investigated as a predictor in a model where overt sex-based discrimination mediates the relationship betw een EEO and five depe ndent variables (job satisfaction, organizational commitment, inte ntion to turnover, physical health, and psychological health). Minority st atus was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived overt sex-based discrimi nation was investigated as a mediator in the relationship between EEO and job satis faction. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.73, t = -4.41, p < .01) and between the mediator and job satisfaction ( B = -.08, t = -4.25, p < .01) were significant. Perceived EEO was related to less perceived di scrimination, which in turn was related to greater job satisfaction. No significant interaction was found between the predictor (EEO) and the moderator (minority status). Thus, minority status is not a

PAGE 236

201 moderator in the current model. Additionall y, the indirect effects for men and women were both significant but similar in size and direction. Results of the moderated mediation are shown in Table 103. Table 103. Results of Moderated Mediation (PREDICTO R = EEO, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 24.172.55 9.46.00 EEO -.73.17 -4.41.00 Minority Status -4.533.00 -1.51.13 EEO Minority Status .33.20 1.65.10 Job Satisfaction Constant 10.051.13 8.86.00 Overt sex discrimination -.08.02 -4.25.00 EEO .12.07 1.78.08 Minority Status .461.21 .38.71 EEO Minority Status .00.08 .01.99 Minority Status Indirect Effect SE z p Men .06.02 3.02.00 Women .03.01 2.78.01 The results for the separate tests of me diation are shown in Table 104. Perceived discrimination was found to mediate the rela tionship between EEO and job satisfaction for both men and women. All paths we re in the hypothesized directions.

PAGE 237

202 Table 104. Results of Simple Mediation for Men and Women (PREDICTOR = EEO, Med = Overt Discrimination, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .18.072.54.01.16.04 3.67 .00 b(MX) -.73.16-4.69.00-.40.11 -3.67 .00 b(YM.X) -.16.04-4.21.00-.06.02 -2.39 .02 b(YX.M) .07.07.88.38.13.04 3.10 .00 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between EEO and organi zational commitment. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.71, t = -4.29, p < .01) and between the mediator and organizational commitment ( B = -.20, t = -3.66, p < .01) were significan t. Perceived EEO was re lated to less perceived discrimination which in turn was related to greater organizational commitment. No significant interaction was found between the predictor (EEO) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Further, the indirect effects for men and women we re both significant but similar in size and direction. Results for the moderated me diation are displayed in Table 105.

PAGE 238

203 Table 105. Results of Moderated Mediation (PREDICTO R = EEO, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 23.642.55 9.27.00 EEO -.71.16 -4.29.00 Minority Status -3.902.99 -1.30.19 EEO Minority Status .30.20 1.51.13 Organizational Constant 25.403.02 8.40.00 Overt sex discrimination -.20.05 -3.66.00 EEO .43.18 2.35.02 Minority Status -.933.23 -.29.77 EEO Minority Status .22.21 1.03.30 Minority Status Indirect Effect SE z p Men .14.05 2.74.01 Women .08.03 2.60.01 The results for the separate tests of me diation are shown in Table 106. Perceived discrimination was found to mediate the re lationship between EEO and organizational commitment for both men and women. Als o, all paths were in the hypothesized directions.

PAGE 239

204 Table 106. Results of Simple Mediation for Men and Women (PREDICTOR = EEO, Med = Overt Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) .57.183.23.00.73.12 6.14 .00 b(MX) -.71.15-4.64.00-.41.11 -3.72 .00 b(YM.X) -.23.10-2.29.02-.18.06 -2.89 .00 b(YX.M) .40.192.16.03.65.12 5.45 .00 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between EEO and intention to turnover. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.73, t = -4.43, p < .01) and between the mediator and intention to turnover ( B = .05, t = 5.22, p < .01) were significant. Perceived EEO was related to less perceived discrimination which in turn was related to less intention to tu rnover. No significant interaction was found between the predictor (EEO) and the moderator (minority status ). Thus, minority status is not a moderator in the current model. The i ndirect effects for men and women were both significant but similar in size and directi on. Results for the moderated mediation are similar for both men and women. Table 107 disp lays results of the moderated mediation.

PAGE 240

205 Table 107. Results of Moderated Mediation (PREDICTO R = EEO, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 24.172.54 9.51.00 EEO -.73.16 -4.43.00 Minority Status -4.422.98 -1.48.14 EEO Minority Status .32.20 1.62.10 Turnover Constant 2.71.53 5.11.00 Overt sex discrimination .05.01 5.22.00 EEO -.04.03 -1.39.16 Minority Status .08.57 .14.89 EEO Minority Status -.01.04 -.31.76 Minority Status Indirect Effect SE z p Men -.04.01 -3.34.00 Women -.02.01 -3.07.00 The results for the separate tests of me diation are shown in Table 108. Perceived discrimination was found to mediate the re lationship between EEO and intention to turnover for both men and women. Moreover, all paths were in the hypothesized directions.

PAGE 241

206 Table 108. Results of Simple Mediation for Men and Women (PREDICTOR = EEO, Med = Overt Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) -.08.03-2.55.01-.08.02 -3.63 .00 b(MX) -.73.16-4.69.00-.41.11 -3.78 .00 b(YM.X) .06.023.69.00.04.01 3.87 .00 b(YX.M) -.03.03-1.05.30-.06.02 -2.77 .01 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between EEO and physical health. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.67, t = -3.74, p < .01) and between the mediator and physical health ( B = -.22, t = -5.19, p < .01) were significant. Perceived EEO was related to less perceived discrimination which in turn was related to better physical health. No si gnificant interaction was found between the predictor (EEO) and the moderator (minority status). Thus, minority status was not a moderator in the current model. The indi rect effects for men and women were both significant but similar in size and directi on. Results for the moderated mediation are displayed in Table 109.

PAGE 242

207 Table 109. Results of Moderated Mediation (PREDICTO R = EEO, Med = Overt Discrimination, DV = Physical Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 23.182.82 8.22.00 EEO -.67.18 -3.74.00 Minority Status -4.013.24 -1.23.22 EEO Minority Status .29.21 1.36.17 Physical Health Constant 69.092.53 27.30.00 Overt sex discrimination -.22.04 -5.19.00 EEO .03.15 .18.86 Minority Status -3.252.68 -1.21.23 EEO Minority Status .16.17 .92.36 Minority Status Indirect Effect SE z p Men .15.05 3.00.00 Women .09.03 2.86.00 The results for the separate tests of mediation are displayed in Table 110. Perceived discrimination was found to me diate the relationship between EEO and physical health for both men and women.

PAGE 243

208 Table 110. Results of Simple Mediation for Men and Women (PREDICTOR = EEO, Med = Overt Discrimination, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .18.151.19.23.27.10 2.87 .00 b(MX) -.67.16-4.08.00-.38.11 -3.37 .00 b(YM.X) -.28.08-3.50.00-.20.05 -3.98 .00 b(YX.M) -.01.15-.08.93.19.09 2.06 .04 Finally, perceived overt sexbased discrimination was inve stigated as a mediator in the relationship between EEO and psychol ogical health. Support fo r overall mediation was found as the paths between EEO and perceived discrimination ( B = -.73, t = -4.33, p < .01) and between the mediator and psychological health ( B = -.21, t = -4.24, p < .01) were significant. Perceived EEO was related to less perceived discrimination which in turn was related to better psychological health. No significant interaction was found between the predictor (EEO) a nd the moderator (minority stat us). Thus, minority status was not a moderator in the current model. A dditionally, the indirect effects for men and women were both significant but similar in si ze and direction. Results for the moderated mediation are displayed in Table 111.

PAGE 244

209 Table 111. Results of Moderated Mediation (PREDICTO R = EEO, Med = Overt Discrimination, DV = Psychological Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 23.992.62 9.17.00 EEO -.73.17 -4.33.00 Minority Status -4.673.09 -1.51.13 EEO Minority Status .35.20 1.74.08 Psychological Health Constant 53.422.79 19.13.00 Overt sex discrimination -.21.05 -4.24.00 EEO .02.17 .09.93 Minority Status -6.362.99 -2.12.03 EEO Minority Status .39.20 1.97.05 Minority Status Indirect Effect SE z p Men .15.05 2.99.00 Women .08.03 2.58.01 Results for the simple tests of media tion are found in Table 112. Perceived discrimination was found to mediate the rela tionship between EEO and physical health for women, but not men. However, the path fr om EEO to the mediator was marginally significant. Also, all paths were in the pred icted directions. Given the fewer number of men than women, the lack of mediation fo r men may be due to a lack of power.

PAGE 245

210 Table 112. Results of Simple Mediation for Men and Women (PREDICTOR = EEO, Med = Overt Discrimination, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) .17.171.01.31.48.11 4.36 .00 b(MX) -.73.15-4.76.00-.38.12 -3.22 .00 b(YM.X) -.19.10-1.93.06-.22.06 -3.79 .00 b(YX.M) .03.18.17.86.40.11 3.64 .00 Subtle Race-based Discrimination Perceived Equal Employment Opportunity (EEO) in the workplace was investigated as a predictor in a model where subtle racebased discrimination mediates the relati onship between EEO and five dependent variables (job satisfaction, or ganizational commitment, intention to turnover, physical health, and psychological health). Minority status was investig ated as a moderator in the relationship between the pred ictor and the mediator. First, perceived subtle race-based discrimi nation was investigated as a mediator in the relationship between EEO and job satis faction. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.21, t = -2.53, p < .05) and between the mediator and job satisfaction ( B = -.12, t = -4.08, p < .01) were significant. Perceived EEO was related to less perceived discrimination which in turn was related to greater job satisfaction. No si gnificant interaction was found between the predictor (EEO) and the moderator (minority status). Thus, minority status is not a

PAGE 246

211 moderator in the current model. Additionall y, the indirect effects for men and women were both significant, and similar in size and direction. Results for the moderated mediation can be found in Table 113. Table 113. Results of Moderated Mediation (PREDIC TOR = EEO, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 13.501.23 10.96.00 EEO -.21.08 -2.53.01 Minority Status 5.091.90 2.67.01 EEO Minority Status -.18.13 -1.43.15 Job Satisfaction Constant 10.12.82 12.36.00 Subtle race discrimination -.12.03 -4.08.00 EEO .14.05 2.81.01 Minority Status .631.12 .56.58 EEO Minority Status -.01.07 -.18.86 Minority Status Indirect Effect SE z p Whites .03.01 2.11.03 Non-whites .05.02 2.84.00 The results for the separate tests of me diation are displayed in Table 114. Subtle race-based discrimination was found to me diate the relationship between EEO and job satisfaction for both white and non-white resp ondents. Also, all paths were in the predicted directions.

PAGE 247

212 Table 114. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = EEO, Med = Subtle Discrimination, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) .16.053.27.00.17.06 2.98 .00 b(MX) -.21.07-3.16.00-.39.12 -3.41 .00 b(YM.X) -.13.05-2.71.01-.11.04 -3.08 .00 b(YX.M) .13.052.68.01.13.06 2.20 .03 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between EEO and orga nizational commitment. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.21, t = -2.56, p < .05) and between the medi ator and job satisfaction ( B = -.23, t = -2.93, p < .01) were significant. Perceived EEO was rela ted to less perceived discrimination which in turn was related to greater organizationa l commitment. No significant interaction was found between the predictor (EEO) and the m oderator (minority status). Thus, minority status is not a moderator in th e current model. While the indire ct effect is significant for non-white individuals, the indi rect effect is approach ing significance for white participants. Results for the moderate d mediation can be found in Table 115.

PAGE 248

213 Table 115. Results of Moderated Mediation (PREDIC TOR = EEO, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 13.371.21 11.07.00 EEO -.21.08 -2.56.01 Minority Status 5.011.88 2.66.01 EEO Minority Status -.17.13 -1.34.18 Organizational Constant 24.222.17 11.18.00 Subtle race discrimination -.23.08 -2.93.00 EEO .53.13 4.12.00 Minority Status -.562.98 -.19.85 EEO Minority Status .20.20 .99.32 Minority Status Indirect Effect SE z p Whites .05.03 1.87.06 Non-whites .09.04 2.31.02 The results for the separate tests of mediation are shown in Table 116. Subtle race-based discrimination was found to mediate the relationship between EEO and organizational commitment for non-white pa rticipants, but the model was not fully supported for white participants. However, al l paths were in the predicted directions.

PAGE 249

214 Table 116. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = EEO, Med = Subtle Discrimination, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) .58.134.34.00.81.15 5.62 .00 b(MX) -.21.06-3.28.00-.38.12 -3.26 .00 b(YM.X) -.24.14-1.66.10-.23.09 -2.50 .01 b(YX.M) .53.143.89.00.73.15 4.95 .00 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between EEO and intention to turnover. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.21, t = -2.51, p < .05) and between the mediator and intention to turnover ( B = .08, t = 5.88, p < .01) were significant. Perceived EEO was related to less perceived discrimination which in turn as related to less intention to turnove r. No significant interaction was found between the predictor (EEO) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Additionally, th e indirect effects are significant for both white and non-white participants, and the indirect effects are similar in size. The results of the moderated mediation are shown in Table 117.

PAGE 250

215 Table 117. Results of Moderated Mediation (PREDIC TOR = EEO, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 13.521.23 11.02.00 EEO -.21.08 -2.51.01 Minority Status 4.911.90 2.58.01 EEO Minority Status -.17.13 -1.36.17 Turnover Constant 3.03.37 8.09.00 Subtle race discrimination .08.01 5.88.00 EEO -.07.02 -3.28.00 Minority Status -.95.51 -1.86.06 EEO Minority Status .04.03 1.16.25 Minority Status Indirect Effect SE z p Whites -.01.01 -2.28.02 Non-whites -.03.01 -3.24.00 The results for the separate tests of mediation are shown in Table 118. Subtle race-based discrimination was found to mediate the relationship between EEO and intention to turnover for both white and non-wh ite participants. More over, all paths were in predicted directions.

PAGE 251

216 Table 118. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = EEO, Med = Subtle Discrimination, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) -.09.02-3.91.00-.06.03 -2.36 .02 b(MX) -.21.07-3.09.00-.38.11 -3.33 .00 b(YM.X) .10.024.35.00.07.02 4.08 .00 b(YX.M) -.07.02-3.10.00-.04.03 -1.42 .16 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between EEO and physical health. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.19, t = -2.33, p < .05) and between the mediator and physical health ( B = -.31, t = -4.72, p < .01) were jointly significant. Perceived EEO was relate d to less perceived di scrimination which was in turn related to better physical health. No significant interaction was found between the predictor (EEO) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Further, the indirect effects are si gnificant for both white and non-white participants, and the indirect effects are similar in size. Results of the moderated mediation are shown in Table 119.

PAGE 252

217 Table 119. Results of Moderated Mediation (PREDIC TOR = EEO, Med = Subtle Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 13.131.21 10.84.00 EEO -.19.08 -2.33.02 Minority Status 4.761.93 2.45.01 EEO Minority Status -.16.13 -1.25.21 Physical Health Constant 67.531.72 39.29.00 Subtle race discrimination -.31.06 -4.72.00 EEO .12.10 1.16.25 Minority Status -2.422.40 -1.01.31 EEO Minority Status .17.16 1.04.30 Minority Status Indirect Effect SE z p Whites .06.03 2.06.04 Non-whites .11.04 2.79.01 The results for the separate tests of me diation are displayed in Table 120. Subtle race-based discrimination was found to mediate the relationship between EEO and physical health for non-white and white part icipants. Paths are in the hypothesized directions.

PAGE 253

218 Table 120. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = EEO, Med = Subtle Discrimination, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) .18.101.79.07.39.13 2.93 .00 b(MX) -.19.07-2.80.01-.35.12 -2.99 .00 b(YM.X) -.27.10-2.72.01-.33.09 -3.77 .00 b(YX.M) .12.101.26.21.28.13 2.09 .04 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between EEO and psyc hological health. There was not support found for overall mediation. Although the path from EEO to the mediator was significant ( B = .22, t = -2.49, p < .05), the path between the mediator and psychological health only approached significance ( B = -.14, t = -1.96, p = .05). No significant interaction was found between the predictor (EEO) and the m oderator (minority status). Thus, minority status is not a moderator in the current model. Neither the indirect effects for white nor non-white participants were significant. Result s of the moderated mediation are displayed in Table 121.

PAGE 254

219 Table 121. Results of Moderated Mediation (PREDIC TOR = EEO, Med = Subtle Discrimination, DV = Psychological Health, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 13.651.28 10.65.00 EEO -.22.09 -2.49.01 Minority Status 4.852.00 2.43.02 EEO Minority Status -.17.13 -1.24.22 Psychological Health Constant 43.742.02 21.60.00 Subtle race discrimination -.14.07 -1.96.05 EEO .56.12 4.61.00 Minority Status 6.722.78 2.41.02 EEO Minority Status -.47.19 -2.54.01 Minority Status Indirect Effect SE z p Whites .03.02 1.47.14 Non-whites .05.03 1.69.09 The results for the separate tests of me diation are shown in Table 122. There was not support for a mediating effect in either white or non-white i ndividuals, although the non-white individuals exhibi ted one significant path and one path approaching significance. However, all paths we re in the predicted directions.

PAGE 255

220 Table 122. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = EEO, Med = Subtle Discrimination, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) .59.124.84.00.14.14 .99 .32 b(MX) -.22.07-3.05.00-.38.12 -3.13 .00 b(YM.X) -.10.12-.81.42-.17.09 -1.88 .06 b(YX.M) .57.124.57.00.08.14 .53 .60 Overt Race-based Discrimination Perceived Equal Employment Opportunity (EEO) in the workplace was investigated as a predictor in a model where overt racebased discrimination mediates the relati onship between EEO and five dependent variables (job satisfaction, or ganizational commitment, intention to turnover, physical health, and psychological health). Minority status was investig ated as a moderator in the relationship between the pred ictor and the mediator. First, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between EEO and job satis faction. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.33, t = -2.98, p < .01) and between the mediator and job satisfaction ( B = -.09, t = -4.34, p < .01) were jointly significant. Perceived EEO was relate d to less perceived di scrimination which in turn was related to greater job satisfacti on. No significant interaction was found between the predictor (EEO) and the moderator (minority status). Thus, minority status is not a

PAGE 256

221 moderator in the current model. The indirect effects are significant for both white and non-white participants, and were similar in si ze. Results for the moderated mediation are shown in Table 123. Table 123. Results of Moderated Mediation (PREDICTO R = EEO, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Overt race discrimination Constant 16.521.68 9.85.00 EEO -.33.11 -2.98.00 Minority Status 6.292.63 2.39.02 EEO Minority Status -.24.18 -1.35.18 Job Satisfaction Constant 10.02.80 12.59.00 Overt race discrimination -.09.02 -4.34.00 EEO .13.05 2.74.01 Minority Status .611.13 .54.59 EEO Minority Status -.02.08 -.26.80 Minority Status Indirect Effect SE z p Whites .03.01 2.42.02 Non-whites .05.02 2.99.00 The results for the separate tests of mediation are shown in Table 124. A significant mediating effect was found for bot h white and non-white participants. All paths were in the predicted directions.

PAGE 257

222 Table 124. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = EEO, Med = Overt Discrimination, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) .16.053.31.00.17.06 2.86 .00 b(MX) -.34.10-3.29.00-.57.15 -3.84 .00 b(YM.X) -.13.03-4.17.00-.06.03 -2.08 .04 b(YX.M) .12.052.46.01.13.06 2.19 .03 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between EEO and organi zational commitment. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.34, t = -3.03, p < .01) and between the mediator and organizational commitment ( B = -.20, t = -3.50, p < .01) were jointly significant. Perc eived EEO was related to less perceived discrimination which in turn was related to greater organizational commitment. No significant interaction was found between the predictor (EEO) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects were significant for both white and non-white participants, and were similar in size. Results ar e displayed in Table 125.

PAGE 258

223 Table 125. Results of Moderated Mediation (PREDICTO R = EEO, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Overt race discrimination Constant 16.451.66 9.93.00 EEO -.34.11 -3.03.00 Minority Status 6.052.62 2.31.02 EEO Minority Status -.21.18 -1.21.23 Organizational Constant 24.932.08 11.97.00 Overt race discrimination -.20.06 -3.50.00 EEO .49.13 3.82.00 Minority Status -1.422.97 -.48.63 EEO Minority Status .23.20 1.17.24 Minority Status Indirect Effect SE z p Whites .07.03 2.24.03 Non-whites .11.04 2.61.01 The results for the separate tests of mediation are displa yed in Table 126. A significant mediating effect was found for bot h white and non-white participants. Also, all paths were in the expected directions.

PAGE 259

224 Table 126. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = EEO, Med = Overt Discrimination, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) .56.134.18.00.83.15 5.63 .00 b(MX) -.34.10-3.38.00-.55.15 -3.64 .00 b(YM.X) -.26.09-2.94.00-.15.07 -2.06 .04 b(YX.M) .47.133.49.00.74.15 4.94 .00 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between EEO and intention to turnover. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.34, t = -3.00, p < .01) and between the mediator and intention to turnover ( B = .05, t = 5.61, p < .01) were jointly significant. Perceived EEO wa s related to less perceived discrimination which in turn was related to less intention to turnover. No significant interaction was found between the predictor (EEO) and the m oderator (minority status). Thus, minority status is not a moderator in th e current model. The indirect effects are significant for both white and non-white participants, and were similar in size. Results for the moderated mediation are displayed in Table 127.

PAGE 260

225 Table 127. Results of Moderated Mediation (PREDICTO R = EEO, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Overt race discrimination Constant 16.681.68 9.91.00 EEO -.34.11 -3.00.00 Minority Status 5.952.65 2.24.03 EEO Minority Status -.22.18 -1.23.22 Turnover Constant 3.11.36 8.59.00 Overt race discrimination .05.01 5.61.00 EEO -.07.02 -3.04.00 Minority Status -.77.52 -1.49.14 EEO Minority Status .03.03 .97.33 Minority Status Indirect Effect SE z p Whites -.02.01 -2.62.01 Non-whites -.03.01 -3.27.00 The results for the separate tests of me diation are shown in Table 128. Perceived overt race-based discrimina tion was found to mediate the relationship between EEO and organizational commitment for both white and non-white individuals. Moreover, all paths were in hypothesized directions.

PAGE 261

226 Table 128. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = EEO, Med = Overt Discrimination, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) -.09.02-3.78.00-.06.03 -2.34 .02 b(MX) -.34.10-3.26.00-.56.15 -3.75 .00 b(YM.X) .07.015.37.00.04.01 2.70 .01 b(YX.M) -.06.02-2.78.01-.04.03 -1.56 .12 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between EEO and physical health. Support for overall mediation was found as the paths between EEO and perceived discrimination ( B = -.33, t = -2.82, p < .05) and between the mediator and physical health ( B = -.28, t = -6.35, p < .01) were jointly significant. Perceived EEO was relate d to less perceived di scrimination which in turn was related to better physical health. No significant inter action was found between the predictor (EEO) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are significant for both white and non-white participants, and were similar in size. Results are displayed in Table 129.

PAGE 262

227 Table 129. Results of Moderated Mediation (PREDICTO R = EEO, Med = Overt Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 16.281.72 9.47.00 EEO -.33.11 -2.82.01 Minority Status 5.792.79 2.08.04 EEO Minority Status -.20.19 -1.08.28 Physical Health Constant 67.621.62 41.72.00 Overt race discrimination -.28.04 -6.35.00 EEO .12.10 1.25.21 Minority Status -1.512.37 -.64.53 EEO Minority Status .10.16 .66.51 Minority Status Indirect Effect SE z p Whites .09.04 2.56.01 Non-whites .15.05 3.12.00 The results for the separate tests of mediation are displa yed in Table 130. A significant mediating effect was found for bot h white and non-white participants. All paths were in the predicted directions.

PAGE 263

228 Table 130. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = EEO, Med = Overt Discrimination, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) .21.102.21.03.37.14 2.72 .01 b(MX) -.33.11-3.07.00-.53.16 -3.30 .00 b(YM.X) -.28.06-4.65.00-.28.07 -4.29 .00 b(YX.M) .12.091.30.20.22.14 1.66 .10 Finally, perceived overt race-based discrimi nation was investigated as a mediator in the relationship between EEO and psychol ogical health. Support fo r overall mediation was found as the paths between EEO and perceived discrimination ( B = -.36, t = -3.06, p < .01) and between the mediator and psychological health ( B = -.17, t = -3.24, p < .01) were jointly significant. Perceived EEO wa s related to less perceived discrimination which in turn was related to better psychol ogical health. No signi ficant interaction was found between the predictor (EEO) and the m oderator (minority status). Thus, minority status is not a moderator in the current m odel. Additionally, the indirect effects are significant for both white and nonwhite participants, and were similar in size. Results are shown in Table 131.

PAGE 264

229 Table 131. Results of Moderated Mediation (PREDICTO R = EEO, Med = Overt Discrimination, DV = Psychological Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 17.071.76 9.71.00 EEO -.36.12 -3.06.00 Minority Status 5.842.78 2.10.04 EEO Minority Status -.21.18 -1.12.26 Psychological Health Constant 44.911.95 22.99.00 Overt race discrimination -.17.05 -3.24.00 EEO .52.12 4.36.00 Minority Status 6.092.78 2.19.03 EEO Minority Status -.43.18 -2.32.02 Minority Status Indirect Effect SE z p Whites .06.03 2.17.03 Non-whites .09.04 2.46.01 A significant mediating effect was found for non-white participants; however, there was not support for the mediating effect in white participants. All paths were in hypothesized directions. Results for the simple tests of mediation are shown in Table 132.

PAGE 265

230 Table 132. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = EEO, Med = Overt Discrimination, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) .58.124.84.00.18.14 1.27 .21 b(MX) -.36.11-3.31.00-.57.16 -3.65 .00 b(YM.X) -.12.07-1.67.10-.21.07 -2.93 .00 b(YX.M) .53.124.38.00.06.15 .43 .67 Results for Minority Segmentation Subtle Sex-based Discrimination Perceived minority segmentation in the workplace was investigated as a predicto r in a model where subtle sex-based discrimination mediates the relationship between minority segmentation and five dependent variables (job satis faction, organizational commitm ent, intention to turnover, physical health, and psychologi cal health). Minority stat us was investigated as a moderator in the relationship betwee n the predictor and the mediator. First, perceived subtle sex-based discrimi nation was investigated as a mediator in the relationship between minority segmentati on and job satisfaction. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .29, t = 3.10, p < .01) and between the mediator and job satisfaction ( B = -.11, t = -3.63, p < .01) were jointly si gnificant. Minority se gmentation was related to greater perceived discrimination which in turn was related to le ss job satisfaction. No

PAGE 266

231 significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are significant for both men and women, and were similar in size. The results of the moderated mediation are shown in Table 133. Table 133. Results of Moderated Mediation (PREDICTO R = Minority Segmenta tion, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 8.031.16 6.94.00 MINORITY SEG .29.09 3.10.00 Minority Status .841.42 .60.55 MINORITY SEG Minority Status .04.11 .34.73 Job satisfaction Constant 12.35.76 16.25.00 Subtle sex discrimination -.11.03 -3.63.00 MINORITY SEG -.02.06 -.37.71 Minority Status 1.34.88 1.52.14 MINORITY SEG Minority Status -.09.07 -1.28.20 Minority Status Indirect Effect SE z p Men -.03.01 -2.31.02 Women -.04.01 -2.93.00 A significant mediating effect was found for both men and women. Thus, subtle sex-based discrimination medi ates the relationship between minority segmentation and job satisfaction. Paths are not all in the hypothe sized directions. Path coefficients were predicted to be different in directions based on minority stat us. However, paths are in the same direction for both majority and mi nority group members. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceive d discrimination, whereas majority group

PAGE 267

232 members would perceive a nega tive relationship. However, both groups demonstrated a positive relationship. Results for the simple te sts of mediation are shown in Table 134. Table 134. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Subtle Discri mination, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) -.05.06-.87.39-.15.04 -3.78 .00 b(MX) .29.093.30.00.32.06 5.00 .00 b(YM.X) -.19.06-3.21.00-.08.04 -2.26 .02 b(YX.M) .00.06.04.97-.12.04 -2.99 .00 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between minority segmentati on and organizational commitment. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .26, t = 2.79, p < .05) and between the mediator and organizational commitment ( B = -.29, t = -3.41, p < .01) were jointly significant. Minority segmentation was related to greater perceived discrimination which in turn was related to less organizational commitment. No significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current m odel. Additionally, the indirect effects are significant for both men and women, and were si milar in size. Results are shown in Table 135.

PAGE 268

233 Table 135. Results of Moderated Mediation (PREDICTO R = Minority Segmenta tion, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 8.201.15 7.16.00 MINORITY SEG .26.09 2.79.01 Minority Status .581.40 .41.68 MINORITY SEG Minority .08.11 .70.49 Organizational Constant 32.652.05 15.89.00 Subtle sex discrimination -.29.08 -3.41.00 MINORITY SEG -.01.16 -.05.96 Minority Status 6.732.36 2.85.00 MINORITY SEG Minority -.41.19 -2.18.03 Minority Status Indirect Effect SE z p Men -.07.04 -2.11.04 Women -.10.03 -2.84.00 The results for the separate tests of mediation are shown in Table 136. A significant mediating effect was found for bot h men and women. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceived discrimination, whereas majority group member s would perceive a negative relationship. However, both groups demonstrated a positive relationship.

PAGE 269

234 Table 136. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Subtle Discrimina tion, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) -.08.15-.54.59-.51.11 -4.71 .00 b(MX) .26.083.10.00.33.06 5.14 .00 b(YM.X) -.33.16-2.06.04-.27.10 -2.74 .01 b(YX.M) .00.15.02.98-.42.11 -3.75 .00 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between minority segmentati on and intention to turnover. Support for overall mediation was found as the paths be tween minority segmentation and perceived discrimination ( B = .29, t = 3.08, p < .01) and between the mediator and intention to turnover ( B = .06, t = 4.14, p < .01) were jointly significan t. Minority segmentation was associated with greater perceived discrimina tion which in turn was related to greater turnover intention. No signifi cant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Further, the indirect effects were significant for both men and women, and were similar in size Results are displayed in Table 137.

PAGE 270

235 Table 137. Results of Moderated Mediation (PREDICTO R = Minority Segmenta tion, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 8.071.16 6.95.00 MINORITY SEG .29.09 3.08.00 Minority Status .821.42 .57.57 MINORITY SEG Minority Status .04.11 .36.72 Turnover Constant 2.07.35 5.83.00 Subtle sex discrimination .06.01 4.14.00 MINORITY SEG -.01.03 -.26.78 Minority Status -.80.41 -1.96.05 MINORITY SEG Minority Status .07.03 2.06.04 Minority Status Indirect Effect SE z p Men .02.01 2.43.02 Women .02.01 3.18.00 The results for the separate tests of mediation are shown in Table 138. A significant mediating effect was found for bot h men and women. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceived discrimination, whereas majority group member s would perceive a negative relationship. However, both groups demonstrated a positive relationship.

PAGE 271

236 Table 138. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Subtle Discrimina tion, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) .01.03.37.71.08.02 4.20 .00 b(MX) .29.093.27.00.33.07 4.99 .00 b(YM.X) .10.033.72.00.05.02 2.63 .01 b(YX.M) -.02.03-.66.51.06.02 3.31 .00 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between minority segmentati on and physical health. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .26, t = 2.82, p < .05) and between the mediator and physical health ( B = -.33, t = -5.06, p < .01) were jointly si gnificant. Minority se gmentation was related to greater perceived discrimination which in turn was related to poorer physical health. No significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Additionally, the indir ect effects are significant for both men and women, and were similar in size. Results for the m oderated mediation are shown in Table 139.

PAGE 272

237 Table 139. Results of Moderated Mediation (PREDICTO R = Minority Segmenta tion, Med = Subtle Discrimination, DV = Physical Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 8.291.17 7.09.00 MINORITY SEG .26.09 2.82.01 Minority Status .581.44 .40.69 MINORITY SEG Minority .05.11 .45.65 Physical Health Constant 70.651.55 45.44.00 Subtle sex discrimination -.33.07 -5.06.00 MINORITY SEG .00.12 .03.97 Minority Status -.091.79 -.05.96 MINORITY SEG Minority -.09.14 -.63.53 Minority Status Indirect Effect SE z p Men -.09.04 -2.43.02 Women -.10.03 -3.42.00 A significant mediating effect was f ound for women, and the results for men approached significance. Paths are not al l in the hypothesized directions. Path coefficients were predicted to be differe nt in directions ba sed on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentati on and perceived discrimination, whereas majority group members would perceive a ne gative relationship. However, both groups demonstrated a positive relationship. Results for the simple tests of mediation are shown in Table 140.

PAGE 273

238 Table 140. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Subtle Discri mination, DV = Physical Health) Men Women B SE t p B SE t p b(YX) -.08.11-.74.46-.19.09 -2.17 .03 b(MX) .26.092.91.00.31.07 2.69 .00 b(YM.X) -.23.12-1.95.05-.38.08 -4.73 .00 b(YX.M) -.02.12-.21.84-.07.09 -.82 .41 Finally, perceived subtle sex-based discri mination was investigated as a mediator in the relationship between minority segmen tation and psychologica l health. Support for overall mediation was found as the paths be tween minority segmentation and perceived discrimination ( B = .26, t = 2.77, p < .05) and between the mediator and psychological health ( B = -.26, t = -3.43, p < .01) were jointly significan t. Minority segmentation was related to greater perceived discriminati on which in turn was related to poorer psychological health. No significant inte raction was found between the predictor (minority segmentation) and the moderator (mi nority status). Thus, mi nority status is not a moderator in the current model. The indir ect effects are significant for both men and women, and were similar in size. Re sults are displayed in Table 141.

PAGE 274

239 Table 141. Results of Moderated Mediation (PREDICTO R = Minority Segmenta tion, Med = Subtle Discrimination, DV = Psychologi cal Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 8.181.15 7.12.00 MINORITY SEG .26.09 2.77.01 Minority Status .951.42 .67.50 MINORITY SEG Minority Status .06.11 .49.63 Psychological Health Constant 56.181.83 30.76.00 Subtle sex discrimination -.26.08 -3.43.00 MINORITY SEG -.19.14 -1.34.18 Minority Status -.652.13 -.31.76 MINORITY SEG Minority Status -.01.17 -.07.95 Minority Status Indirect Effect SE z p Men -.07.03 -2.10.04 Women -.08.03 -2.74.01 A significant mediating effect was found for women, but the model was not fully supported for men. Thus, subtle sex-based discrimination mediates the relationship between minority segmentation and psychologi cal health for women, but not for men. Paths are not all in the hypothe sized directions. Path coeffici ents were predicted to be different in directions based on minority status. However, path s are in the same direction for both majority and minority group member s. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceived discriminati on, whereas majority group members would perceive a negative relationship. Howeve r, both groups demonstrated a positive relationship. Results of the simple test s of mediation are shown in Table 142.

PAGE 275

240 Table 142. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Subtle Discrimi nation, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) -.25.14-1.84.07-.28.10 -2.82 .01 b(MX) .26.083.20.00.31.07 4.49 .00 b(YM.X) -.10.16-.61.54-.32.09 -3.62 .00 b(YX.M) -.23.14-1.59.11-.18.10 -1.80 .07 Overt Sex-based Discrimination Perceived minority segmentation in the workplace was investigated as a predic tor in a model where overt sex-based discrimination mediates the relationship between minority segmentation and five dependent variables (job satis faction, organizational commitm ent, intention to turnover, physical health, and psychologi cal health). Minority stat us was investigated as a moderator in the relationship betwee n the predictor and the mediator. First, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between minority segmentati on and job satisfaction. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .54, t = 4.02, p < .01) and between the mediator and job satisfaction ( B = -.09, t = -4.48, p < .01) were jointly si gnificant. Minority se gmentation was related to greater perceived discrimination which was related to less job satisfaction. No significant interaction was found between the predictor (minority segmentation) and the

PAGE 276

241 moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects were significant for both men and women, and were similar in size. Results are shown in Table 143. Table 143. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 6.941.69 4.10.00 MINORITY SEG .54.13 4.02.00 Minority Status 1.892.09 .91.36 MINORITY SEG Minority -.12.16 -.75.45 Job Satisfaction Constant 12.11.70 17.32.00 Overt sex discrimination -.09.02 -4.48.00 MINORITY SEG -.01.06 -.20.84 Minority Status 1.15.84 1.36.18 MINORITY SEG Minority -.07.07 -1.02.31 Minority Status Indirect Effect SE z p Men -.05.02 -2.95.00 Women -.04.01 -3.10.00 The results for the separate tests of mediation are shown in Table 144. A significant mediating effect was found for both men and women. T hus, overt sex-based discrimination mediated th e relationship between mi nority segmentation and job satisfaction. Additionally, pa ths are not all in the hypot hesized directions. Path coefficients were predicted to be differe nt in directions ba sed on minority status. However, paths are in the same direction for both majority and minority group members.

PAGE 277

242 Table 144. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Overt Discrimination, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) -.06.06-1.00.32-.12.04 -3.08 .00 b(MX) .54.134.17.00.42.10 4.33 .00 b(YM.X) -.17.04-4.63.00-.05.02 -2.34 .02 b(YX.M) .03.06.58.56-.09.04 -2.42 .02 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between minority segmentati on and organizational commitment. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .52, t = 3.80, p < .01) and between the mediator and organizational commitment ( B = -.24, t = -4.40, p < .01) were jointly significant. Minority segmentation was related to greater perceived discrimination which was in turn related to less organizational commitment. No significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects were significant for both men and women, and were similar in size. Results for the moderated mediation are shown in Table 145.

PAGE 278

243 Table 145. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 7.061.70 4.15.00 MINORITY SEG .52.14 3.80.00 Minority Status 1.692.09 .81.42 MINORITY SEG Minority -.09.17 -.54.59 Organizational Constant 32.051.92 16.74.00 Overt sex discrimination -.24.05 -4.40.00 MINORITY SEG .03.15 .21.83 Minority Status 6.542.31 2.83.00 MINORITY SEG Minority -.40.18 -2.17.03 Minority Status Indirect Effect SE z p Men -.12.04 -2.84.00 Women -.10.03 -3.11.00 The results for the separate tests of mediation are shown in Table 146. A significant mediating effect was found for bot h men and women. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceived discrimination, whereas majority group member s would perceive a negative relationship. However, both groups demonstrated a positive relationship.

PAGE 279

244 Table 146. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Overt Discrimina tion, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) -.09.15-.62.54-.47.11 -4.30 .00 b(MX) .52.134.00.00.43.10 4.40 .00 b(YM.X) -.32.10-3.28.00-.21.07 -3.17 .00 b(YX.M) .08.15.50.62-.38.11 -3.42 .00 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between minority segmentati on and intention to turnover. Support for overall mediation was found as the paths be tween minority segmentation and perceived discrimination ( B = .54, t = 4.03, p < .01) and between the mediator and intention to turnover ( B = .05, t = 5.35, p < .01) were jointly significan t. Minority segmentation was related to greater perceived discrimination wh ich in turn was related to greater turnover intention. No significant in teraction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are significant for both men and women, and were similar in size. Re sults are displayed in Table 147.

PAGE 280

245 Table 147. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 6.941.69 4.11.00 MINORITY SEG .54.13 4.03.00 Minority Status 1.812.08 .87.39 MINORITY SEG Minority -.11.16 -.69.49 Turnover Constant 2.25.33 6.90.00 Overt sex discrimination .05.01 5.35.00 MINORITY SEG -.02.03 -.80.43 Minority Status -.83.39 -2.11.04 MINORITY SEG Minority .07.03 2.21.03 Minority Status Indirect Effect SE z p Men .03.01 3.19.00 Women .02.01 3.42.00 A significant mediating effect was f ound for both men and women. Thus, overt sex-based discrimination medi ates the relationship between minority segmentation and intention to turnover. Paths are not all in th e hypothesized directions Path coefficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority a nd minority group members. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceive d discrimination, whereas majority group members would perceive a nega tive relationship. However, both groups demonstrated a positive relationship. Results for the tests of simple mediation are shown in Table 148.

PAGE 281

246 Table 148. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Overt Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) .01.03.25.80.07.02 3.74 .00 b(MX) .54.134.17.00.43.10 4.44 .00 b(YM.X) .07.024.37.00.04.01 3.61 .00 b(YX.M) -.03.03-1.25.21.05.02 2.77 .01 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between minority segmentati on and physical health. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .50, t = 3.62, p < .01) and between the mediator and physical health ( B = -.24, t = -5.56, p < .01) were jointly si gnificant. Minority se gmentation was related to greater perceived discrimination which in turn was related to poorer physical health. No significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Additionally, the indir ect effects were significan t for both men and women, and were similar in size. Results are shown in Table 149.

PAGE 282

247 Table 149. Results of Moderated Mediation (PREDIC TOR = Physical Health, Med = Overt Discrimination, DV = Physical Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 7.061.75 4.04.00 MINORITY SEG .50.14 3.62.00 Minority Status 1.762.15 .82.42 MINORITY SEG Minority -.09.17 -.54.59 Physical Health Constant 69.551.49 46.70.00 Overt sex discrimination -.24.04 -5.56.00 MINORITY SEG .02.12 .20.84 Minority Status .191.80 .11.92 MINORITY SEG Minority -.11.14 -.79.43 Minority Status Indirect Effect SE z p Men -.12.04 -3.00.00 Women -.10.03 -3.28.00 The results for the separate tests of mediation are shown in Table 150. A significant mediating effect was found for bot h men and women. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceived discrimination, whereas majority group member s would perceive a negative relationship. However, both groups demonstrated a positive relationship.

PAGE 283

248 Table 150. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Overt Discrimi nation, DV = Physical Health) Men Women B SE t p B SE t p b(YX) -.10.12-.85.40-.19.09 -2.14 .03 b(MX) .50.133.87.00.41.10 4.02 .00 b(YM.X) -.28.08-3.59.00-.23.05 -4.33 .00 b(YX.M) .04.12.38.71-.09.09 -1.08 .28 Finally, perceived overt sexbased discrimination was inve stigated as a mediator in the relationship between minority segmen tation and psychologica l health. Support for overall mediation was found as the paths be tween minority segmentation and perceived discrimination ( B = .52, t = 3.77, p < .01) and between the mediator and psychological health ( B = -.21, t = -4.33, p < .01) were jointly significan t. Minority segmentation was related to greater perceived discriminati on which in turn was related to poorer psychological health. No significant inte raction was found between the predictor (minority segmentation) and the moderator (mi nority status). Thus, mi nority status is not a moderator in the current model. Further, th e indirect effects are significant for both men and women, and were similar in size. Results for the moderated mediation are shown in Table 151.

PAGE 284

249 Table 151. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Overt Discrimination, DV = Psychologi cal Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 7.051.71 4.12.00 MINORITY SEG .52.14 3.77.00 Minority Status 2.232.14 1.04.30 MINORITY SEG Minority -.13.17 -.75.46 Psychological Health Constant 55.421.69 32.78.00 Overt sex discrimination -.21.05 -4.33.00 MINORITY SEG -.13.14 -.98.33 Minority Status -.332.07 -.16.87 MINORITY SEG Minority -.06.16 -.37.71 Minority Status Indirect Effect SE z p Men -.11.04 -2.80.01 Women -.08.03 -2.86.00 A significant mediating effect was found for women. However, the model was not fully supported for men as the path fr om minority segmentation to perceived discrimination is not significant. Paths are not all in the hypothesi zed directions. Path coefficients were predicted to be differe nt in directions ba sed on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentati on and perceived discrimination, whereas majority group members would perceive a ne gative relationship. However, both groups demonstrated a positive relationship. Table 152 di splays the results for the simple tests of mediation.

PAGE 285

250 Table 152. Results of Simple Mediation for Men and Women (PREDICTOR = Minority Segmentation, Med = Overt Discrimina tion, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) -.24.13-1.83.07-.28.10 -2.79 .01 b(MX) .52.134.03.00.39.10 3.80 .00 b(YM.X) -.15.09-1.58.12-.24.06 -4.11 .00 b(YX.M) -.17.14-1.20.23-.18.10 -1.85 .07 Subtle Race-based Discrimination Perceived minority segmentation in the workplace was investigated as a predicto r in a model where subtle race-based discrimination mediates the relationship between minority segmentation and five dependent variables (job satis faction, organizational commitm ent, intention to turnover, physical health, and psychologi cal health). Minority stat us was investigated as a moderator in the relationship betwee n the predictor and the mediator. First, perceived subtle race-based discrimi nation was investigated as a mediator in the relationship between minority segmenta tion and job satisfaction. Support was not found for the overall model. The path betw een the mediator and job satisfaction ( B = -.12, t = -4.31, p < .01) was significant, but the path between minority segmentation and the mediator was not ( B = .13, t = 1.73, p > .05). No significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the cu rrent model. The indirect effects, however,

PAGE 286

251 indicate that there may be a difference be tween white and non-white participants. The indirect effects for non-whites participants were significant (Indirect effect = -.04, z = 2.83, p < .01), but this was not the case for white participants. Results for the moderated mediation are displayed in Table 153. Table 153. Results of Moderated Mediation (PREDICTO R = Minority Segmenta tion, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 8.98.94 9.60.00 MINORITY SEG .13.07 1.73.08 Minority Status .091.42 .06.95 MINORITY SEG Minority .18.11 1.59.11 Job Satisfaction Constant 13.50.60 22.52.00 Subtle race discrimination -.12.03 -4.31.00 MINORITY SEG -.12.04 -2.59.01 Minority Status -.19.82 -.23.82 MINORITY SEG Minority .06.06 .93.35 Minority Status Indirect Effect SE z p Whites -.02.01 -1.57.12 Non-whites -.04.01 -2.83.00 The results for the separate tests of medi ation indicate that the model is supported for both white and non-white participants. Path s are not all in the hypothesized directions. Path coefficients were predicted to be diffe rent in direction base d on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentati on and perceived discrimination, whereas

PAGE 287

252 majority group members would perceive a ne gative relationship. However, both groups demonstrated a positive relationship. Results are shown in Table 154. Table 154. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Subtle Discri mination, DV = Job Satisfaction) White Non-whites B SE t p B SE t p b(YX) -.13.05-2.91.00-.09.05 -1.98 .05 b(MX) .13.062.15.03.31.10 3.24 .00 b(YM.X) -.13.05-2.65.01-.12.04 -3.42 .00 b(YX.M) -.11.05-2.54.01-.06.05 -1.19 .24 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between minority segm entation and organizational commitment. Similar to the results for job satisfaction, there was not support for overall mediation. The path between minority segmentation and perc eived discrimination was not significant, but the path between the mediator and organizational commitment ( B = -.29, t = -3.65, p < .01) was significant. No significant inte raction was found between the predictor (minority segmentation) and the moderator (mi nority status). Thus, mi nority status is not a moderator in the current model. The indirect effects, however, indicate that there is a difference between white and non-white partic ipants. The indirect effect for non-white participants is significan t (Indirect effect = -.09, z = -2.59, p < .05), but this is not the case for white participants. Results are displayed in Table 155.

PAGE 288

253 Table 155. Results of Moderated Mediation (PREDICTO R = Minority Segmenta tion, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 8.91.92 9.65.00 MINORITY SEG .13.08 1.77.08 Minority Status .161.41 .11.91 MINORITY SEG Minority .17.11 1.55.12 Organizational Constant 35.921.63 21.99.00 Subtle race discrimination -.29.08 -3.65.00 MINORITY SEG -.31.12 -2.53.01 Minority Status 3.522.25 1.57.12 MINORITY SEG Minority -.04.18 -.24.81 Minority Status Indirect Effect SE z p Whites -.04.03 -1.55.12 Non-whites -.09.03 -2.59.01 The results for the separate tests of mediation indicate that there was a significant mediating effect was found for both white and n on-white participants. Paths are not all in the hypothesized directions. Path coefficients we re predicted to be di fferent in direction based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceived discrimination, whereas majority group member s would perceive a negative relationship. However, both groups demonstrated a positiv e relationship. Results are shown in Table 156.

PAGE 289

254 Table 156. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Subtle Discrimina tion, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) -.34.13-2.74.01-.44.12 -3.50 .00 b(MX) .13.062.25.03.30.10 3.17 .00 b(YM.X) -.30.14-2.10.04-.29.09 -3.06 .00 b(YX.M) -.30.13-2.42.02-.35.13 -2.78 .01 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between minority segmenta tion and intention to turnover. Support was not found for overall mediation. The path be tween minority segmentation and perceived discrimination was not significant, but the pa th between the mediator and intention to turnover was significant ( B = .08, t = 5.95, p < .01). No significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the curren t model. The indirect effects were different for white and non-white participants. The indi rect effects were si gnificant for non-white participants (Indirect effect = .03, z = 3.28, p < .01), but were not significant for white participants. Results of the moderate d mediation are shown in Table 157.

PAGE 290

255 Table 157. Results of Moderated Mediation (PREDICTO R = Intention to Turnover, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 8.93.94 9.54.00 MINORITY SEG .14.08 1.86.06 Minority Status -.001.42 -.00.99 MINORITY SEG Minority .18.11 1.61.11 Turnover Constant 1.31.28 4.74.00 Subtle race discrimination .08.01 5.95.00 MINORITY SEG .06.02 2.80.01 Minority Status .08.38 .21.83 MINORITY SEG Minority -.04.03 -1.43.15 Minority Status Indirect Effect SE z p Whites .01.01 1.75.08 Non-whites .03.01 3.28.00 The separate tests of mediat ion revealed full support fo r the mediating effect for both white and non-white participants. Paths are not all in the hypot hesized directions. Path coefficients were predicted to be diffe rent in direction base d on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentati on and perceived discrimination, whereas majority group members would perceive a ne gative relationship. However, both groups demonstrated a positive relationship. Results are shown in Table 158.

PAGE 291

256 Table 158. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Subtle Discrimina tion, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) .07.023.23.00.04.02 1.80 .07 b(MX) .14.062.28.02.32.10 3.37 .00 b(YM.X) .09.024.12.00.07.02 4.34 .00 b(YX.M) .06.022.68.01.02.02 .78 .44 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between minority segmentation and physical health. The overall model of mediation was not supported. The path be tween minority segmentation and perceived discrimination was not significant, but the path between the mediator and physical health was significant ( B = -.36, t = -5.42, p < .01). Although the path from minority segmentation to perceived discrimination wa s not significant, the interaction between minority segmentation and minority status in predicting the mediator was significant ( B = .30, t = 2.68, p < .05). Additionally, the indirect effect was significant for non-white participants only (Indi rect effect = -.14, z = -3.59, p < .01). Results are displayed in Table 159.

PAGE 292

257 Table 159. Results of Moderated Mediation (PREDICTO R = Minority Segmenta tion, Med = Subtle Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 9.24.90 10.27.00 MINORITY SEG .10.07 1.41.16 Minority Status -1.511.41 -1.07.29 MINORITY SEG Minority .30.11 2.68.01 Physical Health Constant 71.621.28 55.98.00 Subtle race discrimination -.36.07 -5.42.00 MINORITY SEG -.16.09 -1.74.08 Minority Status -1.961.77 -1.11.27 MINORITY SEG Minority .19.14 1.33.18 Minority Status Indirect Effect SE z p Whites -.04.03 -1.34.18 Non-whites -.14.04 -3.59.00 Results for the separate tests of media tion indicated a mediating effect for nonwhite participants but not for white particip ants. Paths are not al l in the hypothesized directions. Path coefficients were predicted to be different in dire ctions based on minority status. However, paths are in the same di rection for both majority and minority group members. Specifically, it wa s hypothesized that minority gr oup members would perceive a positive relationship between minority se gmentation and perceived discrimination, whereas majority group members would perceive a negative relations hip. However, both groups demonstrated a positive relationship. Ta ble 160 displays the results for the simple tests of mediation.

PAGE 293

258 Table 160. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Subtle Discri mination, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) -.20.09-2.21.03-.12.12 -1.03 .30 b(MX) .10.061.65.10.40.10 4.20 .00 b(YM.X) -.27.10-2.75.01-.42.09 -4.60 .00 b(YX.M) -.17.09-1.92.06.05.11 .42 .68 Finally, perceived subtle race -based discrimination was i nvestigated as a mediator in the relationship between minority segmen tation and psychologica l health. Support was not found for overall mediation. The pa th between minority segmentation and psychological health was not si gnificant, however the path between the mediator and psychological health was significant ( B = -.16, t = -2.19, p < .05). No significant interaction was found between the predictor (minority segm entation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are supportive of the lack of moderation. Th e indirect effects were not significant for either white or non-white part icipants. Results are shown in Table 161.

PAGE 294

259 Table 161. Results of Moderated Mediation (PREDICTO R = Minority Segmenta tion, Med = Subtle Discrimination, DV = Psychol ogical Health, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 8.99.97 9.30.00 MINORITY SEG .14.08 1.76.08 Minority Status .501.49 .33.74 MINORITY SEG Minority .14.12 1.21.23 Psychological Health Constant 54.171.48 36.61.00 Subtle race discrimination -.16.07 -2.19.03 MINORITY SEG -.20.11 -1.80.07 Minority Status 1.262.06 .61.54 MINORITY SEG Minority -.07.16 -.46.65 Minority Status Indirect Effect SE z p Whites -.02.02 -1.29.19 Non-whites -.04.02 -1.78.08 There was not a significant mediating effect for either white or non-white participants. Paths are not al l in the hypothesized directions. Path coefficients were predicted to be different in directions based on minority stat us. However, paths are in the same direction for both majority and mi nority group members. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceive d discrimination, whereas majority group members would perceive a nega tive relationship. However, both groups demonstrated a positive relationship. Results for the simple te sts of mediation are shown in Table 162.

PAGE 295

260 Table 162. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Subtle Discrimi nation, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) -.22.11-1.93.06-.32.11 -2.82 .01 b(MX) .14.072.16.03.28.10 2.79 .01 b(YM.X) -.18.12-1.51.13-.14.09 -1.64 .10 b(YX.M) -.19.12-1.69.09-.28.11 -2.42 .02 Overt Race-based Discrimination Perceived minority segmentation in the workplace was investigated as a predicto r in a model where overt race-based discrimination mediates the relationship between minority segmentation and five dependent variables (job satis faction, organizational commitm ent, intention to turnover, physical health, and psychologi cal health). Minority stat us was investigated as a moderator in the relationship betwee n the predictor and the mediator. First, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between minority segmentati on and job satisfaction. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .24, t = 2.28, p < .05) and between the mediator and job satisfaction ( B = -.10, t = -4.59, p < .01) were jointly si gnificant. Minority se gmentation was related to greater perceived discrimination which in turn was related to le ss job satisfaction. No significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current

PAGE 296

261 model. However, the interaction effect approached significance in this case ( B = .26, t = 1.69, p = .09). The indirect effects indicate a difference between groups. While the indirect effect for non-whites is significant (Indirect effect = -.05, z = -3.17, p < .01), the indirect effect for white participants is approaching significance. Results for the moderated mediation are shown in Table 163. Table 163. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Overt race discrimination Constant 8.991.27 7.10.00 MINORITY SEG .24.10 2.28.02 Minority Status -.831.95 -.42.67 MINORITY SEG Minority .26.15 1.69.09 Job Satisfaction Constant 13.29.57 23.23.00 Overt race discrimination -.10.02 -4.59.00 MINORITY SEG -.11.04 -2.50.01 Minority Status -.54.83 -.65.51 MINORITY SEG Minority .08.06 1.20.23 Minority Status Indirect Effect SE z p Whites -.02.01 -2.00.05 Non-whites -.05.02 -3.17.00 The results for the separate tests of medi ation indicated support for the mediation model for both white and non-white participan ts. Paths are not all in the hypothesized directions. Path coefficients were predicted to be different in dire ctions based on minority status. However, paths are in the same di rection for both majority and minority group members. Specifically, it wa s hypothesized that minority gr oup members would perceive a positive relationship between minority se gmentation and perceived discrimination,

PAGE 297

262 whereas majority group members would perceive a negative relations hip. However, both groups demonstrated a positive relationship. Ta ble 164 displays the results for the simple tests of mediation. Table 164. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Overt Discrimination, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) -.32.12-2.58.01-.08.05 -1.71 .09 b(MX) .24.092.55.01.49.12 4.10 .00 b(YM.X) -.30.09-3.47.00-.07.03 -2.38 .02 b(YX.M) -.25.12-2.02.04-.05.05 -.97 .33 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between minority segmentati on and organizational commitment. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .24, t = 2.32, p < .05) and between the mediator and organizational commitment ( B = -.24, t = -4.11, p < .01) were jointly significant. Minority segmentation was related to greater perceived discrimination which in turn was related to less organizational commitment. No significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. However, similar to the findings for job satisfaction, the interaction eff ect approached significance ( B = .25, t = 1.64, p = .10).

PAGE 298

263 The indirect effects indica te a difference between groups The indirect effect is significant for non-white particip ants (Indirect effect = -.11, z = -2.96, p < .01), but not so for white participants. Results are displayed in Table 165. Table 165. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Overt race discrimination Constant 8.891.26 7.06.00 MINORITY SEG .24.10 2.32.02 Minority Status -.701.96 -.36.72 MINORITY SEG Minority .25.15 1.64.10 Organizational Constant 35.351.54 22.97.00 Overt race discrimination -.24.06 -4.11.00 MINORITY SEG -.27.12 -2.22.03 Minority Status 3.112.25 1.38.17 MINORITY SEG Minority -.05.17 -.27.79 Minority Status Indirect Effect SE z p Whites -.06.03 -1.97.05 Non-whites -.11.04 -2.96.00 The results for the separate tests of mediation indicated full support for the models for both white and non-white particip ants. Paths are not al l in the hypothesized directions. Path coefficients were predicted to be different in dire ctions based on minority status. However, paths are in the same di rection for both majority and minority group members. Specifically, it wa s hypothesized that minority gr oup members would perceive a positive relationship between minority se gmentation and perceived discrimination, whereas majority group members would perceive a negative relations hip. However, both groups demonstrated a positive relationship. Results are shown in Table 166.

PAGE 299

264 Table 166. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Overt Discrimina tion, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) -.32.12-2.58.01-.43.12 -3.43 .00 b(MX) .24.092.55.01.49.12 3.98 .00 b(YM.X) -.30.09-3.47.00-.18.08 -2.39 .02 b(YX.M) -.25.12-2.02.04-.34.13 -2.65 .01 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between minority segmentati on and intention to turnover. Support for overall mediation was found as the paths be tween minority segmentation and perceived discrimination ( B = .26, t = 2.50, p < .05) and between the mediator and intention to turnover ( B = .05, t = 5.58, p < .01). Moreover, all paths were in the predicted direction. No significant interaction was found between the predictor (minority segmentation) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects were both signifi cant and similar in size and direction. Results are shown in Table 167.

PAGE 300

265 Table 167. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Overt race discrimination Constant 8.801.28 6.90.00 MINORITY SEG .26.10 2.50.01 Minority Status -.841.97 -.43.67 MINORITY SEG Minority .25.15 1.63.10 Turnover Constant 1.54.26 5.86.00 Overt race discrimination .05.01 5.58.00 MINORITY SEG .05.02 2.56.01 Minority Status .18.38 .47.64 MINORITY SEG Minority -.04.03 -1.36.17 Minority Status Indirect Effect SE z p Whites .01.01 2.26.02 Non-whites .03.01 3.50.00 A significant mediating effect wa s found for both white and non-white participants. Paths are not al l in the hypothesized directions. Path coefficients were predicted to be different in directions based on minority stat us. However, paths are in the same direction for both majority and mi nority group members. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceive d discrimination, whereas majority group members would perceive a nega tive relationship. However, both groups demonstrated a positive relationship. Results for the simple te sts of mediation are shown in Table 168.

PAGE 301

266 Table 168. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Overt Discrimination, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) .07.023.16.00.04.02 1.75 .08 b(MX) .26.102.68.01.51.12 4.25 .00 b(YM.X) .07.015.06.00.04.01 2.91 .00 b(YX.M) .05.022.38.02.02.02 .83 .41 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between minority segmentati on and physical health. Support for overall mediation was found as the paths between minority segmentation and perceived discrimination ( B = .22, t = 2.08, p < .05) and between the mediator and physical health ( B = -.31, t = -6.82, p < .01) were jointly si gnificant. Minority se gmentation was related to greater perceived discrimination which in turn was related to poorer physical health. There was a significant interaction was found between the predictor (minority segmentation) and the moderator ( B = .39, t = 2.46, p < .05). Thus, minority status is a moderator in the current model. The indirect effects are supportive of the interaction effect. The indirect effects are significant for non-white participants, but not so for white participants. Results are shown in Table 169.

PAGE 302

267 Table 169. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Overt Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 9.171.28 7.15.00 MINORITY SEG .22.11 2.08.04 Minority Status -2.712.05 -1.32.19 MINORITY SEG Minority .39.16 2.46.01 Physical Health Constant 70.871.18 60.21.00 Overt race discrimination -.31.04 -6.82.00 MINORITY SEG -.10.09 -1.11.27 Minority Status -2.161.77 -1.23.22 MINORITY SEG Minority .18.14 1.33.18 Minority Status Indirect Effect SE z p Whites -.07.03 -1.97.05 Non-whites -.19.05 -4.08.00 The results for the separate tests of mediation indicated full support of the mediation model for both white and non-white participants. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority group members would perceive a positive relationship between minority segmentation and perceived discrimination, whereas majority group member s would perceive a negative relationship. However, both groups demonstrated a positiv e relationship. Results are shown in Table 170.

PAGE 303

268 Table 170. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Overt Discrimi nation, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) -.17.09-1.88.06-.11.12 -.91 .37 b(MX) .22.102.19.03.61.13 4.84 .00 b(YM.X) -.28.06-4.65.00-.34.07 -4.92 .00 b(YX.M) -.11.09-1.24.22.10.12 .88 .38 Finally, perceived overt race-based discrimi nation was investigated as a mediator in the relationship between minority segmen tation and psychologica l health. Support for overall mediation was found as the paths be tween minority segmentation and perceived discrimination ( B = .27, t = 2.48, p < .05) and between the mediator and psychological health ( B = -.17, t = -3.40, p < .01) were jointly significan t. Minority segmentation was related to greater perceived discriminati on which in turn was related to poorer psychological health. No significant inte raction was found between the predictor (minority segmentation) and the moderator (mi nority status). Thus, mi nority status is not a moderator in the current model. The indir ect effects were different for white and nonwhite participants, however. Non-white particip ants exhibited a significant indirect effect (Indirect effect = -.07, z = -2.49, p < .05), whereas white participants did not. Results for the moderated mediation are shown in Table 171.

PAGE 304

269 Table 171. Results of Moderated Mediation (PREDIC TOR = Minority Segmen tation, Med = Overt Discrimination, DV = Psychol ogical Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 8.811.32 6.69.00 MINORITY SEG .27.11 2.48.01 Minority Status .012.07 .00.99 MINORITY SEG Minority .18.16 1.12.26 Psychological Health Constant 54.601.39 39.28.00 Overt race discrimination -.17.05 -3.40.00 MINORITY SEG -.18.11 -1.70.09 Minority Status .932.07 .45.65 MINORITY SEG Minority -.06.16 -.36.72 Minority Status Indirect Effect SE z p Whites -.05.02 -1.94.05 Non-whites -.07.03 -2.49.01 Both white and non-white participants exhibited support for the mediation model in the simple test of mediation. Paths are not all in the hypothesi zed directions. Path coefficients were predicted to be differe nt in directions ba sed on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it was hypothesized that minority group members would perceive a positive relationship between minority segmentati on and perceived discrimination, whereas majority group members would perceive a ne gative relationship. However, both groups demonstrated a positive relationship. Results are displayed in Table 172.

PAGE 305

270 Table 172. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Minority Segmentation, Med = Overt Discrimina tion, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) -.23.11-2.06.04-.32.12 -2.79 .01 b(MX) .27.102.66.01.45.13 3.50 .00 b(YM.X) -.17.08-2.28.02-.18.07 -2.56 .01 b(YX.M) -.19.11-1.64.10-.24.12 -2.07 .04 Results for Diversity Climate Subtle Sex-based Discrimination Diversity climate in the workplace was investigated as a predictor in a model where subtle sex-based discri mination mediates the relationship between diversity climate and five dependent variab les (job satisfaction, organizational commitment, intention to tu rnover, physical hea lth, and psychological health). Minority status was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived subtle sex-based discrimi nation was investigated as a mediator in the relationship between diversity climat e and job satisfaction. Support for overall mediation was found as the paths between dive rsity climate and perceived discrimination ( B = -.26, t = -4.12, p < .01) and between the medi ator and job satisfaction ( B = -.08, t = -2.71, p < .05) were jointly signifi cant. Diversity climate was related to less perceived discrimination which in turn was related to greater job satisfaction. No significant

PAGE 306

271 interaction was found between the predictor (diversity cl imate) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are significant for both men and women, and were similar in size. Results are shown in Table 173. Table 173. Results of Moderated Mediation (PREDIC TOR = Diversity Climate, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 19.782.08 9.49.00 DIVERSITY CLIM -.26.06 -4.12.00 Minority Status .962.44 .39.70 DIVERSITY CLIM Minority .01.07 .16.87 Job satisfaction Constant 7.421.39 5.32.00 Subtle sex discrimination -.08.03 -2.71.01 DIVERSITY CLIM .13.04 3.48.00 Minority Status .901.47 .61.54 DIVERSITY CLIM Minority -.02.04 -.46.65 Minority Status Indirect Effect SE z p Men .02.01 2.22.03 Women .02.01 2.47.01 The mediation model was not fully suppor ted for either men or women. Although the path from the perceived discrimination to job satisfaction wa s significant for both men and women, the path from diversity clim ate to the mediator was only approaching significance. Paths are not all in the hypothesized directions Path coefficients were predicted to be different in directions based on minority stat us. However, paths are in the same direction for both majority and mi nority group members. Specifically, it was hypothesized that minority individuals woul d report a negative relationship between

PAGE 307

272 diversity climate and perceived discriminati on whereas majority i ndividuals would report a positive relationship. However, the relatio nship was negative for both groups. Results for the simple tests of medi ation are shown in Table 174. Table 174. Results of Simple Mediation for Men and Women (PREDICTOR = Diversity Climate, Med = Subtle Discrimination, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .15.043.99.00.13.02 5.76 .00 b(MX) -.26.06-4.43.00-.25.04 -6.17 .00 b(YM.X) -.12.06-1.99.05-.07.04 -1.95 .05 b(YX.M) .12.043.01.00.12.02 4.74 .00 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between diversity climate and organizational commitment. Support for overall mediation was not found. Although th e path between diversity climate and perceived discrimination ( B = -.25, t = -4.04, p < .01) was significant, the path between perceived discrimination and organizationa l commitment was not. No significant interaction was found between the predictor (diversity cl imate) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Additionally, the indirect eff ects were not significant for either men or women. Results are displayed in Table 175.

PAGE 308

273 Table 175. Results of Moderated Mediation (PREDIC TOR = Diversity Climate, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 19.262.04 9.42.00 DIVERSITY CLIM -.25.06 -4.04.00 Minority Status 1.072.39 .45.65 DIVERSITY CLIM Minority .01.07 .17.86 Organizational Constant 18.683.55 5.26.00 Subtle sex discrimination -.14.08 -1.72.09 DIVERSITY CLIM .38.10 3.84.00 Minority Status -3.263.74 -.87.38 DIVERSITY CLIM Minority .15.11 1.34.18 Minority Status Indirect Effect SE z p Men .03.02 1.54.12 Women .03.02 1.64.10 The mediation model was not supported for men or women. Thus, subtle sexbased discrimination does not mediate the re lationship between diversity climate and organizational commitment. Paths are not all in the hypothesized directions. Path coefficients were predicted to be differe nt in directions ba sed on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it was hypothesized that minor ity individuals woul d report a negative relationship between diversity climate and perceived discrimination whereas majority individuals would report a posi tive relationship. However, th e relationship was negative for both groups. Results are shown in Table 176.

PAGE 309

274 Table 176. Results of Simple Mediation for Men and Women (PREDICTOR = Diversity Climate, Med = Subtle Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) .41 .094.36.00.56.069.40 .00 b(MX) -.25 .05-4.59.00-.23.04-5.97 .00 b(YM.X) -.09 .16-.55.58-.15.09-1.65 .10 b(YX.M) .39 .103.80.00.52.068.30 .00 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between diversity climate a nd intention to turnover. Support for overall mediation was found as the paths between dive rsity climate and perceived discrimination ( B = -.26, t = -4.09, p < .01) and between the mediator and intention to turnover ( B = .05, t = 3.48, p < .01) were jointly significant. Diversit y climate was related to less perceived discrimination which in turn was related to less turnover intention. No significant interaction was found between the predictor (diversity cl imate) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are significant for both men and women, and were similar in size. Results for the moderated mediation are shown in Table 177.

PAGE 310

275 Table 177. Results of Moderated Mediation (PREDIC TOR = Diversity Climate, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 19.762.09 9.47.00 DIVERSITY CLIM -.26.06 -4.09.00 Minority Status .692.43 .28.78 DIVERSITY CLIM Minority .02.07 .25.80 Turnover Constant 3.70.66 5.60.00 Subtle sex discrimination .05.01 3.48.00 DIVERSITY CLIM -.05.02 -2.78.01 Minority Status -.19.70 -.27.79 DIVERSITY CLIM Minority .01.02 .30.77 Minority Status Indirect Effect SE z p Men -.01.00 -2.61.01 Women -.01.00 -3.01.00 The results for the separate tests of mediation are shown in Table 178. A significant mediating effect was found for bot h men and women. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. For example, minority individuals were hypot hesized to report a positive relationship between minority segm entation and perceived discrimination, whereas majority participants should report a negative relationship. However, paths are in the same direction for both majo rity and minority group members.

PAGE 311

276 Table 178. Results of Simple Mediation for Men and Women (PREDICTOR = Diversity Climate, Med = Subtle Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) -.06.02-3.72.00-.06.01 -5.01 .00 b(MX) -.26.06-4.40.00-.24.04 -6.08 .00 b(YM.X) .06.032.47.01.04.02 2.59 .01 b(YX.M) -.05.02-2.61.01-.05.01 -3.85 .00 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between diversity climat e and physical health. Support for overall mediation was found as the paths between dive rsity climate and perceived discrimination ( B = -.25, t = -3.90, p < .01) and between the mediat or and physical health ( B = -.26, t = 3.93, p < .01) were jointly significant. Diversit y climate was related to less perceived discrimination which in turn was related to better physical health. No significant interaction was found between the predictor (diversity cl imate) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Additionally, the indirect effects are si gnificant for both men and women, and were similar in size. Results ar e displayed in Table 179.

PAGE 312

277 Table 179. Results of Moderated Mediation (PREDIC TOR = Diversity Climate, Med = Subtle Discrimination, DV = Physical Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 19.452.12 9.18.00 DIVERSITY CLIM -.25.06 -3.90.00 Minority Status .532.47 .22.83 DIVERSITY CLIM Minority .02.07 .24.81 Physical Health Constant 66.622.95 22.57.00 Subtle sex discrimination -.26.07 -3.93.00 DIVERSITY CLIM .10.08 1.24.21 Minority Status -3.273.09 -1.06.29 DIVERSITY CLIM Minority .06.09 .70.49 Minority Status Indirect Effect SE z p Men .06.02 2.73.01 Women .06.02 3.24.00 The results for the separate tests of me diation are shown in Table 180. Support for the mediation model was found for women, but not for men. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority individuals would report a negative rela tionship between diversity climate and perceived discrimination whereas majority individuals would report a positive relationship. However, the relationship was negative for both groups.

PAGE 313

278 Table 180. Results of Simple Mediation for Men and Women (PREDICTOR = Diversity Climate, Med = Subtle Discrimination, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .16.082.16.03.22.05 4.47 .00 b(MX) -.25.06-4.07.00-.23.04 -5.84 .00 b(YM.X) -.19.12-1.58.12-.29.08 -3.62 .00 b(YX.M) .12.081.45.15.16.05 3.02 .00 Finally, perceived subtle sex-based discri mination was investigated as a mediator in the relationship between diversity climate and psyc hological health. Support for overall mediation was found as the paths between diversity climate and perceived discrimination ( B = -.22, t = -3.58, p < .01) and between the me diator and psychological health ( B = -.18, t = -2.42, p < .05) were jointly significant. Diversity climate was related to less perceived discrimination which in turn was related to better psychological health. No significant interaction was found between the predictor (divers ity climate) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are signifi cant for women, but were only approaching significance for the men. Results are shown in Table 181.

PAGE 314

279 Table 181. Results of Moderated Mediation (PREDIC TOR = Diversity Climate, Med = Subtle Discrimination, DV = Psychologi cal Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 18.492.10 8.81.00 DIVERSITY CLIM -.22.06 -3.58.00 Minority Status 2.162.46 .88.38 DIVERSITY CLIM Minority -.02.07 -.26.80 Psychological Health Constant 49.793.39 14.68.00 Subtle sex discrimination -.18.08 -2.42.02 DIVERSITY CLIM .10.09 1.10.27 Minority Status -6.653.61 -1.84.07 DIVERSITY CLIM Minority .19.11 1.72.09 Minority Status Indirect Effect SE z p Men .04.02 1.95.05 Women .04.02 2.23.03 A significant mediating effect was f ound for women, but the mediation model was not supported for men. Paths are not all in the hypothesized directions. Path coefficients were predicted to be differe nt in directions ba sed on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it was hypothesized that minor ity individuals woul d report a negative relationship between diversity climate and perceived discrimination whereas majority individuals would report a posi tive relationship. However, th e relationship was negative for both groups. Results for the simple te sts of mediation are shown in Table 182.

PAGE 315

280 Table 182. Results of Simple Mediation for Men and Women (PREDICTOR = Diversity Climate, Med = Subtle Discrimination, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) .14.101.49.14.33.06 5.95 .00 b(MX) -.22.05-4.15.00-.24.04 -5.88 .00 b(YM.X) -.07.17-.41.68-.22.08 -2.60 .01 b(YX.M) .13.101.24.22.28.06 4.73 .00 Overt Sex-based Discrimination Diversity climate in the workplace was investigated as a predictor in a model wher e overt sex-based discri mination mediates the relationship between diversity climate and five dependent variab les (job satisfaction, organizational commitment, intention to tu rnover, physical hea lth, and psychological health). Minority status was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between diversity climat e and job satisfaction. Support for overall mediation was found as the paths between dive rsity climate and perceived discrimination ( B = -.43, t = -4.85, p < .01) and between the medi ator and job satisfaction ( B = -.06, t = 3.07, p < .01) were jointly significant. Diversit y climate was related to less perceived discrimination which in turn was related to greater job satisfaction. No significant interaction was found between the predictor (diversity cl imate) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The

PAGE 316

281 indirect effects are significant for both men and women, and were similar in size. Results are shown in Table 183. Table 183. Results of Moderated Mediation (PREDIC TOR = Diversity Clim ate, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 27.042.91 9.30.00 DIVERSITY CLIM -.43.09 -4.85.00 Minority Status -2.533.46 -.73.46 DIVERSITY CLIM Minority .10.10 .93.35 Job Satisfaction Constant 7.291.27 5.72.00 Overt sex discrimination -.06.02 -3.07.00 DIVERSITY CLIM .13.04 3.73.00 Minority Status 1.391.38 1.02.31 DIVERSITY CLIM Minority -.03.04 -.82.41 Minority Status Indirect Effect SE z p Men .03.01 2.55.01 Women .02.01 2.68.01 The results for the separate tests of me diation are shown in Table 184. Support for the mediation model was found for men, but the model was not fully supported for women. Paths are not all in the hypothesized directions. Path coefficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it was hypothesized that minority individuals woul d report a negative relationship between diversity climate and perceived discriminati on whereas majority i ndividuals would report a positive relationship. However, the re lationship was negative for both groups.

PAGE 317

282 Table 184. Results of Simple Mediation for Men and Women (PREDICTOR = Diversity Climate, Med = Overt Discrimination, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .16.044.38.00.12.02 5.37 .00 b(MX) -.43.08-5.05.00-.33.06 -5.72 .00 b(YM.X) -.11.04-3.03.00-.04.02 -1.69 .09 b(YX.M) .11.042.88.00.11.02 4.54 .00 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between diversity climate and organizational commitment. Support for overall mediation was found as the paths between diversity climate and perceived discrimination ( B = -.42, t = -4.81, p < .01) and between the me diator and organizational commitment ( B = -.12, t = -2.30, p < .05) were jointly signific ant. Diversity climate was related to less perceived discrimination which in turn was related to greater organizational commitment. No significant interaction was found be tween the predictor (diversity climate) and the moderator (minor ity status). Thus, minority status is not a moderator in the current model. Further, the indirect effects are significant for both men and women, and were similar in si ze. Results are shown in Table 185.

PAGE 318

283 Table 185. Results of Moderated Mediation (PREDIC TOR = Diversity Clim ate, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 26.602.89 9.22.00 DIVERSITY CLIM -.42.09 -4.81.00 Minority Status -2.243.42 -.65.51 DIVERSITY CLIM Minority .09.10 .89.38 Organizational Constant 18.333.30 5.55.00 Overt sex discrimination -.12.05 -2.30.02 DIVERSITY CLIM .39.09 4.20.00 Minority Status -2.013.56 -.57.57 DIVERSITY CLIM Minority .11.11 1.04.30 Minority Status Indirect Effect SE z p Men .05.02 2.04.04 Women .04.02 2.12.03 The results for the separate tests of me diation indicate that that the mediation model was not fully supported for either me n or women. In both cases, the path from perceived discrimination to organizational co mmitment was significant, but the path from diversity climate to perceived discrimi nation was not. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority individuals would report a negative rela tionship between diversity climate and perceived discrimination whereas majority individuals would report a positive relationship. However, the relationship was negative fo r both groups. Results are shown in Table 186.

PAGE 319

284 Table 186. Results of Simple Mediation for Men and Women (PREDICTOR = Diversity Climate, Med = Overt Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) .44.094.92.00.54.06 9.26 .00 b(MX) -.42.08-5.07.00-.33.06 -5.73 .00 b(YM.X) -.12.10-1.26.21-.12.06 -1.92 .06 b(YX.M) .39.103.96.00.50.06 8.17 .00 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between diversity climate a nd intention to turnover. Support for overall mediation was found as the paths between dive rsity climate and perceived discrimination ( B = -.43, t = -4.88, p < .01) and between the mediator and intention to turnover ( B = .04, t = 4.04, p < .01) were jointly significant. Diversit y climate was related to less perceived discrimination which in turn was related to less turnover intention. No significant interaction was found between the predictor (diversity cl imate) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are significant for both men and women, and were similar in size. Results for the moderated mediation are shown in Table 187.

PAGE 320

285 Table 187. Results of Moderated Mediation (PREDIC TOR = Diversity Clim ate, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 27.042.89 9.36.00 DIVERSITY CLIM -.43.09 -4.88.00 Minority Status -2.573.41 -.75.45 DIVERSITY CLIM Minority .10.10 .94.35 Turnover Constant 3.83.61 6.33.00 Overt sex discrimination .04.01 4.04.00 DIVERSITY CLIM -.05.02 -3.11.00 Minority Status -.35.65 -.54.59 DIVERSITY CLIM Minority .01.02 .58.57 Minority Status Indirect Effect SE z p Men -.02.01 -3.08.00 Women -.01.00 -3.31.00 The results for the separate tests of mediation are shown in Table 188. A significant mediating effect was found for bot h men and women. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. For example, th e path between minority segmentation and perceived discrimination was predicted to be positive for minority respondents and negative for majority respondents. It was found to be negative for both.

PAGE 321

286 Table 188. Results of Simple Mediation for Men and Women (PREDICTOR = Diversity Climate, Med = Overt Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) -.07.02-4.31.00-.05.01 -4.96 .00 b(MX) -.43.08-5.05.00-.33.06 -5.85 .00 b(YM.X) .04.022.55.01.04.01 3.17 .00 b(YX.M) -.05.02-2.97.00-.04.01 -3.70 .00 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between diversity climat e and physical health. Support for overall mediation was found as the paths between dive rsity climate and perceived discrimination ( B = -.36, t = -3.81, p < .01) and between the mediat or and physical health ( B = -.20, t = 4.46, p < .01) were jointly significant. Diversit y climate was related to less perceived discrimination which in turn was related to better physical health. No significant interaction was found between the predictor (diversity cl imate) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are significant for both men and women, and were similar in size. Results are shown in Table 189.

PAGE 322

287 Table 189. Results of Moderated Mediation (PREDIC TOR = Diversity Clim ate, Med = Overt Discrimination, DV = Physical Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 24.753.18 7.80.00 DIVERSITY CLIM -.36.09 -3.81.00 Minority Status -1.033.70 -.28.78 DIVERSITY CLIM Minority .05.11 .47.64 Physical Health Constant 65.452.88 22.71.00 Overt sex discrimination -.20.04 -4.46.00 DIVERSITY CLIM .12.08 1.43.15 Minority Status -2.423.10 -.78.44 DIVERSITY CLIM Minority .04.09 .43.67 Minority Status Indirect Effect SE z p Men .07.03 2.86.00 Women .06.02 3.40.00 A significant mediating effect was found for both men and women. Paths are not all in the hypothesized directi ons. Path coefficients were pr edicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Speci fically, it was hypothesized that minority individuals would report a nega tive relationship between dive rsity climate and perceived discrimination whereas majority individuals would report a positive relationship. However, the relationship was negative for both groups. Results for the mediation model are shown in Table 190.

PAGE 323

288 Table 190. Results of Simple Mediation for Men and Women (PREDICTOR = Diversity Climate, Med = Overt Discrimination, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .19.082.39.02.22.05 4.35 .00 b(MX) -.36.09-4.02.00-.31.06 -5.27 .00 b(YM.X) -.28.08-3.47.00-.17.05 -3.14 .00 b(YX.M) .09.081.10.27.17.05 3.19 .00 Finally, perceived overt sexbased discrimination was inve stigated as a mediator in the relationship between diversity climate and psyc hological health. Support for overall mediation was found as the paths between diversity climate and perceived discrimination ( B = -.41, t = -4.55, p < .01) and between the me diator and psychological health ( B = -.17, t = -3.35, p < .01) were jointly significant. Diversity climate was related to less perceived discrimination which in turn was related to better psychological health. No significant interaction was found between the predictor (divers ity climate) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Additionally, the indir ect effects are significant for both men and women, and were similar in size. Results are displayed in Table 191.

PAGE 324

289 Table 191. Results of Moderated Mediation (PREDIC TOR = Diversity Clim ate, Med = Overt Discrimination, DV = Psychologi cal Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 26.232.98 8.81.00 DIVERSITY CLIM -.41.09 -4.55.00 Minority Status -1.803.53 -.51.61 DIVERSITY CLIM Minority .08.11 .75.45 Psychological Health Constant 51.003.16 16.13.00 Overt sex discrimination -.17.05 -3.35.00 DIVERSITY CLIM .07.09 .80.42 Minority Status -6.613.42 -1.93.05 DIVERSITY CLIM Minority .18.10 1.72.09 Minority Status Indirect Effect SE z p Men .07.03 2.65.01 Women .05.02 2.84.00 The results for the separate tests of mediation are shown in Table 192. A significant mediating effect was found for wo men, but the model was not fully supported for men. Paths are not all in th e hypothesized directions. Path coefficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it was hypothesized that minority individuals woul d report a negative relationship between diversity climate and perceived discriminati on whereas majority i ndividuals would report a positive relationship. However, the re lationship was negative for both groups.

PAGE 325

290 Table 192. Results of Simple Mediation for Men and Women (PREDICTOR = Diversity Climate, Med = Overt Discrimination, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) .14.091.55.12.30.06 5.39 .00 b(MX) -.41.08-4.87.00-.33.06 -5.43 .00 b(YM.X) -.15.10-1.49.14-.18.06 -3.03 .00 b(YX.M) .08.10.81.42.25.06 4.20 .00 Subtle Race-based Discrimination Diversity climate in the workplace was investigated as a predictor in a model wher e subtle race-based discrimination mediates the relationship between diversity climate and five dependent variab les (job satisfaction, organizational commitment, intention to tu rnover, physical hea lth, and psychological health). Minority status was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived subtle race-based discrimi nation was investigated as a mediator in the relationship between diversity climat e and job satisfaction. Support for overall mediation was found as the paths between dive rsity climate and perceived discrimination ( B = -.15, t = -3.39, p < .01) and between the medi ator and job satisfaction ( B = -.08, t = 2.58, p < .05) were jointly significant. Diversit y climate was related to less perceived discrimination which in turn was related to greater job satisfaction. A significant interaction was found between the predictor (minority segm entation) and the minority

PAGE 326

291 status ( B = -.21, t = -3.26, p < .00). Thus, minority status does serve as a moderator in the current model. The indirect effects were diffe rent between groups. The indirect effect is significant for non-white participants and nonsignificant for white participants, although the indirect effect in this case approach ed significance. Results for the moderated mediation are shown in Table 193. Table 193. Results of Moderated Mediation (PREDIC TOR = Diversity Climate, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 15.281.47 10.40.00 DIVERSITY CLIM -.15.04 -3.39.00 Minority Status 9.852.19 4.50.00 DIVERSITY CLIM Minority -.21.07 -3.26.00 Job Satisfaction Constant 8.10.97 8.33.00 Subtle race discrimination -.08.03 -2.58.01 DIVERSITY CLIM .11.03 4.23.00 Minority Status -.221.32 -.16.87 DIVERSITY CLIM Minority .01.04 .32.75 Minority Status Indirect Effect SE z p Whites .01.01 2.00.05 Non-whites .03.01 2.42.02 The results for the separate tests of mediation are shown in Table 194. A significant mediating effect was found for non-white participants, but the mediation model was not supported for white participan ts. Paths are not all in the hypothesized directions. Path coefficients were predicted to be different in dire ctions based on minority status. However, paths are in the same di rection for both majority and minority group members. Specifically, it wa s hypothesized that minority individuals would report a

PAGE 327

292 negative relationship between diversity climate and percei ved discrimination whereas majority individuals would report a positive relationship. However, the relationship was negative for both groups. Table 194. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Diversity Climate, Med = Subtle Discrimi nation, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) .12.034.60.00.15.03 5.54 .00 b(MX) -.15.04-4.14.00-.37.06 -6.42 .00 b(YM.X) -.07.05-1.49.14-.08.04 -2.15 .03 b(YX.M) .11.034.04.00.12.03 4.11 .00 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between di versity climate and organiza tional commitment. Support for overall mediation was not found as the path between diversity climate and perceived discrimination ( B = -.15, t = -3.48, p < .01) was significant bu t the path between the mediator and organizational commitment was not. A significant in teraction was found between the predictor (diversity climate) and minority status ( B = -.20, t = -3.05, p < .01). Thus, minority status is a moderator in the cu rrent model. The indirect effects, however, were non-significant and similar in si ze. Results are shown in Table 195.

PAGE 328

293 Table 195. Results of Moderated Mediation (PREDIC TOR = Diversity Climate, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 15.161.43 10.63.00 DIVERSITY CLIM -.15.04 -3.48.00 Minority Status 9.272.14 4.33.00 DIVERSITY CLIM Minority -.20.06 -3.05.00 Organizational Constant 15.562.48 6.28.00 Subtle race discrimination -.08.08 -1.10.27 DIVERSITY CLIM .47.07 6.87.00 Minority Status -1.373.35 -.41.68 DIVERSITY CLIM Minority .09.10 .89.37 Minority Status Indirect Effect SE z p Whites .01.01 1.01.31 Non-whites .03.03 1.08.28 The results for the separate tests of me diation are shown in Table 196. Findings from both white and non-white participants failed to suppor t the mediation model. Paths are not all in the hypothesized di rections. Path coefficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Speci fically, it was hypothesized that minority individuals would report a nega tive relationship between dive rsity climate and perceived discrimination whereas majority individuals would report a positive relationship. However, the relationship was negative for both groups.

PAGE 329

294 Table 196. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Diversity Climate, Med = Subtle Discriminati on, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) .48.076.76.00.58.07 8.76 .00 b(MX) -.15.03-4.39.00-.35.06 -6.08 .00 b(YM.X) -.06.14-.43.67-.10.09 -1.12 .27 b(YX.M) .47.076.35.00.55.07 7.50 .00 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between diversity climate and inten tion to turnover. Support for overall mediation was found as the paths between diversity climate and perceived discrimination ( B = -.16, t = -3.55, p < .01) and between the me diator and intention to turnover ( B = .07, t = 4.79, p < .01) were jointly significant. Diversity climate was related to less perceived discrimination which in turn was related to less turnover intention. A significant interaction was found between the predictor (diversity climate) and the moderator ( B = -.20, t = -3.01, p < .01). Thus, minority status is a moderator in the current model. The indirect effects, howeve r, were significant for both white and nonwhite participants, and were similar in size. Results are shown in Table 197.

PAGE 330

295 Table 197. Results of Moderated Mediation (PREDIC TOR = Diversity Climate, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 15.511.45 10.70.00 DIVERSITY CLIM -.16.04 -3.55.00 Minority Status 9.192.16 4.24.00 DIVERSITY CLIM Minority -.20.07 -3.01.00 Turnover Constant 3.83.45 8.52.00 Subtle race discrimination .07.01 4.79.00 DIVERSITY CLIM -.06.01 -4.47.00 Minority Status -1.09.60 -1.82.07 DIVERSITY CLIM Minority .02.02 1.33.18 Minority Status Indirect Effect SE z p Whites -.01.00 -2.81.00 Non-whites -.02.01 -3.99.00 The results for the separate tests of mediation are shown in Table 198. A significant mediating effect was found for bot h white and non-white participants. Paths are not all in the hypothesized di rections. Path coefficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Fo r example, the path between minority segmentation and perceived discrimination wa s hypothesized to be positive for minority participants and negative for majority participants. Results indicated a negative relationship for both groups.

PAGE 331

296 Table 198. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Diversity Climate, Med = Subtle Discriminati on, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) -.07.01-5.26.00-.05.01 -4.04 .00 b(MX) -.16.04-4.31.00-.35.06 -6.28 .00 b(YM.X) .07.023.07.00.06.02 3.68 .00 b(YX.M) -.05.01-4.29.00-.03.01 -2.21 .03 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between diversity climate and physical health. Support for overall mediation was found as the paths between dive rsity climate and perceived discrimination ( B = -.14, t = -3.30, p < .01) and between the mediat or and physical health ( B = -.31, t = 4.58, p < .01) were jointly significant. Diversit y climate was related to less perceived discrimination which in turn was related to better physical health. A significant interaction was found between the predictor (diversity cl imate) and the moderator ( B = .18, t = -2.77, p < .05). Thus, minority status is a m oderator in the current model. The indirect effects are significan t for both white and non-white pa rticipants, and were similar in size. Results are displayed in Table 199.

PAGE 332

297 Table 199. Results of Moderated Mediation (PREDIC TOR = Diversity Climate, Med = Subtle Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 14.961.42 10.52.00 DIVERSITY CLIM -.14.04 -3.30.00 Minority Status 8.682.20 3.94.00 DIVERSITY CLIM Minority -.18.06 -2.77.01 Physical Health Constant 63.782.08 30.67.00 Subtle race discrimination -.31.07 -4.58.00 DIVERSITY CLIM .17.06 3.08.00 Minority Status 2.642.87 .92.36 DIVERSITY CLIM Minority -.08.09 -1.00.32 Minority Status Indirect Effect SE z p Whites .04.02 2.64.01 Non-whites .10.03 3.72.00 The results for the separate tests of mediation are shown in Table 200. A significant mediating effect was found for bot h white and non-white participants. Paths are not all in the hypothesized di rections. Path coefficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Speci fically, it was hypothesized that minority individuals would report a nega tive relationship between dive rsity climate and perceived discrimination whereas majority individuals would report a positive relationship. However, the relationship was negative for both groups.

PAGE 333

298 Table 200. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Diversity Climate, Med = Subtle Discrimi nation, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) .22 .054.20.00.19.072.62 .01 b(MX) -.14 .04-3.90.00-.33.06-5.60 .00 b(YM.X) -.24 .10-2.41.02-.36.10-3.75 .00 b(YX.M) .18 .053.46.00.07.08.93 .35 Finally, perceived subtle race -based discrimination was i nvestigated as a mediator in the relationship between diversity climate and psyc hological health. Support for overall mediation was not found, as the path between diversity climate and perceived discrimination was significant ( B = -.16, t = -3.52, p < .01), but the path between the mediator and psychological health was not significant. A significant interaction was found between the predictor (minority segmentation) and minority status ( B = -.21, t = 3.14, p < .01). Thus, minority status is a moderator in the current model. The indirect effects were similar in size and non-significan t for both white and nonwhite participants. Results are shown in Table 201.

PAGE 334

299 Table 201. Results of Moderated Mediation (PREDIC TOR = Diversity Climate, Med = Subtle Discrimination, DV = Psychol ogical Health, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 15.601.47 10.58.00 DIVERSITY CLIM -.16.05 -3.52.00 Minority Status 9.762.25 4.34.00 DIVERSITY CLIM Minority -.21.07 -3.14.00 Psychological Health Constant 40.542.39 16.99.00 Subtle race discrimination -.08.07 -1.02.31 DIVERSITY CLIM .34.07 5.14.00 Minority Status 6.443.27 1.97.05 DIVERSITY CLIM Minority -.21.10 -2.13.03 Minority Status Indirect Effect SE z p Whites .01.01 .95.34 Non-whites .03.03 1.01.31 The results for the separate tests of me diation are shown in Table 202. Findings from both white and non-white participants indicated no suppo rt for the mediation model. Paths are not all in the hypothe sized directions. Path coeffici ents were predicted to be different in directions based on minority status. However, path s are in the same direction for both majority and minority group member s. Specifically, it was hypothesized that minority individuals would re port a negative relationship be tween diversity climate and perceived discrimination whereas majori ty individuals would report a positive relationship. However, the relatio nship was negative for both groups.

PAGE 335

300 Table 202. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Diversity Climate, Med = Subtle Discrimina tion, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) .35 .065.44.00.15.072.14 .03 b(MX) -.16 .04-4.22.00-.37.06-6.21 .00 b(YM.X) .03 .12.25.80-.14.09-1.52 .13 b(YX.M) .35 .075.28.00.10.081.26 .21 Overt Race-based Discrimination Diversity climate in the workplace was investigated as a predictor in a model where overt race-based discrimination mediates the relationship between diversity climate and five dependent variab les (job satisfaction, organizational commitment, intention to tu rnover, physical hea lth, and psychological health). Minority status was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between diversity climat e and job satisfaction. Support for overall mediation was found as the paths between dive rsity climate and perceived discrimination ( B = -.29, t = -4.86, p < .01) and between the medi ator and job satisfaction ( B = -.07, t = 3.07, p < .01) were jointly significant. Diversit y climate was related to less perceived discrimination which in turn was related to greater job satisfaction. A significant interaction was found between the predictor (diversity cl imate) and the moderator ( B = .22, t = -2.47, p < .05). Thus, minority status is a m oderator in the current model. The

PAGE 336

301 indirect effects are significan t for both white and non-white pa rticipants, and were similar in size. Results are shown in Table 203. Table 203. Results of Moderated Mediation (PREDIC TOR = Diversity Clim ate, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Overt race discrimination Constant 21.051.96 10.73.00 DIVERSITY CLIM -.29.06 -4.86.00 Minority Status 10.152.96 3.43.00 DIVERSITY CLIM Minority -.22.09 -2.47.01 Job Satisfaction Constant 8.27.96 8.59.00 Overt race discrimination -.07.02 -3.07.00 DIVERSITY CLIM .11.03 4.04.00 Minority Status -.131.30 -.10.92 DIVERSITY CLIM Minority .01.04 .17.86 Minority Status Indirect Effect SE z p Whites .02.01 2.56.01 Non-whites .03.01 2.83.00 The results for the separate tests of mediation are shown in Table 204. A significant mediating effect was found for white participants, but not for non-white participants. Paths are not al l in the hypothesized directions. Path coefficients were predicted to be different in directions based on minority stat us. However, paths are in the same direction for both majority and mi nority group members. Specifically, it was hypothesized that minority individuals woul d report a negative relationship between diversity climate and perceived discriminati on whereas majority i ndividuals would report a positive relationship. However, the re lationship was negative for both groups.

PAGE 337

302 Table 204. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Diversity Climate, Med = Overt Discrimi nation, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) .13.034.74.00.15.03 5.30 .00 b(MX) -.29.06-5.14.00-.51.07 -7.25 .00 b(YM.X) -.10.03-3.30.00-.03.03 -1.01 .31 b(YX.M) .10.033.51.00.13.03 4.15 .00 Next, perceived over race-based discrimina tion was investigated as a mediator in the relationship between diversity climate and organizational commitment. Support for overall mediation was not found as the path between diversity climate and perceived discrimination was significant ( B = -.29, t = -5.00, p < .01), but the path between the mediator and organizational commitment was not. A significant in teraction was found between the predictor (diversity climate) and the moderator ( B = -.21, t = -2.40, p < .05). Thus, minority status is a moderator in the current model. The indirect effects are not significant for white or non-white participants and were similar in size. Results are shown in Table 205.

PAGE 338

303 Table 205. Results of Moderated Mediation (PREDIC TOR = Diversity Clim ate, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Overt race discrimination Constant 21.031.92 10.95.00 DIVERSITY CLIM -.29.06 -5.00.00 Minority Status 9.932.92 3.40.00 DIVERSITY CLIM Minority -.21.09 -2.40.02 Organizational Constant 17.022.46 6.92.00 Overt race discrimination -.09.06 -1.51.13 DIVERSITY CLIM .43.07 6.30.00 Minority Status -1.923.32 -.58.56 DIVERSITY CLIM Minority .10.10 .10.32 Minority Status Indirect Effect SE z p Whites .03.02 1.42.16 Non-whites .04.03 1.47.14 The results for the separate tests of mediation are shown in Table 206. The mediating effect was non-significant for both white and non-white participants. However, the model approached significance for white participants. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Specifically, it wa s hypothesized that minority individuals would report a negative rela tionship between diversity climate and perceived discrimination whereas majority individuals would report a positive relationship. However, the relationship was negative for both groups.

PAGE 339

304 Table 206. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Diversity Climate, Med = Overt Discriminatio n, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) .45.076.46.00.57.07 8.45 .00 b(MX) -.29.05-5.36.00-.51.07 -7.12 .00 b(YM.X) -.16.09-1.84.07-.02.07 -.22 .83 b(YX.M) .41.075.49.00.56.08 7.30 .00 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between diversity climate a nd intention to turnover. Support for overall mediation was found as the paths between dive rsity climate and perceived discrimination ( B = -.31, t = -5.24, p < .01) and between the mediator and intention to turnover ( B = .04, t = 4.26, p < .01) were jointly significant. Diversit y climate was related to less perceived discrimination which in turn was related to less turnover intention. A significant interaction was found between the predictor and the moderator ( B = -.19, t = -2.20, p < .05). Thus, minority status is a moderator in the current model. The indirect effects are significant for both white and nonwhite participants, and were similar in size. Results are shown in Table 207.

PAGE 340

305 Table 207. Results of Moderated Mediation (PREDIC TOR = Diversity Clim ate, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Overt race discrimination Constant 21.751.94 11.20.00 DIVERSITY CLIM -.31.06 -5.24.00 Minority Status 9.292.94 3.16.00 DIVERSITY CLIM Minority -.19.09 -2.20.03 Turnover Constant 3.84.45 8.59.00 Overt race discrimination .04.01 4.26.00 DIVERSITY CLIM -.05.01 -4.07.00 Minority Status -.79.60 -1.33.18 DIVERSITY CLIM Minority .02.02 1.01.31 Minority Status Indirect Effect SE z p Whites -.01.00 -3.27.00 Non-whites -.02.01 -3.71.00 A significant mediating effect was found for white partic ipants, but the model was not fully supported for non-wh ite participants as the path from diversity climate to perceived discrimination was approaching significance. Paths are not all in the hypothesized directions. Path coe fficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. For example, th e path between minority segmentation and perceived discrimination was predicted to be positive for minority individuals but negative for majority individuals. It was found to be negative for both groups. Results for the simple mediation tests are shown in Table 208.

PAGE 341

306 Table 208. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Diversity Climate, Med = Overt Discriminatio n, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) -.06.01-5.22.00-.05.01 -3.99 .00 b(MX) -.31.06-5.47.00-.51.07 -7.31 .00 b(YM.X) .06.014.13.00.03.01 1.89 .06 b(YX.M) -.05.01-3.66.00-.04.02 -2.59 .01 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between diversity climat e and physical health. Support for overall mediation was found as the paths between dive rsity climate and perceived discrimination ( B = -.29, t = -4.93, p < .01) and between the mediat or and physical health ( B = -.28, t = 5.81, p < .01) were jointly significant. Diversit y climate was related to less perceived discrimination which in turn was related to better physical health. A significant interaction was found between the predictor (diversity cl imate) and the moderator ( B = .20, t = -2.17, p < .05). Thus, minority status is a moderator in the current model. The indirect effects are significan t for both white and non-white pa rticipants, and were similar in size. Results are shown in Table 209.

PAGE 342

307 Table 209. Results of Moderated Mediation (PREDIC TOR = Diversity Clim ate, Med = Overt Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 21.141.97 10.75.00 DIVERSITY CLIM -.29.06 -4.93.00 Minority Status 9.543.09 3.08.00 DIVERSITY CLIM Minority -.20.09 -2.17.03 Physical Health Constant 64.432.02 31.95.00 Overt race discrimination -.28.05 -5.81.00 DIVERSITY CLIM .16.06 2.82.01 Minority Status 3.342.79 1.20.23 DIVERSITY CLIM Minority -.11.08 -1.34.18 Minority Status Indirect Effect SE z p Whites .08.02 3.73.00 Non-whites .14.03 4.45.00 The results for the separate tests of mediation are shown in Table 210. A significant mediating effect was found for bot h white and non-white participants. Paths are not all in the hypothesized di rections. Path coefficients were predicted to be different in directions based on minority status. However, paths are in the same direction for both majority and minority group members. Speci fically, it was hypothesized that minority individuals would report a nega tive relationship between dive rsity climate and perceived discrimination whereas majority individuals would report a positive relationship. However, the relationship was negative for both groups.

PAGE 343

308 Table 210. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Diversity Climate, Med = Overt Discrimi nation, DV = Phys ical Health) Whites Non-whites B SE t p B SE t p b(YX) .24 .054.70.00.18.072.46 .02 b(MX) -.30 .06-5.14.00-.50.07-6.64 .00 b(YM.X) -.22 .06-3.66.00-.34.08-4.47 .00 b(YX.M) .17 .053.31.00.01.08.14 .89 Finally, perceived overt race-based discrimi nation was investigated as a mediator in the relationship between diversity climate and psyc hological health. Support for overall mediation was found as the paths between diversity climate and perceived discrimination ( B = -.31, t = -5.17, p < .05) and between the me diator and psychological health ( B = -.13, t = -2.39, p < .05) were jointly significant. Diversity climate was related to less perceived discrimination which in turn was related to better psychological health. A significant interaction was found between the predictor (d iversity climate) and the moderator ( B = -.21, t = -2.28, p < .05). Thus, minority status is a moderator in the current model. The indirect effects are significant for both white and non-white participants, and were similar in si ze. Results are shown in Table 211.

PAGE 344

309 Table 211. Results of Moderated Mediation (PREDIC TOR = Diversity Clim ate, Med = Overt Discrimination, DV = Psychol ogical Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 21.931.98 11.07.00 DIVERSITY CLIM -.31.06 -5.17.00 Minority Status 9.893.07 3.22.00 DIVERSITY CLIM Minority -.21.09 -2.28.02 Psychological Health Constant 42.462.35 18.08.00 Overt race discrimination -.13.05 -2.39.02 DIVERSITY CLIM .30.06 4.69.00 Minority Status 6.153.20 1.93.06 DIVERSITY CLIM Minority -.20.10 -2.09.04 Minority Status Indirect Effect SE z p Whites .04.02 2.14.03 Non-whites .07.03 2.26.02 A significant mediating effect was found for non-white participants, but not for white participants. Paths are not all in the hypothesized directi ons. Path coefficients were predicted to be different in directions based on minority stat us. However, paths are in the same direction for both majority and mi nority group members. Specifically, it was hypothesized that minority individuals woul d report a negative relationship between diversity climate and perceived discriminati on whereas majority i ndividuals would report a positive relationship. However, the relatio nship was negative for both groups. Results for the simple tests of medi ation are shown in Table 212.

PAGE 345

310 Table 212. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Diversity Climate, Med = Overt Discrimina tion, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) .34.065.50.00.17.07 2.30 .02 b(MX) -.31.06-5.38.00-.53.07 -7.12 .00 b(YM.X) -.06.07-.89.38-.20.08 -2.56 .01 b(YX.M) .32.074.86.00.06.08 .76 .45 Results for Instrumental Social Support Subtle Sex-based Discrimination Instrumental social support in the workplace was investigated as a predictor in a m odel where subtle sex-based discrimination mediates the relationship between instrume ntal social support and five dependent variables (job satisfaction, or ganizational commitment, intention to turnover, physical health, and psychological health). Minority status was investig ated as a moderator in the relationship between the pred ictor and the mediator. First, perceived subtle sex-based discrimi nation was investigated as a mediator in the relationship between instrumental soci al support and job sa tisfaction. Support for overall mediation was found as the paths be tween instrumental social support and perceived discrimination ( B = -.30, t = -2.53, p < .05) and between the mediator and job satisfaction ( B = -.06, t = -2.32, p < .05) were jointly signific ant. Instrumental social support was related to less perc eived discrimination which in turn was related to greater

PAGE 346

311 job satisfaction. No significant interaction wa s found between the pred ictor (instrumental social support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects indicate a difference for men and women. The indirect effect was signifi cant for women (Indirect effect = .02, z = 2.01, p < .05), but was non-significant for men. Results for the moderated mediation are shown in Table 213. Table 213. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 15.441.64 9.42.00 INSTRUMENTAL -.30.12 -2.53.01 Minority Status 1.951.95 1.00.32 INSTRUMENTAL Minority -.04.14 -.25.80 Job satisfaction Constant 4.61.96 4.81.00 Subtle sex discrimination -.06.03 -2.32.02 INSTRUMENTAL .51.06 8.09.00 Minority Status 1.921.03 1.86.06 INSTRUMENTAL Minority -.13.07 -1.79.08 Minority Status Indirect Effect SE z p Men .02.01 1.64.10 Women .02.01 2.01.04 The results for the separate tests of medi ation also indicated a difference for men and women. Here, the mediation model wa s supported for men but not supported for women. However, all paths we re in the hypothesized direc tions. Results are shown in Table 214.

PAGE 347

312 Table 214. Results of Simple Mediation for Men and Women (PREDICTOR = Instrumental Social Support, Med = Subtle Discrimina tion, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .53.068.54.00.40.04 9.87 .00 b(MX) -.30.11-2.66.01-.34.08 -4.34 .00 b(YM.X) -.10.05-2.10.04-.05.03 -1.44 .15 b(YX.M) .50.067.93.00.38.04 9.20 .00 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between instrumental soci al support and organi zational commitment. Support for overall mediation was found as the paths between instrumental social support and perceived discrimination ( B = -.29, t = -2.45, p < .05) and between the mediator and organizational commitment ( B = -.14, t = -2.01, p < .05) were jointly significant. Social support was related to less perc eived discrimination which in turn was related to greater organizational commitment. No significant in teraction was found betw een the predictor and (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are non-significant for both me n and women, and are similar in size and direction. Results are shown in Table 215.

PAGE 348

313 Table 215. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 15.061.61 9.36.00 INSTRUMENTAL -.29.12 -2.45.01 Minority Status 2.191.92 1.15.25 INSTRUMENTAL Minority -.04.14 -.30.77 Organizational Constant 16.562.45 6.77.00 Subtle sex discrimination -.14.07 -2.01.04 INSTRUMENTAL 1.08.16 6.69.00 Minority Status -2.352.64 -.89.37 INSTRUMENTAL Minority .26.19 1.36.17 Minority Status Indirect Effect SE z p Men .04.03 1.49.14 Women .05.03 1.79.07 Results for both men and women failed to support the mediation model. However, all paths were in the hypothesized directions. Table 216 displays the re sults for the simple tests of mediation. Table 216. Results of Simple Mediation for Men and Women (PREDICTOR = Instrumental Social Support, Med = Subtle Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) 1.12.167.10.001.39.10 13.39 .00 b(MX) -.29.11-2.71.01-.33.08 -4.24 .00 b(YM.X) -.13.14-.96.34-.14.08 -1.76 .08 b(YX.M) 1.08.166.66.001.34.11 12.57 .00

PAGE 349

314 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between instrumental social support and intention to turnover. Support for overall mediation was found as the path s between instrumental social support and perceived discrimination ( B = -.32, t = -2.65, p < .05) and between the mediator and intention to turnover ( B = .04, t = 3.19, p < .01) were jointly significant. Social support was related to less perceived discrimination wh ich in turn was related to less turnover intention. No significant inter action was found between the pred ictor (instrumental social support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects indi cate a difference between men and women, as the indirect effects are significant for wome n, but are not for men. Results are shown in Table 217. Table 217. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 15.721.66 9.46.00 INSTRUMENTAL -.32.12 -2.65.01 Minority Status 1.721.96 .88.38 INSTRUMENTAL Minority -.02.14 -.16.87 Turnover Constant 4.98.48 10.43.00 Subtle sex discrimination .04.01 3.19.00 INSTRUMENTAL -.21.03 -6.65.00 Minority Status -.70.51 -1.38.17 INSTRUMENTAL Minority .06.04 1.56.12 Minority Status Indirect Effect SE z p Men -.01.01 -1.98.05 Women -.01.01 -2.57.01

PAGE 350

315 The separate tests of medi ation indicated a significant mediating effect for both men and women. All paths were in the hypothe sized directions. Results are shown in Table 218. Table 218. Results of Simple Mediation for Men and Women (PREDICTOR = Instrumental Social Support, Med = Subtle Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) -.22.03-7.76.00-.17.02 -8.16 .00 b(MX) -.32.11-2.78.01-.34.08 -4.47 .00 b(YM.X) .06.022.52.01.04.02 2.25 .03 b(YX.M) -.20.03-7.07.00-.15.02 -7.35 .00 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between instrumental soci al support and physical health. Support for overall mediation was found as the paths be tween instrumental social support and perceived discrimination ( B = -.26, t = -2.08, p < .05) and between the mediator and physical health ( B = -.29, t = -4.36, p < .01) were jointly signif icant. Social support was related to less perceived discrimination whic h in turn was related to better physical health. No significant interaction was found be tween the predictor (instrumental social support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects were different for men and women, as the indirect effect was significant for wome n, but non-significant for men. Results are shown in Table 219.

PAGE 351

316 Table 219. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Subtle Discrimination, DV = Phys ical Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 14.841.70 8.74.00 INSTRUMENTAL -.26.12 -2.08.04 Minority Status 1.832.01 .91.36 INSTRUMENTAL Minority -.04.14 -.29.77 Physical Health Constant 67.872.30 29.51.00 Subtle sex discrimination -.29.07 -4.36.00 INSTRUMENTAL .17.15 1.15.25 Minority Status -2.892.46 -1.17.24 INSTRUMENTAL Minority .11.18 .62.54 Minority Status Indirect Effect SE z p Men .07.04 1.84.07 Women .08.03 2.86.00 The separate tests of medi ation revealed full support of the mediation model for women as both paths were significant. On the other hand, there was no support of the model for men. However, all paths were in the hypothesized direc tions. Results for the simple tests of mediation are shown in Table 220.

PAGE 352

317 Table 220. Results of Simple Mediation for Men and Women (PREDICTOR = Instrumental Social Support, Med = Subtle Discrimination, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .25.141.71.09.37.10 3.73 .00 b(MX) -.26.12-2.12.04-.30.08 -3.87 .00 b(YM.X) -.20.11-1.72.09-.33.08 -4.06 .00 b(YX.M) .20.151.34.18.27.10 2.76 .01 Finally, perceived subtle sex-based discri mination was investigated as a mediator in the relationship between in strumental social support and psychological health. Support for overall mediation was found as the path s between instrumental social support and perceived discrimination ( B = -.29, t = -2.39, p < .05) and between the mediator and psychological health ( B = -.20, t = -2.62, p < .05) were jointly signi ficant. Social support was related to less perceived discriminati on which in turn was related to better psychological health. No significant inte raction was found between the predictor (instrumental social support) and the moderator (minority stat us). Thus, minority status is not a moderator in the current model. The i ndirect effects were different for men and women, as the indirect effect was significant for women, but not for men. Results are displayed in Table 221.

PAGE 353

318 Table 221. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Subtle Discrimination, DV = Psychological Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 15.041.67 9.04.00 INSTRUMENTAL -.29.12 -2.39.02 Minority Status 1.871.98 .95.35 INSTRUMENTAL Minority -.01.14 -.10.92 Psychological Health Constant 46.312.65 17.48.00 Subtle sex discrimination -.20.08 -2.62.01 INSTRUMENTAL .51.17 2.94.00 Minority Status -.672.84 -.24.81 INSTRUMENTAL Minority -.02.21 -.10.93 Minority Status Indirect Effect SE z p Men .06.03 1.70.09 Women .06.03 2.13.03 The results for the separate tests of mediation are shown in Table 222. A significant mediating effect was found for wo men, but not for men. However, all paths were in the hypothesized directions.

PAGE 354

319 Table 222. Results of Simple Mediation for Men and Women (PREDICTOR = Instrumental Social Support, Med = Subtle Discriminati on, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) .57.183.24.00.55.11 4.96 .00 b(MX) -.29.11-2.73.01-.30.08 -3.72 .00 b(YM.X) -.05.16-.31.75-.24.09 -2.86 .00 b(YX.M) .56.183.05.00.48.11 4.25 .00 Overt Sex-based Discrimination Instrumental social s upport in the workplace was investigated as a predictor in a model wher e overt sex-based discri mination mediates the relationship between instrumental social support and five depe ndent variables (job satisfaction, organizational commitment, inte ntion to turnover, physical health, and psychological health). Minority st atus was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between instrumental soci al support and job sa tisfaction. Support for overall mediation was found as the paths be tween instrumental social support and perceived discrimination ( B = -.65, t = -3.68, p < .01) and between the mediator and job satisfaction ( B = -.04, t = -2.19, p < .05) were jointly signif icant. Social support was related to less perceived di scrimination which in turn was related to greater job satisfaction. No significant in teraction was found between the predictor (instrumental social support) and the moderator (minority status). Thus, minority status is not a

PAGE 355

320 moderator in the current model. The indirect effects are similar for men and women, in that they are both significant, and nearly identical in size. Results for the moderated mediation are shown in Table 223. Table 223. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 21.912.42 9.05.00 INSTRUMENTAL -.65.18 -3.68.00 Minority Status -1.412.90 -.49.63 INSTRUMENTAL Minority .17.21 .81.42 Job Satisfaction Constant 4.28.91 4.69.00 Overt sex discrimination -.04.02 -2.19.03 INSTRUMENTAL .52.06 8.45.00 Minority Status 2.261.00 2.27.02 INSTRUMENTAL Minority -.15.07 -2.08.04 Minority Status Indirect Effect SE z p Men .02.01 1.84.07 Women .02.01 1.90.06 The results for the separate tests of me diation indicate a difference between men and women. Here, the mediation model is fu lly supported for men but not fully supported for women. However, all paths were in the hypo thesized directions. Results are shown in Table 224.

PAGE 356

321 Table 224. Results of Simple Mediation for Men and Women (PREDICTOR = Instrumental Social Support, Med = Overt Discrimina tion, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .54.069.11.00.39.04 9.70 .00 b(MX) -.65.17-3.87.00-.48.12 -4.06 .00 b(YM.X) -.07.03-2.34.02-.02.02 -1.15 .25 b(YX.M) .49.067.99.00.37.04 9.15 .00 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between instrumental soci al support and organi zational commitment. Support for overall mediation was found as the paths between instrumental social support and perceived discrimination ( B = -.63, t = -3.64, p < .01) and between the mediator and organizational commitment ( B = -.10, t = -2.11, p < .05) were jointly significant. Social support was related to less perc eived discrimination which in turn was related to greater organizational commitment. No significant interaction was found be tween the predictor (instrumental social support) and the moderator (minority stat us). Thus, minority status is not a moderator in the current model. The i ndirect effects are similar for men and women, as they are both non-significant and simila r in size. Results are shown in Table 225.

PAGE 357

322 Table 225. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 21.562.41 8.95.00 INSTRUMENTAL -.63.17 -3.64.00 Minority Status -.992.89 -.34.73 INSTRUMENTAL Minority .15.21 .73.47 Organizational Constant 15.762.39 6.60.00 Overt sex discrimination -.10.05 -2.11.04 INSTRUMENTAL 1.11.16 6.94.00 Minority Status -1.172.61 -.45.65 INSTRUMENTAL Minority .18.19 .96.34 Minority Status Indirect Effect SE z p Men .06.03 1.78.08 Women .05.03 1.85.06 The results for the separate tests of mediation are shown in Table 226. The mediating effect was not fully s upported for either men or women. Table 226. Results of Simple Mediation for Men and Women (PREDICTOR = Instrumental Social Support, Med = Overt Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) 1.17.157.69.001.33.10 12.72 .00 b(MX) -.63.16-3.88.00-.48.12 -4.12 .00 b(YM.X) -.09.08-1.04.30.10.05 -1.83 .07 b(YX.M) 1.11.166.91.001.29.11 11.95 .00

PAGE 358

323 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between instrumental social support and intention to turnover. Support for overall mediation was found as the path s between instrumental social support and perceived discrimination ( B = -.65, t = -3.71, p < .01) and between the mediator and intention to turnover ( B = .03, t = 3.60, p < .01) were jointly significant. Social support was related to less perceived discrimination wh ich in turn was related to less turnover intention. No significant inter action was found between the pred ictor (instrumental social support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects we re significant for both men and women, and were similar in size and direction. Results of the moderated mediati on are shown in Table 227. Table 227. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 21.912.40 9.13.00 INSTRUMENTAL -.65.17 -3.71.00 Minority Status -.982.86 -.34.73 INSTRUMENTAL Minority .14.21 .66.51 Turnover Constant 5.02.45 11.10.00 Overt sex discrimination .03.01 3.60.00 INSTRUMENTAL -.21.03 -6.82.00 Minority Status -.81.49 -1.64.10 INSTRUMENTAL Minority .06.04 1.77.08 Minority Status Indirect Effect SE z p Men -.02.01 -2.54.01 Women -.01.01 -2.78.01

PAGE 359

324 The results for the separate tests of mediation are shown in Table 228. A significant mediating effect was found for both men and women. Moreover, all paths were the hypothesized directions. Table 228. Results of Simple Mediation for Men and Women (PREDICTOR = Instrumental Social Support, Med = Overt Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) -.23.03-8.23.00-.16.02 -7.86 .00 b(MX) -.65.17-3.87.00-.51.11 -4.44 .00 b(YM.X) .03.012.24.03.03.01 2.86 .00 b(YX.M) -.21.03-7.17.00-.14.02 -6.95 .00 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between instrumental soci al support and physical health. Support for overall mediation was found as the paths be tween instrumental social support and perceived discrimination ( B = -.50, t = -2.72, p < .05) and between the mediator and physical health ( B = -.22, t = -4.97, p < .01) were jointly signif icant. Social support was related to less perceived discrimination whic h in turn was related to better physical health. No significant interaction was found be tween the predictor (instrumental social support) and the moderator (minority status ). Additionally, the i ndirect effects are significant for both men and women, and were similar in size. Results are displayed in Table 229.

PAGE 360

325 Table 229. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Overt Discrimination, DV = Phys ical Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 19.902.57 7.74.00 INSTRUMENTAL -.50.19 -2.72.01 Minority Status -.443.05 -.14.89 INSTRUMENTAL Minority .09.22 .41.68 Physical Health Constant 67.392.27 29.75.00 Overt sex discrimination -.22.04 -4.97.00 INSTRUMENTAL .16.15 1.04.30 Minority Status -2.972.48 -1.20.23 INSTRUMENTAL Minority .12.18 .67.50 Minority Status Indirect Effect SE z p Men .11.05 2.35.02 Women .09.03 2.84.00 The results for the separate tests of mediation are shown in Table 230. A significant mediating effect was found for bot h men and women. All paths were in the hypothesized directions.

PAGE 361

326 Table 230. Results of Simple Mediation for Men and Women (PREDICTOR = Instrumental Social Support, Med = Overt Discrimina tion, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .27.151.78.08.37.10 3.65 .00 b(MX) -.50.18-2.86.01-.41.12 -3.45 .00 b(YM.X) -.26.08-3.40.00-.20.05 -3.75 .00 b(YX.M) .13.15.90.37.29.10 2.84 .00 Finally, perceived overt sexbased discrimination was inve stigated as a mediator in the relationship between in strumental social support and psychological health. Support for overall mediation was found as the path s between instrumental social support and perceived discrimination ( B = -.61, t = -3.47, p < .01) and between the mediator and psychological health ( B = -.17, t = -3.57, p < .01) were jointly signi ficant. Social support was related to less perceived discriminati on which in turn was related to better psychological health. No significant inte raction was found between the predictor (instrumental social support) and the moderator (minority stat us). Thus, minority status is not a moderator in the current model. Additio nally, the indirect effects were significant for both men and women, and were similar in size. Results are displayed in Table 231.

PAGE 362

327 Table 231. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Overt Discrimination, DV = Psychological Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 21.242.46 8.65.00 INSTRUMENTAL -.61.18 -3.47.00 Minority Status -1.182.95 -.40.69 INSTRUMENTAL Minority .17.21 .81.42 Psychological Health Constant 47.762.54 18.84.00 Overt sex discrimination -.17.05 -3.57.00 INSTRUMENTAL .41.17 2.42.02 Minority Status -1.772.78 -.64.52 INSTRUMENTAL Minority .05.20 .26.80 Minority Status Indirect Effect SE z p Men .10.04 2.44.01 Women .08.03 2.54.01 The results for the separate tests of me diation indicated a difference between men and women. The model was fully supported for women, but not supported for men. All paths were in the hypothesized directions Results are displayed in Table 232. Table 232. Results of Simple Mediation for Men and Women (PREDICTOR = Instrumental Social Support, Med = Overt Discriminatio n, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) .52.173.07.00.54.11 4.78 .00 b(MX) -.61.17-3.76.00-.44.12 -3.62 .00 b(YM.X) -.13.09-1.37.17-.19.06 -3.36 .00 b(YX.M) .44.182.46.02.46.11 4.01 .00

PAGE 363

328 Subtle Race-based Discrimination Instrumental social support in the workplace was investigated as a predictor in a mode l where subtle race-based discrimination mediates the relationship between instrume ntal social support and five dependent variables (job satisfaction, or ganizational commitment, intention to turnover, physical health, and psychological health). Minority status was investig ated as a moderator in the relationship between the pred ictor and the mediator. First, perceived subtle race-based discrimi nation was investigated as a mediator in the relationship between instrumental soci al support and job sa tisfaction. Support for overall mediation was found as the paths be tween instrumental social support and perceived discrimination ( B = -.18, t = -2.03, p < .05) and between the mediator and job satisfaction ( B = -.07, t = -2.87, p < .01) were jointly signif icant. Social support was related to less perceived di scrimination which in turn was related to greater job satisfaction. A significant interaction was found between the pred ictor (instrumental social support) and the moderator ( B = -.32, t = -2.34, p < .05). Thus, minority status is a moderator in the current model. The indirect effects were also di fferent for white and non-white participants. The i ndirect effect was significan t for non-white participants (Indirect effect = .04, z = 2.43, p < .05), but non-significant for white participants. Results are shown in Table 233.

PAGE 364

329 Table 233. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 12.901.19 10.85.00 INSTRUMENTAL -.18.09 -2.03.04 Minority Status 6.821.88 3.63.00 INSTRUMENTAL Minority -.32.13 -2.34.02 Job Satisfaction Constant 6.02.68 8.91.00 Subtle race discrimination -.07.03 -2.87.00 INSTRUMENTAL .42.04 9.61.00 Minority Status .18.95 .19.85 INSTRUMENTAL Minority -.01.07 -.15.88 Minority Status Indirect Effect SE z p Whites .01.01 1.59.11 Non-whites .04.01 2.43.02 The results for the separate tests of mediation are shown in Table 234. A significant mediating effect was found for non-white participants, but not for white participants. However, all paths we re in the hypothesized directions.

PAGE 365

330 Table 234. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Instrumental Social Support, Med = Subtle Discri mination, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) .43.049.87.00.45.05 8.67 .00 b(MX) -.18.07-2.46.01-.49.12 -4.10 .00 b(YM.X) -.08.04-1.92.06-.07.03 -2.14 .03 b(YX.M) .42.049.48.00.41.05 7.74 .00 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between in strumental social support and organizational commitment. Support for overall mediation was found as the paths between instrumental social support and perceived discrimination ( B = -.19, t = -2.22, p < .05) and between the mediator and organizational commitment ( B = -.15, t = -2.33, p < .05) were jointly significant. Social support was related to less perc eived discrimination which in turn was related to greater organizational commitment. A significant in teraction was found between the predictor (instrumental social supp ort) and the moderator ( B = -.27, t = -2.04, p < .05). Thus, minority status is a moderator in the current model. Additionally, the indirect effects are different for white and non-white participants, as the indirect effect is significant for nonwhite participants, but not for white part icipants. Results are shown in Table 235.

PAGE 366

331 Table 235. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 12.981.16 11.14.00 INSTRUMENTAL -.19.09 -2.22.03 Minority Status 6.301.86 3.39.00 INSTRUMENTAL Minority -.27.13 -2.04.04 Organizational Constant 14.171.74 8.16.00 Subtle race discrimination -.15.06 -2.33.02 INSTRUMENTAL 1.28.11 11.42.00 Minority Status 1.872.45 .76.45 INSTRUMENTAL Minority -.05.17 -.27.79 Minority Status Indirect Effect SE z p Whites .03.02 1.54.12 Non-whites .07.04 2.04.04 The results for the separate tests of me diation indicate support of the mediation model for non-white participants but not for white participan ts. However, all paths were in the hypothesized directions. Results of th e simple tests of mediation are shown in Table 236.

PAGE 367

332 Table 236. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Instrumental Social Support, Med = Subtle Discrimina tion, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) 1.31.1111.57.001.31.13 9.96 .00 b(MX) -.19.07-2.77.01-.46.12 -3.82 .00 b(YM.X) -.09.11-.82.41-.19.08 -2.33 .02 b(YX.M) 1.29.1211.21.001.22.13 9.05 .00 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between in strumental social support and intention to turnover. Support for overall mediation was found as the path s between instrumental social support and perceived discrimination ( B = -.21, t = -2.37, p < .05) and between the mediator and intention to turnover ( B = .06, t = 4.97, p < .01) were jointly significant. Social support was related to less perceived discrimination which in turn was related to less intention to turnover. A significant interaction was found be tween the predictor (instrumental social support) and the moderator ( B = -.32, t = -2.40, p < .05). Thus, minority status is a moderator in the current model. The indirect effects are similar for white and non-white participants, as they are both significant a nd similar in size and direction. Results are shown in Table 237.

PAGE 368

333 Table 237. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 13.321.18 11.33.00 INSTRUMENTAL -.21.09 -2.37.02 Minority Status 6.801.86 3.66.00 INSTRUMENTAL Minority -.32.13 -2.40.02 Turnover Constant 4.77.33 14.53.00 Subtle race discrimination .06.01 4.97.00 INSTRUMENTAL -.20.02 -9.44.00 Minority Status -1.40.46 -3.06.00 INSTRUMENTAL Minority .08.03 2.70.01 Minority Status Indirect Effect SE z p Whites -.01.01 -2.11.03 Non-whites -.03.01 -3.55.00 The results for the separate tests of mediation indicate full support for the mediation model for both white and non-wh ite participants. A ll paths were in hypothesized directions. Table 238 displays the results for the simple tests of mediation.

PAGE 369

334 Table 238. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Instrumental Social Support, Med = Subtle Discrimina tion, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) -.21.02-10.33.00-.14.03 -5.36 .00 b(MX) -.21.07-2.83.01-.53.12 -4.43 .00 b(YM.X) .07.023.55.00.06.02 3.46 .00 b(YX.M) -.20.02-9.74.00-.11.03 -4.15 .00 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between in strumental social support and physical health. Support was not found for overall mediation as the path between instrumental social support and perceived discrimination was non-significant, but the path between the mediator and physical health ( B = -.32, t = -4.75, p < .01) was significant. No significant interaction was found between the predic tor (instrumental social support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Additionally, the indirect effect s were different for white a nd non-white participants, as the indirect was significant for non-white participants (Indirect effect = .11, z = 2.70, p < .05), but was non-significant for white participants. Results are displayed in Table 239.

PAGE 370

335 Table 239. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Subtle Discrimination, DV = Ph ysical Health, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 12.671.16 10.91.00 INSTRUMENTAL -.17.09 -1.94.05 Minority Status 4.911.90 2.58.01 INSTRUMENTAL Minority -.19.14 -1.40.16 Physical Health Constant 65.481.67 39.25.00 Subtle race discrimination -.32.07 -4.75.00 INSTRUMENTAL .29.11 2.70.01 Minority Status 1.582.38 .66.51 INSTRUMENTAL Minority -.13.17 -.75.45 Minority Status Indirect Effect SE z p Whites .05.03 1.76.08 Non-whites .11.04 2.70.01 The results for the separate tests of medi ation indicate significa nt mediating effect was found for both white and non-white particip ants. All paths were in the hypothesized directions. Table 240 displays the results. Table 240. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Instrumental Social Support, Med = Subtle Discrimination, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) .34.103.37.00.27.15 1.85 .07 b(MX) -.17.07-2.27.02-.35.12 -2.89 .00 b(YM.X) -.23.10-2.31.02-.38.09 -4.09 .00 b(YX.M) .30.102.99.00.14.14 .97 .34

PAGE 371

336 Finally, perceived subtle race -based discrimination was i nvestigated as a mediator in the relationship between in strumental social support and psychological health. Support was not found for overall mediation as the path between instrumental social support and perceived discrimination was significant ( B = -.22, t = -2.42, p < .05) but the path between the mediator and psychological heal th was not. No significant interaction was found between the predictor (instrumental so cial support) and the moderator (minority status). Thus, minority status is not a moderator in the curren t model. The indirect effects were not significant for either white or nonwhite participants. Results are displayed in Table 241. Table 241. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Subtle Discrimination, DV = Psychological Health, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 13.501.22 11.08.00 INSTRUMENTAL -.22.09 -2.42.02 Minority Status 5.841.94 3.01.00 INSTRUMENTAL Minority -.25.14 -1.79.07 Psychological Health Constant 43.571.92 22.66.00 Subtle race discrimination -.08.07 -1.16.25 INSTRUMENTAL .57.12 4.61.00 Minority Status 1.772.69 .66.51 INSTRUMENTAL Minority -.15.19 -.77.44 Minority Status Indirect Effect SE z p Whites .02.02 .98.33 Non-whites .04.04 1.09.27

PAGE 372

337 The results for the separate tests of mediation are shown in Table 242. A significant mediating effect was not found for either white or non-white participants. However, all paths were in the hypothesized directions. Table 242. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Instrumental Social Support, Med = Subtle Discrimi nation, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) .59.124.70.00.46.14 3.17 .00 b(MX) -.22.07-2.90.00-.47.13 -3.69 .00 b(YM.X) -.08.12-.65.52-.09.09 -.98 .33 b(YX.M) .57.134.47.00.42.15 2.77 .01 Overt Race-based Discrimination Instrumental social support in the workplace was investigated as a predictor in a m odel where overt race-b ased discrimination mediates the relationship between instrume ntal social support and five dependent variables (job satisfaction, or ganizational commitment, intention to turnover, physical health, and psychological health). Minority status was investig ated as a moderator in the relationship between the pred ictor and the mediator. First, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between instrumental soci al support and job sa tisfaction. Support for overall mediation was found as the paths be tween instrumental social support and perceived discrimination ( B = -.37, t = -3.11, p < .01) and between the mediator and job satisfaction ( B = -.04, t = -2.40, p < .05) were jointly signif icant. Social support was

PAGE 373

338 related to less perceived di scrimination which in turn was related to greater job satisfaction. No significant in teraction was found between the predictor (instrumental social support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects were different for white and nonwhite participants, as the indirect effect was significant for non-white participants (Indirect effect = .03, z = 2.12, p < .05), but non-significant for white participants. Results of the moderated mediation are shown in Table 243. Table 243. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Overt race discrimination Constant 16.701.62 10.32.00 INSTRUMENTAL -.37.12 -3.11.00 Minority Status 7.532.60 2.90.00 INSTRUMENTAL Minority -.34.19 -1.80.07 Job Satisfaction Constant 5.98.67 8.90.00 Overt race discrimination -.04.02 -2.40.02 INSTRUMENTAL .41.04 9.19.00 Minority Status -.29.97 -.29.77 INSTRUMENTAL Minority .02.07 .24.81 Minority Status Indirect Effect SE z p Whites .02.01 1.84.07 Non-whites .03.01 2.12.03 The results for the separate tests of mediation indicated full support for the mediation model for white participants, but not for non-white participants. However, all paths were in the hypothesized directions. Ta ble 244 displays the results for the simple tests of mediation.

PAGE 374

339 Table 244. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Instrumental Social Support, Med = Overt Discri mination, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) .42.049.49.00.45.05 8.69 .00 b(MX) -.37.11-3.34.00-.70.16 -4.54 .00 b(YM.X) -.08.03-2.95.00-.01.03 -.48 .63 b(YX.M) .39.058.77.00.45.06 8.04 .00 Next, perceived over race-based discrimina tion was investigated as a mediator in the relationship between instrumental soci al support and organi zational commitment. Support for overall mediation was found as the paths between instrumental social support and perceived discrimination ( B = -.39, t = -3.30, p < .01) and between the mediator and organizational commitment ( B = -.10, t = -2.02, p < .05) were jointly significant. Social support was related to less perc eived discrimination which in turn was related to greater organizational commitment. No significant interaction was found be tween the predictor (instrumental social support) and the moderator (minority stat us). Thus, minority status is not a moderator in the current model. The i ndirect effects are similar for white and nonwhite participants, in that they are both nonsignificant and similar in size and direction. Results are shown in Table 245.

PAGE 375

340 Table 245. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Overt race discrimination Constant 16.851.60 10.55.00 INSTRUMENTAL -.39.12 -3.30.00 Minority Status 7.682.59 2.96.00 INSTRUMENTAL Minority -.33.19 -1.79.07 Organizational Constant 14.551.71 8.50.00 Overt race discrimination -.10.05 -2.02.04 INSTRUMENTAL 1.23.11 10.90.00 Minority Status .852.48 .34.73 INSTRUMENTAL Minority -.00.18 -.00.99 Minority Status Indirect Effect SE z p Whites .04.02 1.67.10 Non-whites .07.04 1.85.07 The results for the separate tests of mediation are shown in Table 246. The mediating effect was non-significant for both white and non-white participants. However, all paths were in the hypothesized directions.

PAGE 376

341 Table 246. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Instrumental Social Support, Med = Overt Discrimina tion, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) 1.26.1111.13.001.30.13 9.73 .00 b(MX) -.39.11-3.59.00-.72.16 -4.58 .00 b(YM.X) -.13.07-1.75.08-.07.06 -1.11 .27 b(YX.M) 1.22.1210.45.001.24.14 8.82 .00 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between instrumental social support and intention to turnover. Support for overall mediation was found as the path s between instrumental social support and perceived discrimination ( B = -.42, t = -3.51, p < .01) and between the mediator and intention to turnover ( B = .03, t = 3.84, p < .01) were jointly significant. Social support was related to less perceived discrimination which in turn was related to less intention to turnover. No significant interaction was found between the predictor (instrumental social support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Additionall y, the indirect effects are si gnificant for both white and non-white participants, and we re similar in size. Results are shown in Table 247.

PAGE 377

342 Table 247. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Overt race discrimination Constant 17.401.61 10.82.00 INSTRUMENTAL -.42.12 -3.51.00 Minority Status 7.622.59 2.95.00 INSTRUMENTAL Minority -.35.19 -1.86.06 Turnover Constant 4.84.33 14.88.00 Overt race discrimination .03.01 3.84.00 INSTRUMENTAL -.19.02 -8.84.00 Minority Status -.98.46 -2.12.03 INSTRUMENTAL Minority .07.03 1.97.05 Minority Status Indirect Effect SE z P Whites -.01.01 -2.55.01 Non-whites -.02.01 -3.08.00 The results for the separate tests of me diation indicate a fully supported mediation model for white participants, but not for non-white participan ts. However, all paths were in hypothesized directions. Tabl e 248 displays the results.

PAGE 378

343 Table 248. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Instrumental Social Support, Med = Overt Discrimina tion, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) -.20.02-9.84.00-.15.03 -5.52 .00 b(MX) -.42.11-3.71.00-.76.15 -4.99 .00 b(YM.X) .05.014.02.00.02.01 1.55 .12 b(YX.M) -.18.02-8.89.00-.13.03 -4.64 .00 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between instrumental soci al support and physical health. Support for overall mediation was found as the paths be tween instrumental social support and perceived discrimination ( B = -.36, t = -3.04, p < .01) and between the mediator and physical health ( B = -.28, t = -6.13, p < .01) were jointly signif icant. Social support was related to less perceived discrimination whic h in turn was related to better physical health. No significant interaction was found be tween the predictor (instrumental social support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects are significant and similar in size for both white and non-white participants. Results for the m oderated mediation are shown in Table 249.

PAGE 379

344 Table 249. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Overt Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 16.581.64 10.09.00 INSTRUMENTAL -.36.12 -3.04.00 Minority Status 6.942.74 2.54.01 INSTRUMENTAL Minority -.30.20 -1.52.13 Physical Health Constant 65.841.59 41.37.00 Overt race discrimination -.28.05 -6.13.00 INSTRUMENTAL .26.10 2.53.01 Minority Status 2.212.35 .94.35 INSTRUMENTAL Minority -.18.17 -1.04.30 Minority Status Indirect Effect SE z p Whites .10.04 2.69.01 Non-whites .19.06 3.48.00 The results for the separate tests of mediation are shown in Table 250. A significant mediating effect was found for bot h white and non-white participants. All paths were in hypothesized directions.

PAGE 380

345 Table 250. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Instrumental Social Support, Med = Overt Discrimination, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) .36.103.67.00.27.15 1.80 .07 b(MX) -.36.11-3.23.00-.66.17 -3.99 .00 b(YM.X) -.24.06-4.01.00-.32.07 -4.56 .00 b(YX.M) .28.102.81.01.06.15 .41 .68 Finally, perceived overt race-based discrimi nation was investigated as a mediator in the relationship between in strumental social support and psychological health. Support for overall mediation was found as the path s between instrumental social support and perceived discrimination ( B = -.44, t = -3.55, p < .01) and between the mediator and psychological health ( B = -.13, t = -2.51, p < .05) were jointly signi ficant. Social support was related to less perceived discriminati on which in turn was related to better psychological health. No significant inte raction was found between the predictor (instrumental social support) and the moderator (minority stat us). Thus, minority status is not a moderator in the current model. Additiona lly, the indirect effect was significant for non-white participants (Indirect effect = .09, z = 2.14, p < .05), but was only approaching significance for white particip ants (Indirect effect = .06, z = 2.00, p = .05). Results are displayed in Table 251.

PAGE 381

346 Table 251. Results of Moderated Mediation (PREDICTO R = Instrumental Social Support, Med = Overt Discrimination, DV = Psychological Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 17.731.67 10.64.00 INSTRUMENTAL -.44.12 -3.55.00 Minority Status 6.092.71 2.25.03 INSTRUMENTAL Minority -.24.20 -1.22.22 Psychological Health Constant 44.681.87 23.93.00 Overt race discrimination -.13.05 -2.51.01 INSTRUMENTAL .55.12 4.47.00 Minority Status 1.982.67 .74.46 INSTRUMENTAL Minority -.17.19 -.87.38 Minority Status Indirect Effect SE z p Whites .06.03 2.00.05 Non-whites .09.04 2.14.03 The results for the separate tests of mediation are shown in Table 252. A significant mediating effect was found for non-white participants, but the mediation model was not fully supported for white pa rticipants. However, all paths were in hypothesized directions.

PAGE 382

347 Table 252. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Instrumental Social Support, Med = Overt Discrimi nation, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) .60.124.92.00.46.15 3.11 .00 b(MX) -.44.12-3.77.00-.68.16 -4.11 .00 b(YM.X) -.09.07-1.25.21-.17.07 -2.33 .02 b(YX.M) .56.134.45.00.35.15 2.28 .02 Results for Emotional Social Support Subtle Sex-based Discrimination Emotional social support in the workplace was investigated as a predictor in a model where subtle sex-based discri mination mediates the relationship between emoti onal social support and five dependent variables (job satisfaction, organizational commitment, inte ntion to turnover, physical health, and psychological health). Minority st atus was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived subtle sex-based discrimi nation was investigated as a mediator in the relationship between emotional social s upport and job satisfaction. Support for overall mediation was found as the paths between emotional social support and perceived discrimination ( B = -.33, t = -2.65, p < .05) and between the medi ator and job satisfaction ( B = -.06, t = -2.02, p < .05) were jointly si gnificant. Social support was related to less perceived discrimination which in turn wa s related to greater job satisfaction. No significant interaction was found between the pr edictor (emotional soci al support) and the

PAGE 383

348 moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects were simila r for men and women, as they were both nonsignificant and similar in size. Results for the moderated mediation are shown in Table 253. Table 253. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 16.461.91 8.61.00 EMOTIONAL -.33.12 -2.65.01 Minority Status 3.592.30 1.56.12 EMOTIONAL Minority Status -.14.15 -.98.33 Job satisfaction Constant 5.051.17 4.33.00 Subtle sex discrimination -.06.03 -2.02.04 EMOTIONAL .43.07 6.14.00 Minority Status .711.29 .55.58 EMOTIONAL Minority Status -.04.08 -.49.63 Minority Status Indirect Effect SE z p Men .02.01 1.55.12 Women .03.01 1.89.06 The results for the separate tests of me diation are shown in Table 254. There was a lack of support for the mediation model for both men and women. However, all paths were in the hypothesized directions.

PAGE 384

349 Table 254. Results of Simple Mediation for Men and Women (PREDICTOR = Emotional Social Support, Med = Subtle Discrimina tion, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .45.076.30.00.42.05 9.07 .00 b(MX) -.33.12-2.71.01-.48.08 -5.70 .00 b(YM.X) -.10.05-1.81.07-.04.03 -1.24 .22 b(YX.M) .42.075.72.00.40.05 8.15 .00 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between emotional social support and organizational commitment. Support for overall mediation was not found as the path between emo tional social support and perceived discrimination ( B = -.32, t = -2.60, p < .05) was significant but the path between the mediator and organizational comm itment was not significant. No significant interaction was found between th e predictor (emotional social support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Moreover, the indirect effect s were non-significant but similar in size for both men and women. Results are shown in Table 255.

PAGE 385

350 Table 255. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 16.041.88 8.56.00 EMOTIONAL -.32.12 -2.60.01 Minority Status 3.822.25 1.70.09 EMOTIONAL Minority Status -.15.15 -1.02.30 Organizational Constant 18.323.16 5.80.00 Subtle sex discrimination -.16.08 -1.98.05 EMOTIONAL .86.19 4.54.00 Minority Status -2.443.47 -.70.48 EMOTIONAL Minority Status .25.23 1.12.26 Minority Status Indirect Effect SE z p Men .05.03 1.51.13 Women .07.04 1.85.06 The results for the separate tests of me diation revealed lack of support for the mediation model in both me n and women. However, all pa ths were in the hypothesized directions. Table 256 displays the results for the simple tests of mediation.

PAGE 386

351 Table 256. Results of Simple Mediation for Men and Women (PREDICTOR = Emotional Social Support, Med = Subtle Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) .91.185.12.001.19.13 9.33 .00 b(MX) -.32.11-2.77.01-.47.08 -5.66 .00 b(YM.X) -.14.15-.94.35-.17.10 -1.73 .08 b(YX.M) .87.184.71.001.10.13 8.26 .00 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between emotional social s upport and intention to turnover. Support for overall mediation was found as the paths betw een emotional social support and perceived discrimination ( B = -.33, t = -2.64, p < .05) and between the me diator and intention to turnover ( B = .04, t = 3.14, p < .01) were jointly significant. Social support was related to less perceived discrimination which in turn was related to less turnover intention. No significant interaction was found between the pr edictor (emotional soci al support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effect was signifi cant for women (Indirect effect = -.02, z = -2.74, p < .05) but was marginally significant for men (Indirect effect = -.01, z = -1.97, p = .05). Results are displayed in Table 257.

PAGE 387

352 Table 257. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 16.461.91 8.62.00 EMOTIONAL -.33.12 -2.64.01 Minority Status 3.602.29 1.57.12 EMOTIONAL Minority Status -.15.15 -1.02.31 Turnover Constant 3.89.58 6.74.00 Subtle sex discrimination .04.01 3.14.00 EMOTIONAL -.12.03 -3.36.00 Minority Status .45.63 .71.48 EMOTIONAL Minority Status -,03.04 -.62.53 Minority Status Indirect Effect SE z p Men -.01.01 -1.97.05 Women -.02.01 -2.74.01 The results for the separate tests of mediation were similar for both men and women, as support was found for the mediati on models in both groups. Moreover, all paths were in hypothesized directions. Table 258 displays the results for the simple tests of mediation.

PAGE 388

353 Table 258. Results of Simple Mediation for Men and Women (PREDICTOR = Emotional Social Support, Med = Subtle Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) -.13.03-3.81.00-.16.02 -7.10 .00 b(MX) -.33.12-2.69.01-.48.08 -5.82 .00 b(YM.X) .07.032.55.01.04.02 2.11 .04 b(YX.M) -.11.03-3.15.00-.15.02 -6.01 .00 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between emotional social s upport and physical healt h. Support for overall mediation was not found as the path between emotional social support and perceived discrimination was non-significant, but the path between the mediator and physical health ( B = -.29, t = -4.31, p < .01) was significant. No significant interaction was found between the predictor (emotiona l social support) and the mo derator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects indicated a difference between groups, as wome n exhibited a significa nt indirect effect (Indirect effect = .13, z = 3.32, p < .01), but the indirect effect was non-significant for men. Results are shown in Table 259.

PAGE 389

354 Table 259. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Subtle Discrimination, DV = Phys ical Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 15.321.98 7.73.00 EMOTIONAL -.26.13 -1.99.05 Minority Status 4.082.36 1.73.09 EMOTIONAL Minority Status -.19.15 -1.24.21 Physical Health Constant 65.602.68 24.47.00 Subtle sex discrimination -.29.07 -4.31.00 EMOTIONAL .32.16 1.99.05 Minority Status -.882.96 -.30.77 EMOTIONAL Minority Status -.05.19 -.24.81 Minority Status Indirect Effect SE z p Men .08.04 1.76.08 Women .13.04 3.32.00 The results for the separate tests of mediation are shown in Table 260. The mediation model was supported for women but not for men. However, all paths were in the hypothesized directions.

PAGE 390

355 Table 260. Results of Simple Mediation for Men and Women (PREDICTOR = Emotional Social Support, Med = Subtle Discrimination, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .40.152.62.01.41.11 3.68 .00 b(MX) -.26.13-1.95.05-.45.08 -5.38 .00 b(YM.X) -.22.11-1.92.06-.33.08 -3.88 .00 b(YX.M) .34.152.24.03.26.11 2.29 .02 Finally, perceived subtle sex-based discri mination was investigated as a mediator in the relationship between emotional soci al support and psychol ogical health. Support for overall mediation was found as the pa ths between emotional social support and perceived discrimination ( B = -.29, t = -2.28, p < .05) and between the mediator and psychological health ( B = -.18, t = -2.29, p < .05) were jointly signi ficant. Social support was related to less perceived discriminati on which in turn was related to better psychological health. No significant inte raction was found between the predictor (emotional social support) and th e moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects indicated a difference in minority status, as the indirect effect was significant for women (Indirect effect = .08, z = 2.08, p < .05) but not for men. Resu lts are shown in Table 261.

PAGE 391

356 Table 261. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Subtle Discrimination, DV = Psychological Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 15.531.92 8.09.00 EMOTIONAL -.29.12 -2.28.02 Minority Status 4.312.32 1.86.06 EMOTIONAL Minority Status -.18.15 -1.17.24 Psychological Health Constant 44.883.03 14.81.00 Subtle sex discrimination -.18.08 -2.29.02 EMOTIONAL .54.18 2.95.00 Minority Status -.913.38 -.27.79 EMOTIONAL Minority Status .00.22 .02.98 Minority Status Indirect Effect SE z p Men .05.03 1.54.12 Women .08.04 2.08.04 A significant mediating effect was found for women but not for men. However, all paths were in the hypothesized direct ions. Results are displayed in Table 262. Table 262. Results of Simple Mediation for Men and Women (PREDICTOR = Emotional Social Support, Med = Subtle Discriminati on, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) .59.183.20.00.62.12 5.04 .00 b(MX) -.29.11-2.53.01-.46.09 -5.20 .00 b(YM.X) -.00.16-.02.99-.24.09 -2.67 .01 b(YX.M) .59.193.09.00.52.13 3.99 .00

PAGE 392

357 Overt Sex-based Discrimination Emotional social support in the workplace was investigated as a predictor in a model wher e overt sex-based discri mination mediates the relationship between emoti onal social support and five dependent variables (job satisfaction, organizational commitment, inte ntion to turnover, physical health, and psychological health). Minority st atus was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between emotional social s upport and job satisfaction. Support for overall mediation was found as the paths between emotional social support and perceived discrimination ( B = -.85, t = -4.67, p < .01) and between the medi ator and job satisfaction ( B = -.04, t = -2.23, p < .05) were jointly si gnificant. Social support was related to less perceived discrimination which in turn wa s related to greater job satisfaction. No significant interaction was found between the pr edictor (emotional soci al support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. The indirect effects were similar for men and women, as both were similar in size and non-significant. Results are displayed in Table 263.

PAGE 393

358 Table 263. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 25.972.78 9.33.00 EMOTIONAL -.85.18 -4.67.00 Minority Status -3.763.37 -1.11.27 EMOTIONAL Minority Status .30.22 1.39.17 Job Satisfaction Constant 5.121.12 4.56.00 Overt sex discrimination -.04.02 -2.23.03 EMOTIONAL .42.07 6.11.00 Minority Status .891.23 .72.47 EMOTIONAL Minority Status -.05.08 -.58.56 Minority Status Indirect Effect SE z p Men .04.02 1.98.05 Women .02.01 1.95.05 The results for the separate tests of medi ation were similar for men and women as there was a failure to support the mediati on model in both groups. However, all paths were in the hypothesized direc tions. Table 264 shows the results for the simple tests of mediation.

PAGE 394

359 Table 264. Results of Simple Mediation for Men and Women (PREDICTOR = Emotional Social Support, Med = Overt Discrimina tion, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .45.076.57.00.39.04 8.77 .00 b(MX) -.85.17-4.97.00-.54.13 -4.28 .00 b(YM.X) -.06.04-1.62.11-.04.02 -1.62 .11 b(YX.M) .40.085.33.00.37.05 8.09 .00 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between emotional social support and organizational commitment. Support for overall mediation was found as th e paths between emotional social support and perceived discrimination ( B = -.84, t = -4.64, p < .01) and between the mediator and organizational commitment ( B = -.13, t = -2.46, p < .05) were jointly significant. Social support was related to less perc eived discrimination which in turn was related to greater organizational commitment. No significant interaction was found be tween the predictor (emotional social support) and th e moderator (minority status). Thus, minority status is not a moderator in the current model. Additio nally, the indirect effects for both men and women were similar in size and significant. Results for the moderated mediation are displayed in Table 265.

PAGE 395

360 Table 265. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 25.612.77 9.25.00 EMOTIONAL -.84.18 -4.64.00 Minority Status -3.383.34 -1.01.31 EMOTIONAL Minority Status .29.22 1.34.18 Organizational Constant 18.463.10 5.95.00 Overt sex discrimination -.13.05 -2.46.01 EMOTIONAL .84.19 4.49.00 Minority Status -1.473.38 -.44.66 EMOTIONAL Minority Status .19.22 .86.39 Minority Status Indirect Effect SE z p Men .11.05 2.14.03 Women .07.03 2.12.03 The results for the separate tests of me diation were different for men and women. While the mediation model was fully suppor ted for women, it was not supported for men. However, all paths were in the hypothesized directions. Table 266 displays the results for the simple tests of mediation.

PAGE 396

361 Table 266. Results of Simple Mediation for Men and Women (PREDICTOR = Emotional Social Support, Med = Overt Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) .95.175.54.001.10.13 8.68 .00 b(MX) -.84.17-5.02.00-.55.13 -4.34 .00 b(YM.X) -.06.10-.56.58-.16.06 -2.50 .01 b(YX.M) .91.194.75.001.02.13 7.81 .00 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between emotional social s upport and intention to turnover. Support for overall mediation was found as the paths betw een emotional social support and perceived discrimination ( B = -.85, t = -4.72, p < .01) and between the me diator and intention to turnover ( B = .04, t = 3.84, p < .01) were jointly significant. Social support was related to less perceived discrimination which in turn was related to les turnover intention. No significant interaction was found between the pr edictor (emotional soci al support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Further, the indirect effects were significant for both men and women. Table 267 displays the results for the moderated mediation.

PAGE 397

362 Table 267. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 25.972.75 9.43.00 EMOTIONAL -.85.18 -4.72.00 Minority Status -3.063.32 -.92.36 EMOTIONAL Minority Status .26.22 1.19.24 Turnover Constant 3.87.56 6.88.00 Overt sex discrimination .04.01 3.84.00 EMOTIONAL -.11.03 -3.27.00 Minority Status .34.61 .55.58 EMOTIONAL Minority Status -.02.04 -.52.61 Minority Status Indirect Effect SE z p Men -.03.01 -2.94.00 Women -.02.01 -2.98.00 The results for the separate tests of mediation are shown in Table 268. A significant mediating effect was found for both men and women. Moreover, all paths were hypothesized directions.

PAGE 398

363 Table 268. Results of Simple Mediation for Men and Women (PREDICTOR = Emotional Social Support, Med = Overt Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) -.14.03-4.26.00-.15.02 -6.74 .00 b(MX) -.85.17-4.97.00-.59.12 -4.77 .00 b(YM.X) .04.022.15.03.04.01 3.18 .00 b(YX.M) -.11.04-3.02.00-.13.02 -5.68 .00 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between emotional social s upport and physical healt h. Support for overall mediation was found as the paths between emotional social support and perceived discrimination ( B = -.72, t = -3.70, p < .01) and between the medi ator and physical health ( B = -.22, t = -4.85, p < .01) were jointly si gnificant. Social support was related to less perceived discrimination which in turn was related to better physical health. No significant interaction was found between the pr edictor (emotional soci al support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Additionally, the indir ect effects are significant for both men and women, and were similar in size. Results are shown in Table 269.

PAGE 399

364 Table 269. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Overt Discrimination, DV = Phys ical Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 23.922.99 8.00.00 EMOTIONAL -.72.19 -3.70.00 Minority Status -2.133.57 -.60.55 EMOTIONAL Minority Status .19.23 .82.41 Physical Health Constant 65.522.72 24.09.00 Overt sex discrimination -.22.04 -4.85.00 EMOTIONAL .28.16 1.68.09 Minority Status -1.312.97 -.44.66 EMOTIONAL Minority Status -.01.19 -.07.94 Minority Status Indirect Effect SE z p Men .16.05 2.90.00 Women .12.04 3.13.00 The results for the separate tests of mediation are shown in Table 270. A significant mediating effect was found for bot h men and women. All paths were in the hypothesized directions.

PAGE 400

365 Table 270. Results of Simple Mediation for Men and Women (PREDICTOR = Emotional Social Support, Med = Overt Discrimina tion, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .44.162.77.01.38.11 3.41 .00 b(MX) -.72.18-3.92.00-.53.13 -4.09 .00 b(YM.X) -.25.08-3.01.00-.21.05 -3.85 .00 b(YX.M) .26.161.59.11.27.11 2.40 .02 Finally, perceived overt sexbased discrimination was inve stigated as a mediator in the relationship between emotional soci al support and psychol ogical health. Support for overall mediation was found as the pa ths between emotional social support and perceived discrimination ( B = -.82, t = -4.48, p < .01) and between the mediator and psychological health ( B = -.16, t = -3.14, p < .01) were jointly signi ficant. Social support was related to less perceived discriminati on which in turn was related to better psychological health. No significant inte raction was found between the predictor (emotional social support) and th e moderator (minority status). Thus, minority status is not a moderator in the current model. The i ndirect effects are significant for both men and women, and were similar in size. Results are shown in Table 271.

PAGE 401

366 Table 271. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Overt Discrimination, DV = Psychological Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 25.302.82 8.97.00 EMOTIONAL -.82.18 -4.48.00 Minority Status -2.663.43 -.77.44 EMOTIONAL Minority Status .26.22 1.14.26 Psychological Health Constant 46.652.99 15.59.00 Overt sex discrimination -.16.05 -3.14.00 EMOTIONAL .43.18 2.37.02 Minority Status -1.533.29 -.46.64 EMOTIONAL Minority Status .04.22 .18.85 Minority Status Indirect Effect SE z p Men .13.05 2.53.01 Women .09.04 2.52.01 The results for the separate tests of me diation revealed a significant mediating effect for women, but not for men. Howeve r, all paths were in the hypothesized directions. Table 272 displays the results for the simple tests of mediation. Table 272. Results of Simple Mediation for Men and Women (PREDICTOR = Emotional Social Support, Med = Overt Discriminatio n, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) .56.183.16.00.56.13 4.47 .00 b(MX) -.82.17-4.91.00-.57.13 -4.24 .00 b(YM.X) -.05.10-.51.61-.20.06 -3.37 .00 b(YX.M) .52.202.64.01.45.13 3.52 .00

PAGE 402

367 Subtle Race-based Discrimination Emotional social support in the workplace was investigated as a predictor in a mode l where subtle race-based discrimination mediates the relationship between emotional social support and five dependent variables (job satisfaction, organizational commitment, in tention to turnover, physical health, and psychological health). Minority st atus was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived subtle race-based discri mination was investigat ed as a mediator in the relationship between emotional soci al support and job satisfaction. Support for overall mediation was not found as the path between emotional social support and perceived discrimination was not significant, but the path between the mediator and job satisfaction ( B = -.08, t = -3.07, p < .01) was significant. A significant interaction was found between the predictor (emotional social support) and the moderator ( B = -.34, t = 2.28, p < .05). Thus, minority status is a moderator in the current model. Additionally, the indirect effect results i ndicate a difference between wh ite and non-white participants, as there is non-significant effect for white participants, but a si gnificant effect for nonwhite participants (Indirect effect = .04, z = 2.49, p < .05). Results are shown in Table 273.

PAGE 403

368 Table 273. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 13.081.48 8.83.00 EMOTIONAL -.17.10 -1.71.09 Minority Status 7.612.31 3.30.00 EMOTIONAL Minority Status -.34.15 -2.28.02 Job Satisfaction Constant 5.37.83 6.50.00 Subtle race discrimination -.08.03 -3.07.00 EMOTIONAL .42.05 8.58.00 Minority Status 1.101.19 .93.35 EMOTIONAL Minority Status -.06.08 -.79.43 Minority Status Indirect Effect SE z p Whites .01.01 1.44.15 Non-whites .04.02 2.49.01 The results for the separate tests of mediation are shown in Table 274. A significant mediating effect was found for wo men but not for men. However, all paths were in the hypothesized directions.

PAGE 404

369 Table 274. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Emotional Social Support, Med = Subtle Discri mination, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) .44.059.10.00.41.06 6.69 .00 b(MX) -.17.08-2.07.04-.51.13 -3.83 .00 b(YM.X) -.07.04-1.64.10-.09.03 -2.57 .01 b(YX.M) .43.058.81.00.36.06 5.81 .00 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between emotional soci al support and organi zational commitment. Support for overall mediation was not found as the path between emo tional social support and perceived discrimination was not significan t but the path between the mediator and organizational commitment ( B = -.21, t = -2.86, p < .01) was significant. A significant interaction was found between th e predictor (emotional social support) and the moderator ( B = -.30, t = -2.01, p < .05). Thus, minority status is a moderator in the current model. The indirect effects are di fferent between white and nonwhite respondents. Non-white respondents exhibited a si gnificant indirect effect (Indirect effect = .10, z = 2.33, p < .05), whereas white respondents did not. Re sults are displayed in Table 275.

PAGE 405

370 Table 275. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 13.131.45 9.08.00 EMOTIONAL -.18.10 -1.86.06 Minority Status 6.952.25 3.08.00 EMOTIONAL Minority Status -.30.15 -2.01.04 Organizational Constant 15.682.25 6.98.00 Subtle race discrimination -.21.07 -2.86.00 EMOTIONAL 1.07.13 7.99.00 Minority Status 3.153.21 .98.33 EMOTIONAL Minority Status -.08.21 -.40.69 Minority Status Indirect Effect SE z p Whites .04.02 1.50.13 Non-whites .10.04 2.33.02 The results for the separate tests of mediation are shown in Table 276. A significant mediating effect was found for non-white participants but not for white participants. However, all paths we re in the hypothesized directions.

PAGE 406

371 Table 276. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Emotional Social Support, Med = Subtle Discrimina tion, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) 1.11.148.12.001.09.16 6.98 .00 b(MX) -.18.08-2.32.02-.47.13 -3.58 .00 b(YM.X) -.13.13-1.01.31-.25.09 -2.86 .00 b(YX.M) 1.09.147.84.00.97.16 6.12 .00 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between em otional social support and in tention to turnover. Support for overall mediation was not found as the path between emotiona l social support and perceived discrimination was not significant, but the path between the mediator and intention to turnover ( B = .07, t = 5.03, p < .01) was significant. A significant interaction was found between the predictor (emotiona l social support) and the moderator ( B = -.36, t = -2.41, p < .05). Thus, minority status is a modera tor in the current model. Additionally, the indirect effect was signi ficant for non-white participants (Indirect effect = -.04, z = 3.44, p < .01), but not for white participants. Results are shown in Table 277.

PAGE 407

372 Table 277. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 13.471.45 9.28.00 EMOTIONAL -.19.10 -1.97.05 Minority Status 7.772.28 3.41.00 EMOTIONAL Minority Status -.36.15 -2.41.02 Turnover Constant 4.58.40 11.36.00 Subtle race discrimination .07.01 5.03.00 EMOTIONAL -.17.02 -6.89.00 Minority Status -1.42.58 -2.46.01 EMOTIONAL Minority Status .08.04 2.03.04 Minority Status Indirect Effect SE z p Whites -.01.01 -1.81.07 Non-whites -.04.01 -3.44.00 The results for the separate tests of me diation indicated a significant mediating effect was found for both white and non-white pa rticipants. Moreover, all paths were in hypothesized directions. Results are shown in Table 278.

PAGE 408

373 Table 278. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Emotional Social Support, Med = Subtle Discrimina tion, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) -.18.02-7.40.00-.13.03 -4.15 .00 b(MX) -.19.08-2.36.02-.55.13 -4.17 .00 b(YM.X) .07.023.41.00.06.02 3.68 .00 b(YX.M) -.16.02-6.93.00-.09.03 -2.99 .00 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between emotional soci al support and physical health. Support for overall mediation was not found as the path between emotional social support and perceived discrimination was not significant, but the path between the mediator and physical health ( B = -.33, t = -4.89, p < .01) was significant. No significant interaction was found between the predicto r (emotional social support) and the moderator (minority status). Thus, minority status is not a moderator in the curren t model. The indirect effects indicated a difference between white and non-wh ite participants, as there were significant findings for non-white participants (Indirect effect = .13, z = 2.75, p < .05), but not for white participants. Results for the mode rated mediation are shown in Table 279.

PAGE 409

374 Table 279. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Subtle Discrimination, DV = Ph ysical Health, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 12.561.46 8.62.00 EMOTIONAL -.14.10 -1.43.15 Minority Status 6.042.29 2.64.09 EMOTIONAL Minority Status -.25.15 -1.68.09 Physical Health Constant 65.111.98 32.78.00 Subtle race discrimination -.33.07 -4.89.00 EMOTIONAL .29.12 2.45.01 Minority Status .452.84 .16.88 EMOTIONAL Minority Status -.03.18 -.17.87 Minority Status Indirect Effect SE z p Whites .05.03 1.35.18 Non-whites .13.05 2.75.01 The results for the separate tests of mediation are shown in Table 280. A significant mediating effect was found for non-white participants, but the mediation model was not fully supported for white part icipants. However, all paths were in the hypothesized directions.

PAGE 410

375 Table 280. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Emotional Social Support, Med = Subtle Discrimination, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) .33.112.96.00.39.16 2.47 .01 b(MX) -.14.08-1.68.10-.39.13 -2.95 .00 b(YM.X) -.28.10-2.78.01-.36.09 -3.91 .00 b(YX.M) .30.112.64.01.25.15 1.60 .11 Finally, perceived subtle race -based discrimination was i nvestigated as a mediator in the relationship between emotional soci al support and psychol ogical health. Support for overall mediation was not found as the path between emotiona l social support and perceived discrimination ( B = -.22, t = -2.18, p < .05) was significant, but the path between the mediator and psychological heal th was not significant. No significant interaction was found between th e predictor (emotional social support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Additionally, the indirect effects are similar fo r white and non-white re spondents, as they are both non-significant and similar in si ze. Results are displayed in Table 281.

PAGE 411

376 Table 281. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Subtle Discrimination, DV = Psychological Health, Mod = Race Predictor B SE t p Subtle race discrimination Constant 13.931.52 9.16.00 EMOTIONAL -.22.10 -2.18.03 Minority Status 6.192.40 2.58.01 EMOTIONAL Minority Status -.25.16 -1.62.11 Psychological Health Constant 41.932.27 18.50.00 Subtle race discrimination -.07.07 -.96.34 EMOTIONAL .61.13 4.53.00 Minority Status 1.913.24 .59.56 EMOTIONAL Minority Status -.14.21 -.69.49 Minority Status Indirect Effect SE z p Whites .02.02 .81.42 Non-whites .03.04 .90.37 The mediating effect was non-signif icant for both white and non-white participants. However, all paths were in the hypothesized directions. Results for the simple tests of mediation are shown in Table 282.

PAGE 412

377 Table 282. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Emotional Social Support, Med = Subtle Discrimi nation, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) .63.144.63.00.50.16 3.11 .00 b(MX) -.22.08-2.62.01-.47.14 -3.36 .00 b(YM.X) -.03.12-.22.82-.10.09 -1.05 .29 b(YX.M) .62.144.49.00.45.17 2.74 .01 Overt Race-based Discrimination Emotional social support in the workplace was investigated as a predictor in a model where overt race-based discrimination mediates the relationship between emoti onal social support and five dependent variables (job satisfaction, organizational commitment, inte ntion to turnover, physical health, and psychological health). Minority st atus was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between emotional social s upport and job satisfaction. Support for overall mediation was found as the paths between emotional social support and perceived discrimination ( B = -.34, t = -2.56, p < .05) and between the medi ator and job satisfaction ( B = -.07, t = -3.42, p < .01) were jointly si gnificant. Social support was related to less perceived discrimination which in turn wa s related to greater job satisfaction. No significant interaction was found between the pr edictor (emotional soci al support) and the moderator (minority status). Thus, minority status is not a moderator in the current

PAGE 413

378 model. The indirect effects were somewh at different between white and non-white participants. Although the indi rect effects were similar in size, the findings were significant for non-white particip ants (Indirect effect = .05, z = 2.65, p < .05) only. Results are shown in Table 283. Table 283. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Overt race discrimination Constant 17.052.01 8.50.00 EMOTIONAL -.34.13 -2.56.01 Minority Status 7.873.18 2.48.01 EMOTIONAL Minority Status -.36.21 -1.74.08 Job Satisfaction Constant 5.59.81 6.88.00 Overt race discrimination -.07.02 -3.42.00 EMOTIONAL .41.05 8.26.00 Minority Status .731.19 .62.54 EMOTIONAL Minority Status -.04.08 -.56.58 Minority Status Indirect Effect SE z p Whites .02.01 2.00.05 Non-whites .05.02 2.65.01 The results for the separate tests of mediation revealed full support for the mediation model in white participants, but not in non-white participants. However, all paths were in the hypothesized directions. Ta ble 284 displays the results for the simple tests of mediation.

PAGE 414

379 Table 284. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Emotional Social Support, Med = Overt Discri mination, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) .43.058.86.00.41.06 6.65 .00 b(MX) -.34.12-2.70.01-.70.17 -4.12 .00 b(YM.X) -.09.03-3.52.00-.04.03 -1.42 .16 b(YX.M) .40.058.28.00.38.06 5.93 .00 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between emotional social support and organizational commitment. Support for overall mediation was found as th e paths between emotional social support and perceived discrimination ( B = -.35, t = -2.73, p < .05) and between the mediator and organizational commitment ( B = -.17, t = -3.18, p < .01) were jointly significant. Social support was related to less perc eived discrimination which in turn was related to greater organizational commitment. No significant interaction was found be tween the predictor (emotional social support) and th e moderator (minority status). Thus, minority status is not a moderator in the current model. The i ndirect effects are significant for both white and non-white participants, and were similar in size. Table 285 displays the results for the moderated mediation.

PAGE 415

380 Table 285. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Overt race discrimination Constant 17.211.98 8.69.00 EMOTIONAL -.35.13 -2.73.01 Minority Status 7.303.14 2.32.02 EMOTIONAL Minority Status -.31.20 -1.53.13 Organizational Constant 16.792.19 7.66.00 Overt race discrimination -.17.05 -3.18.00 EMOTIONAL 1.00.13 7.52.00 Minority Status 2.093.19 .66.51 EMOTIONAL Minority Status -.04.21 -.20.84 Minority Status Indirect Effect SE z p Whites .06.03 2.01.04 Non-whites .11.04 2.50.01 The results for the separate tests of medi ation were somewhat different for white and non-white participants. The mediation model is fully supported for white participants but is approaching significance in non-white pa rticipants. However, all paths were in the hypothesized directions. Results are displayed in Table 286.

PAGE 416

381 Table 286. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Emotional Social Support, Med = Overt Discrimina tion, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) 1.06.147.79.001.07.16 6.77 .00 b(MX) -.35.12-2.92.00-.67.17 -3.93 .00 b(YM.X) -.20.08-2.54.01-.14.07 -1.97 .05 b(YX.M) .99.147.21.00.97.16 5,96 .00 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between emotional social s upport and intention to turnover. Support for overall mediation was found as the paths betw een emotional social support and perceived discrimination ( B = -.40, t = -3.04, p < .01) and between the me diator and intention to turnover ( B = .04, t = 4.47, p < .01) were jointly significant. Social support was related to less perceived discrimination which in turn was related to less turnover intention. No significant interaction was found between the pr edictor (emotional soci al support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Further, the indirect effects we re significant for both white and non-white participants, and were similar in size Results are displayed in Table 287.

PAGE 417

382 Table 287. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Overt race discrimination Constant 18.011.98 9.12.00 EMOTIONAL -.40.13 -3.04.00 Minority Status 8.013.15 2.54.01 EMOTIONAL Minority Status -.37.21 -1.81.07 Turnover Constant 4.51.40 11.32.00 Overt race discrimination .04.01 4.47.00 EMOTIONAL -.15.02 -6.25.00 Minority Status -1.00.58 -1.73.08 EMOTIONAL Minority Status .06.04 1.48.14 Minority Status Indirect Effect SE z p Whites -.02.01 -2.48.01 Non-whites -.03.01 -3.24.00 For the separate tests of mediation, a significant mediating effect was found for white participants, but the mediation mode l was only partially supported for non-white participants. All paths were in hypothesized directions. Results are displayed in Table 288.

PAGE 418

383 Table 288. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Emotional Social Support, Med = Overt Discrimina tion, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) -.17.02-6.99.00-.13.03 -4.14 .00 b(MX) -.40.13-3.16.00-.77.17 -4.62 .00 b(YM.X) .06.014.44.00.03.01 1.97 .05 b(YX.M) -.14.02-6.17.00-.11.03 -3.28 .00 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between emotional social s upport and physical healt h. Support for overall mediation was found as the paths between emotional social support and perceived discrimination ( B = -.33, t = -2.44, p < .05) and between the medi ator and physical health ( B = -.29, t = -6.28, p < .01) were jointly si gnificant. Social support was related to less perceived discrimination which in turn was related to better physical health. No significant interaction was found between the pr edictor (emotional soci al support) and the moderator (minority status). Thus, minority status is not a moderator in the current model. Further, the indirect effects we re significant for both white and non-white participants, and were similar in size Results are displayed in Table 289.

PAGE 419

384 Table 289. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Overt Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 16.852.07 8.16.00 EMOTIONAL -.33.13 -2.44.02 Minority Status 7.193.30 2.18.03 EMOTIONAL Minority Status -.32.21 -1.46.14 Physical Health Constant 65.071.89 34.48.00 Overt race discrimination -.29.05 -6.28.00 EMOTIONAL .29.11 2.59.01 Minority Status .802.77 .29.77 EMOTIONAL Minority Status -.07.18 -.37.71 Minority Status Indirect Effect SE z p Whites .09.04 2.25.02 Non-whites .19.06 3.25.00 The results for the separate tests of me diation revealed a significant mediating effect was found for white and non-white pa rticipants. All path s were hypothesized directions. Table 290 displays the results for the simple tests of mediation.

PAGE 420

385 Table 290. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Emotional Social Support, Med = Overt Discrimination, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) .39.113.53.00.41.16 2.59 .01 b(MX) -.33.13-2.54.01-.64.18 -3.67 .00 b(YM.X) -.25.06-4.21.00-.32.07 -4.60 .00 b(YX.M) .31.112.84.01.20.16 1.31 .19 Finally, perceived overt race-based discrimi nation was investigated as a mediator in the relationship between emotional soci al support and psychol ogical health. Support for overall mediation was found as the pa ths between emotional social support and perceived discrimination ( B = -.44, t = -3.21, p < .01) and between the mediator and psychological health ( B = -.13, t = -2.59, p < .05) were jointly signi ficant. Social support was related to less perceived discriminati on which in turn was related to better psychological health. No significant inte raction was found between the predictor (emotional social support) and th e moderator (minority status). Thus, minority status is not a moderator in the current model. The indire ct effects were slightly different for white and non-white participants; the indirect effect was significan t for non-white participants (Indirect effect = .09, z = 2.10, p < .05), but approaching significance for white participants. Results are shown in Table 291.

PAGE 421

386 Table 291. Results of Moderated Mediation (PREDIC TOR = Emotional Social Support, Med = Overt Discrimination, DV = Psychological Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 18.742.08 9.01.00 EMOTIONAL -.44.14 -3.21.00 Minority Status 5.713.34 1.71.09 EMOTIONAL Minority Status -.22.22 -1.01.31 Psychological Health Constant 43.412.19 19.79.00 Overt race discrimination -.13.05 -2.59.01 EMOTIONAL .58.13 4.41.00 Minority Status 2.113.19 .66.51 EMOTIONAL Minority Status -.15.21 -.74.46 Minority Status Indirect Effect SE z p Whites .06.03 1.96.05 Non-whites .09.04 2.10.04 The results for the separate tests of medi ation displayed support for the mediation model in non-white participants but not in white participants However, all paths were in the hypothesized directions. Table 292 shows the results for the simple tests of mediation.

PAGE 422

387 Table 292. Results of Simple Mediation for Whit es and Non-whites (PREDICTOR = Emotional Social Support, Med = Overt Discrimi nation, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) .64.134.84.00.51.16 3.14 .00 b(MX) -.44.13-3.36.00-.66.18 -3.65 .00 b(YM.X) -.11.07-1.51.13-.15.07 -2.16 .03 b(YX.M) .59.144.37.00.41.17 2.44 .02 Results for Token Status Subtle Sex-based Discrimination Token status in the workplace was investigated as a predictor in a model where subtle sexbased discrimination medi ates the relationship between token status and five dependent variables (job satisfaction, organizational commitment, intention to turnover, physical health, and psychological health). Minority status was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived subtle sex-based discrimi nation was investigated as a mediator in the relationship between token status and job satisfaction. Support for overall mediation was not found as the path between token st atus and perceived discrimination was not significant and the path between th e mediator and job satisfaction ( B = -.14, t = -4.70, p < .01) was significant. A significant interacti on was not found between the predictor (token status) and the moderator (minority status), how ever the interaction effect was marginally significant ( B = 4.23, t = 1.97, p = .05). The indirect effects are different between men

PAGE 423

388 and women, as there is a significant effect for wome n (Indirect effect = -.69, z = -2.48, p < .05) but not for men. Results are shown in Table 293. Table 293. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 11.33.45 25.02.00 TOKEN STATUS .671.37 .49.63 Minority Status 1.32.54 2.45.01 TOKEN STATUS Minority Status 4.232.14 1.97.05 Job satisfaction Constant 12.37.44 28.37.00 Subtle sex discrimination -.14.03 -4.70.00 TOKEN STATUS .32.83 .39.70 Minority Status .31.33 .95.35 TOKEN STATUS Minority Status .581.30 .45.66 Minority Status Indirect Effect SE z p Men -.09.20 -.47.64 Women -.69.28 -2.48.01 The results for the separate tests of mediation indicated full support for the mediation model for women, but not for men. Paths are not in the hypothesized directions. The path coefficients were pred icted to differ between minority and majority group members; however the directions are the same. Specifically, the path between token status and perceived discrimination was predicted to be positive for minority individuals and negative for majority indivi duals. However, it was found to positive for both groups. Results are shown in Table 294.

PAGE 424

389 Table 294. Results of Simple Mediation for Men and Women (PREDICTOR = Token Status, Med = Subtle Discrimination, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .23.89.25.80.221.00 .22 .83 b(MX) .671.30.52.614.901.68 2.91 .00 b(YM.X) -.19.06-3.32.00-.12.03 -3.47 .00 b(YX.M) .36.86.42.68.801.00 .81 .42 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between token status and organizational co mmitment. Support for overall mediation was not found as the path between token status and perc eived discrimination was non-significant and the path between th e mediator and organizational commitment ( B = -.36, t = -4.29, p < .01) was significant. No significant interaction was found between the predictor (token status) and the moderator (mi nority status). However, the interaction effect was marginally significant ( B = 4.04, t = 1.92, p = .06). The indirect effects were different between men and women, as the indire ct effect was significant for women (Indirect effect = -1.75, z = -2.45, p < .05), but not for men. Table 295 displays the results for the moderated mediation.

PAGE 425

390 Table 295. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 11.11.45 24.75.00 TOKEN STATUS .891.35 .66.51 Minority Status 1.52.53 2.85.00 TOKEN STATUS Minority 4.042.10 1.92.06 Organizational Constant 33.331.19 28.07.00 Subtle sex discrimination -.36.08 -4.29.00 TOKEN STATUS -.562.25 -.25.80 Minority Status 1.87.90 2.09.04 TOKEN STATUS Minority -.623.52 -.18.86 Minority Status Indirect Effect SE z p Men -.32.50 -.64.52 Women -1.75.72 -2.45.01 The results for the separate tests of me diation displayed full support for the model in women, but not in men. Paths are not in the hypothesized di rections. The path coefficients were predicted to differ between minority a nd majority group members; however the directions are the same. Specifi cally, the path between token status and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals. However, it was found to positive for both groups. Table 296 displays the results for the simple tests of mediation.

PAGE 426

391 Table 296. Results of Simple Mediation for Men and Women (PREDICTOR = Token Status, Med = Subtle Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) -.882.13-.41.68-2.932.82 -1.04 .30 b(MX) .891.22.73.474.931.67 2.95 .00 b(YM.X) -.32.16-2.08.04-.37.10 -3.75 .00 b(YX.M) -.592.11-.28.78-1.122.80 -.40 .69 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between token status and intention to turnov er. Support for overall mediation was not found as the path between token status and perc eived discrimination was not significant, but the path between the mediator and intention to turnover ( B = .07, t = 5.14, p < .01) was significant. No signifi cant interaction was found between the predictor (token status) and the moderator (minority status).Howe ver, the interaction effect was marginally significant ( B = 4.26, t = 1.99, p = .05). The indirect effects indicate a difference between men and women, as there is a significant effect for women (Indirect effect = .35, z = 2.55, p < .05), but the effect is non -significant for men. Results are displayed in Table 297.

PAGE 427

392 Table 297. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 11.35.45 24.97.00 TOKEN STATUS .651.37 .47.64 Minority Status 1.29.54 2.39.02 TOKEN STATUS Minority Status 4.262.14 1.99.05 Turnover Constant 1.82.20 8.92.00 Subtle sex discrimination .07.01 5.14.00 TOKEN STATUS .10.39 .25.80 Minority Status -.00.15 -.01.99 TOKEN STATUS Minority Status -.07.61 -.12.91 Minority Status Indirect Effect SE z p Men .05.10 .46.64 Women .35.14 2.55.01 The results for the separate tests of mediation indicate full support for the mediation model in women but no support fo r men. Paths are not in the hypothesized directions. The path coefficients were pred icted to differ between minority and majority group members; however the directions are the same. Specifically, the path between token status and perceived discrimination was predicted to be positive for minority individuals and negative for majority indivi duals. However, it was found to positive for both groups. Table 298 displays the results for the simple te sts of mediation.

PAGE 428

393 Table 298. Results of Simple Mediation for Men and Women (PREDICTOR = Token Status, Med = Subtle Discrimination, DV = Intention to Turnover) Men Women B SE t p B SE t p b(YX) .15.39.38.71.38.49 .78 .44 b(MX) .651.30.50.624.911.68 2.92 .00 b(YM.X) .09.033.68.00.06.02 3.80 .00 b(YX.M) .09.37.23.82.07.48 .14 .89 Next, perceived subtle sex-ba sed discrimination was invest igated as a mediator in the relationship between token status and physical health. Support for overall mediation was not found as the path between token st atus and perceived discrimination was nonsignificant, but the path between th e mediator and physical health ( B = -.34, t = -5.29, p < .01) was significant. A significant interac tion was found between the predictor (token status) and the moderator ( B = 5.16, t = 2.37, p < .05). Thus, minority status is a moderator in the current model. The indir ect effects are supportive of a difference between men and women. The indirect effect wa s significant for wome n (Indirect effect = -1.73, z = -2.62, p < .05), but not for men. Resu lts are shown in Table 299.

PAGE 429

394 Table 299. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Subtle Discrimination, DV = Physical Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 11.38.46 24.50.00 TOKEN STATUS .001.38 .00.99 Minority Status 1.07.55 1.95.05 TOKEN STATUS Minority 5.162.18 2.37.02 Physical Health Constant 70.67.92 77.20.00 Subtle sex discrimination -.34.06 -5.29.00 TOKEN STATUS .531.67 .32.75 Minority Status -1.19.67 -1.76.08 TOKEN STATUS Minority -.732.66 -.27.78 Minority Status Indirect Effect SE z p Men -.00.47 -.00.99 Women -1.73.66 -2.62.01 The results of the simple tests of mediat ion revealed full support of the mediation model in women, but no support for men. Als o, the path between token status and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals. However, it was found to positive for both groups. Table 300 displays these results.

PAGE 430

395 Table 300. Results of Simple Mediation for Men and Women (PREDICTOR = Token Status, Med = Subtle Discrimination, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .531.63.33.74-1.932.17 -.89 .38 b(MX) .001.35.00.995.171.70 3.03 .00 b(YM.X) -.23.11-2.10.04-.38.08 -4.92 .00 b(YX.M) .531.60.33.74.022.12 .01 .99 Finally, perceived subtle sex-based discri mination was investigated as a mediator in the relationship between token status and psychological hea lth. Support for overall mediation was not found as the path between token status and perc eived discrimination was not significant, but th e path between the mediator and psychological health ( B = -.29, t = -3.82, p < .01) was significant. A significan t interaction was found between the predictor (token status) and the moderator ( B = 4.96, t = 2.21, p < .05). Thus, minority status is a moderator in the current model. A dditionally, the indirect effect was significant for women (Indirect effect = -1.55, z = -2.39, p < .05) only. Results are shown in Table 301.

PAGE 431

396 Table 301. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Subtle Discrimination, DV = Psychologi cal Health, Mod = Gender) Predictor B SE t p Subtle sex discrimination Constant 11.13.45 24.64.00 TOKEN STATUS .461.45 .32.75 Minority Status 1.58.54 2.92.00 TOKEN STATUS Minority Status 4.962.24 2.21.03 Psychological Health Constant 54.221.06 50.96.00 Subtle sex discrimination -.29.08 -3.82.00 TOKEN STATUS .102.12 .05.96 Minority Status -.74.80 -.92.35 TOKEN STATUS Minority Status -3.003.30 -.91.36 Minority Status Indirect Effect SE z p Men -.13.43 -.31.76 Women -1.55.65 -2.39.02 The results for the separate tests of medi ation indicated support for the mediation model in women, but not in men. Paths are not in the hypothesized directions. The path coefficients were predicted to differ between minority a nd majority group members; however the directions are the same. Specifi cally, the path between token status and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals. However, it was found to positive for both groups. Table 302 displays the results for th e simple tests of mediation.

PAGE 432

397 Table 302. Results of Simple Mediation for Men and Women (PREDICTOR = Token Status, Med = Subtle Discrimination, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) -.032.16-.01.99-4.462.54 -1.76 .08 b(MX) .461.27.36.725.421.80 3.02 .00 b(YM.X) -.17.15-1.12.26-.32.09 -3.79 .00 b(YX.M) .052.16.02.98-2.702.52 -1.07 .28 Overt Sex-based Discrimination Token status in the workplace was investigated as a predictor in a model where overt sex-ba sed discrimination medi ates the relationship between token status and five dependent variables (job satisfaction, organizational commitment, intention to turnover, physical health, and psychological health). Minority status was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between token status and job satisfaction. Support for overall mediation was not found as the path between token st atus and perceived discrimination was not significant, but the path between th e mediator and job satisfaction ( B = -.10, t = -5.36, p < .01) was significant. A significant interac tion was found between the predictor (token status) and the moderator ( B = 10.64, t = 3.34, p < .01). Thus, minority status is a moderator in the current model. Further, th e indirect effect is significant for women

PAGE 433

398 (Indirect effect = -1.18, z = -3.46, p < .01) but not for men. Results are shown in Table 303. Table 303. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 13.24.67 19.89.00 TOKEN STATUS .622.04 .30.76 Minority Status .39.79 .49.62 TOKEN STATUS Minority 10.643.19 3.34.00 Job Satisfaction Constant 12.11.37 32.74.00 Overt sex discrimination -.10.02 -5.36.00 TOKEN STATUS .34.81 .41.68 Minority Status .35.32 1.11.27 TOKEN STATUS Minority .921.29 .71.48 Minority Status Indirect Effect SE z p Men -.07.22 -.30.76 Women -1.18.34 -3.46.00 The results for the separate tests of mediation revealed full support of the mediation model in women, but not for men. Pa ths are not in the hypothesized directions. The path coefficients were predicted to differ between minority and majority group members; however the directions are the same. Specifically, the path between token status and perceived discrimina tion was predicted to be positi ve for minority individuals and negative for majority individuals. Howe ver, it was found to positive for both groups. Results are displayed in Table 304.

PAGE 434

399 Table 304. Results of Simple Mediation for Men and Women (PREDICTOR = Token Status, Med = Overt Discrimination, DV = Job Satisfaction) Men Women B SE t p B SE t p b(YX) .27.89.30.76.07.98 .07 .94 b(MX) .622.05.30.7611.262.44 4.61 .00 b(YM.X) -.17.04-4.69.00-.08.02 -3.29 .00 b(YX.M) .37.83.45.65.93.99 .94 .35 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between token status and organizational co mmitment. Support for overall mediation was not found as the path between token status and perc eived discrimination was not significant, but the path between the mediator a nd organizational commitment ( B = -.27, t = -4.98, p < .01) was significant. A significan t interaction was found between the predictor (token status) and the moderator ( B = 10.44, t = 3.30, p < .01). Thus, minority status is a moderator in the current model. Additionally, there was a significant indirect effect for women (Indirect effect = -3.05, z = -3.36, p < .01), but not for men. Results are displayed in Table 305.

PAGE 435

400 Table 305. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 13.03.67 19.48.00 TOKEN STATUS .832.03 .41.68 Minority Status .59.80 .74.46 TOKEN STATUS Minority 10.443.17 3.30.00 Organizational Constant 32.841.02 32.15.00 Overt sex discrimination -.27.05 -4.98.00 TOKEN STATUS -.592.24 -.26.79 Minority Status 1.79.88 2.04.04 TOKEN STATUS Minority .483.53 .14.89 Minority Status Indirect Effect SE z p Men -.23.56 -.40.69 Women -3.05.91 -3.36.00 The results for the separate tests of me diation indicate support for the mediation model in women, but not for men. Paths are no t in the hypothesized directions. The path coefficients were predicted to differ between minority a nd majority group members; however the directions are the same. Specifi cally, the path between token status and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals. However, it was found to positive for both groups. Table 306 displays the results for the simple tests of mediation.

PAGE 436

401 Table 306. Results of Simple Mediation for Men and Women (PREDICTOR = Token Status, Med = Overt Discrimination, DV = Organizational Commitment) Men Women B SE t p B SE t p b(YX) -.812.19-.37.71-3.162.81 -1.13 .26 b(MX) .832.01.41.6811.272.44 4.62 .00 b(YM.X) -.30.09-3.26.00-.26.07 -3.84 .00 b(YX.M) -.562.11-.27.79-.272.84 -.09 .92 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between token status and intention to turnov er. Support for overall mediation was not found as the path between token status and perc eived discrimination was not significant, but the path between the mediator and intention to turnover ( B = .05, t = 6.01, p < .01) was significant. A significan t interaction was found between the predictor (token status) and the moderator ( B = 10.62, t = 3.34, p < .01). Thus, minority status is a moderator in the current model. The indirect effect wa s significant for women (Indirect effect = .62, z = 3.62, p < .01), but was non-significa nt for men. Results are shown in Table 307.

PAGE 437

402 Table 307. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 13.24.66 19.98.00 TOKEN STATUS .622.03 .30.76 Minority Status .42.79 .53.60 TOKEN STATUS Minority 10.623.17 3.34.00 Turnover Constant 1.92.17 11.05.00 Overt sex discrimination .05.01 6.01.00 TOKEN STATUS .10.38 .26.80 Minority Status .00.15 .01.99 TOKEN STATUS Minority -.29.60 -.47.64 Minority Status Indirect Effect SE z p Men .03.11 .30.76 Women .62.17 3.62.00 There was support for the mediation model in women, but all paths were not significant for men. Paths are not in the hypothe sized directions. The path coefficients were predicted to differ between minority and majority group members; however the directions are the same. Specifically, the path between token st atus and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals. However, it was found to positive for both groups. Table 308 displays the results of the simple te sts of mediation for men and women.

PAGE 438

403 Table 308. Results of Simple Mediation for Men and Women (PREDICTOR = Token Status, Med = Overt Discrimination, DV = Intention to Turnover Men Women B SE t p B SE t p b(YX) .13.39.34.73.43.48 .90 .37 b(MX) .622.05.30.7611.242.42 4.63 .00 b(YM.X) .07.024.18.00.05.01 4.45 .00 b(YX.M) .09.37.25.80-.13.48 -.27 .78 Next, perceived overt sex-based discrimina tion was investigated as a mediator in the relationship between token status and physical health. Support for overall mediation was not found as the path between token st atus and perceived discrimination was not significant, but the path between th e mediator and physical health ( B = -.25, t = -5.88, p < .01) was significant. A significant interac tion was found between the predictor (token status) and the moderator ( B = 11.21, t = 3.43, p < .01). Thus, minority status is a moderator in the current model. Additionall y, the indirect effect was significant for women (Indirect effect = -2.78, z = -3.48, p < .01) but not for men. Results are displayed in Table 309.

PAGE 439

404 Table 309. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Overt Discrimination, DV = Physical Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 13.01.69 18.88.00 TOKEN STATUS -.162.07 -.08.94 Minority Status .44.82 .54.59 TOKEN STATUS Minority 11.213.26 3.43.00 Physical Health Constant 69.87.79 88.15.00 Overt sex discrimination -.25.04 -5.88.00 TOKEN STATUS .751.70 .44.66 Minority Status -1.15.67 -1.70.09 TOKEN STATUS Minority .072.72 .03.98 Minority Status Indirect Effect SE z p Men .04.53 .08.94 Women -2.78.80 -3.48.00 The results for the separate tests of mediation are displa yed in Table 310. A significant mediating effect was found for wo men but not for men. All paths were in hypothesized directions.

PAGE 440

405 Table 310. Results of Simple Mediation for Men and Women (PREDICTOR = Token Status, Med = Overt Discrimination, DV = Physical Health) Men Women B SE t p B SE t p b(YX) .791.70.47.64-1.962.20 -.89 .37 b(MX) -.162.02-.08.9411.052.55 4.34 .00 b(YM.X) -.27.07-3.70.00-.24.05 -4.63 .00 b(YX.M) .741.61.46.64.712.20 .32 .75 Finally, perceived overt sexbased discrimination was inve stigated as a mediator in the relationship between token status and psychological hea lth. Support for overall mediation was not found as the path between token status and perc eived discrimination was not significant, but th e path between the mediator and psychological health ( B = -.23, t = -4.64, p < .01) was significant. A significan t interaction was found between the predictor (token status) and the moderator ( B = 12.43, t = 3.68, p < .01). Thus, minority status is a moderator in the current model. A dditionally, the indirect effect was significant for women (Indirect effect = -2.85, z = -3.32, p < .01) but not for men. Results are displayed in Table 311.

PAGE 441

406 Table 311. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Overt Discrimination, DV = Psychologi cal Health, Mod = Gender) Predictor B SE t p Overt sex discrimination Constant 13.06.67 19.42.00 TOKEN STATUS .112.18 .05.96 Minority Status .65.81 .81.42 TOKEN STATUS Minority 12.433.38 3.68.00 Psychological Health Constant 54.03.91 59.28.00 Overt sex discrimination -.23.05 -4.64.00 TOKEN STATUS -.032.10 -.01.99 Minority Status -1.04.78 -1.33.18 TOKEN STATUS Minority -1.603.31 -.48.63 Minority Status Indirect Effect SE z p Men -.02.51 -.05.96 Women -2.85.86 -3.32.00 The results for the separate tests of medi ation displayed support for the model in women, not for men. Paths are not in the hypothe sized directions. The path coefficients were predicted to differ between minority and majority group members; however the directions are the same. Specifically, the path between token st atus and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals. However, it was found to positive for both groups. Table 312 shows the results for the simple tests of mediation.

PAGE 442

407 Table 312. Results of Simple Mediation for Men and Women (PREDICTOR = Token Status, Med = Overt Discrimination, DV = Psychological Health) Men Women B SE t p B SE t p b(YX) -.052.14-.02.98-4.482.56 -1.75 .08 b(MX) .112.15.05.9612.532.60 4.83 .00 b(YM.X) -.19.09-2.15.03-.24.06 -4.15 .00 b(YX.M) -.032.11-.02.99-1.422.59 -.55 .59 Subtle Race-based Discrimination Token status in the workplace was investigated as a predictor in a model where subtle sex-based discri mination mediates the relationship between token status and fi ve dependent variables (job satisfaction, organizational commitment, intention to tu rnover, physical hea lth, and psychological health). Minority status was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived subtle race-based discrimi nation was investigated as a mediator in the relationship between token status and job satisfaction. Support for overall mediation was found as the paths between token st atus and perceived discrimination ( B = 6.25, t = 3.24, p < .01) and between the mediator and job satisfaction ( B = -.13, t = -4.67, p < .01) were jointly significant. A si gnificant interaction was found be tween the predictor (token status) and the moderator ( B = -6.79, t = -3.25, p < .01). Thus, minority status is a moderator in the current model. The indirect effect is significant for white participants

PAGE 443

408 (Indirect effect = -.83, z = -2.61, p < .05), but not for non-white participants. Results are displayed in Table 313. Table 313. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Subtle Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 10.32.34 30.40.00 TOKEN STATUS 6.251.93 3.24.00 Minority Status 2.77.56 4.99.00 TOKEN STATUS Minority -6.792.09 -3.25.00 Job Satisfaction Constant 12.28.35 34.79.00 Subtle race discrimination -.13.03 -4.67.00 TOKEN STATUS -.941.13 -.83.41 Minority Status .41.33 1.24.22 TOKEN STATUS Minority .971.22 .79.43 Minority Status Indirect Effect SE z p Whites -.83.32 -2.61.01 Non-whites .07.11 .66.51 The results for the separate tests of medi ation indicated support for the mediation model in white participants, but not in non-wh ite participants. Moreover, the paths are not in hypothesized directions. The relationshi p between token stat us and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals, however the opposite was found. Results are displayed in Table 314.

PAGE 444

409 Table 314. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Token Status, Med = Subtle Discrimination, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) -1.771.15-1.54.12.10.47 .21 .84 b(MX) 6.251.514.15.00-.54.96 -.56 .58 b(YM.X) -.14.05-2.87.00-.13.03 -3.71 .00 b(YX.M) -.881.17-.75.46.03.45 .06 .95 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between token status and organizational commitment. Support for overall mediation was found as the path s between token status and perceived discrimination ( B = 6.35, t = 3.35, p < .01) and between the me diator and organizational commitment ( B = -.32, t = -4.02, p < .01) were jointly sign ificant. No significant interaction was found between the predictor (token st atus) and the moderator ( B = -6.43, t = -3.13, p < .01). Thus, minority status is a modera tor in the current model. The indirect effect was significant for white par ticipants (Indirect effect = -2.03, z = -2.53, p < .05), but not for non-white particip ants. Results of the modera ted mediation are shown in Table 315.

PAGE 445

410 Table 315. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Subtle Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 10.22.33 30.59.00 TOKEN STATUS 6.351.89 3.35.00 Minority Status 2.67.55 4.86.00 TOKEN STATUS Minority -6.432.05 -3.13.00 Organizational Constant 32.90.97 33.76.00 Subtle race discrimination -.32.08 -4.02.00 TOKEN STATUS -4.883.08 -1.58.11 Minority Status 2.76.91 3.03.00 TOKEN STATUS Minority 3.873.33 1.16.25 Minority Status Indirect Effect SE z p Whites -2.03.80 -2.53.01 Non-whites .03.26 .10.92 The results for the separate tests of me diation indicate a lack of support of the mediation model for both white and non-white participants. However, the mediation model was marginally supported in white par ticipants. Moreover, the paths are not in hypothesized directions. The relationship between token status and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals, however the opposite was found. Results are displayed in Table 316.

PAGE 446

411 Table 316. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Token Status, Med = Subtle Discrimination, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) -6.913.14-2.20.03-.981.25 -.78 .44 b(MX) 6.351.444.42.00-.08.96 -.08 .93 b(YM.X) -.28.14-1.94.05-.34.09 -3.63 .00 b(YX.M) -5.123.26-1.57.12-1.011.21 -.83 .41 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between token status a nd intention to turnover. Support for overall mediation was found as the paths between t oken status and percei ved discrimination ( B = 6.19, t = 3.20, p < .01) and between the mediator and intention to turnover ( B = .08, t = 6.37, p < .01) were jointly significant. A signifi cant interaction was found between the predictor (token status) and the moderator ( B = -6.44, t = -3.08, p < .01). Thus, minority status is a moderator in the current model. A dditionally, the indirect effect was significant for white participants (Indirect effect = .52, z = 2.83, p < .01), but not for non-white participants. Results are displayed in Table 317.

PAGE 447

412 Table 317. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Subtle Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 10.38.34 30.57.00 TOKEN STATUS 6.191.93 3.20.00 Minority Status 2.62.56 4.70.00 TOKEN STATUS Minority -6.442.09 -3.08.00 Turnover Constant 1.89.16 11.61.00 Subtle race discrimination .08.01 6.37.00 TOKEN STATUS .43.52 .83.41 Minority Status -.38.15 -2.52.01 TOKEN STATUS Minority -.45.56 -.81.42 Minority Status Indirect Effect SE z p Whites .52.18 2.83.00 Non-whites -.02.07 -.32.75 The results for the separate tests of me diation were indicated full support of the mediation model in white participants, but not in non-white participants. Moreover, the paths are not in hypothesized directions. Th e relationship between token status and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals, however the opposite was found. Re sults are displayed in Table 318.

PAGE 448

413 Table 318. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Token Status, Med = Subtle Discrimination, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) .95.541.76.08.04.22 -.19 .85 b(MX) 6.191.534.05.00-.25.95 -.27 .79 b(YM.X) .10.024.40.00.07.02 4.71 .00 b(YX.M) .33.54.62.54-.02.21 -.10 .92 Next, perceived subtle race-based discrimi nation was investigat ed as a mediator in the relationship between token status and physical health. Support for overall mediation was found as the paths between t oken status and percei ved discrimination ( B = 4.86, t = 2.48, p < .05) and between the mediator and physical health ( B = -.33, t = -5.15, p < .01) were jointly significant. A si gnificant interaction was found between the predictor (token status) and the moderator ( B = -5.85, t = -2.76, p < .05). Thus, minority status is a moderator in the current model. A dditionally, the indirect effect was significant for white respondents (Indirect effect = -1.59, z = -2.20, p < .05) only. Table 319 displays the results for the moderated mediation.

PAGE 449

414 Table 319. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Subtle Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 10.30.33 30.83.00 TOKEN STATUS 4.861.96 2.48.01 Minority Status 2.74.56 4.87.00 TOKEN STATUS Minority -5.852.12 -2.76.01 Physical Health Constant 69.58.77 90.16.00 Subtle race discrimination -.33.06 -5.15.00 TOKEN STATUS -3.442.41 -1.42.15 Minority Status -.79.71 -1.11.27 TOKEN STATUS Minority 5.362.60 2.06.04 Minority Status Indirect Effect SE z p Whites -1.59.72 -2.20.03 Non-whites .32.27 1.19.23 The results for the separate tests of mediation indicated full support of the mediation model in white participants, but not in non-white participants. Moreover, the paths are not in hypothesized directions. Th e relationship between token status and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals, howeve r the opposite was found. Table 320 displays the results for the simple tests of mediation.

PAGE 450

415 Table 320. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Token Status, Med = Subtle Discrimination, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) -5.032.31-2.18.032.241.07 2.09 .04 b(MX) 4.861.613.01.00-.99.94 -1.04 .30 b(YM.X) -.25.10-2.57.01-.37.09 -4.38 .00 b(YX.M) -3.802.33-1.63.101.871.02 1.83 .07 Finally, perceived subtle race -based discrimination was i nvestigated as a mediator in the relationship between token status and psychological hea lth. Support for overall mediation was found as the paths between t oken status and percei ved discrimination ( B = 6.18, t = 3.17, p < .01) and between the mediator and psychological health ( B = -.17, t = 2.38, p < .05) were jointly significant. A signifi cant interaction was found between the predictor (token status) and the moderator ( B = -6.46, t = -3.05, p < .01). Thus, minority status is a moderator in the current model. The indirect effects we re not significant for white or non-white participants however the indirect effect for white participants was approaching significance (Indirect effect = -1.04, z = -1.84, p = .06). Results are shown in Table 321.

PAGE 451

416 Table 321. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Subtle Discrimination, DV = Psychol ogical Health, Mod = Race) Predictor B SE t p Subtle race discrimination Constant 10.40.35 29.71.00 TOKEN STATUS 6.181.95 3.17.00 Minority Status 2.72.58 4.64.00 TOKEN STATUS Minority -6.462.12 -3.05.00 Psychological Health Constant 52.14.88 59.17.00 Subtle race discrimination -.17.07 -2.38.02 TOKEN STATUS -3.782.74 -1.38.17 Minority Status -.32.83 -.38.70 TOKEN STATUS Minority 4.682.97 1.57.12 Minority Status Indirect Effect SE z p Whites -1.04.56 -1.84.06 Non-whites .05.15 .32.75 The mediation model was not fully s upported for either white or non-white participants. Moreover, the paths are not in hypothesized directions. The relationship between token status and perceived discrimination was predicted to be positive for minority individuals and negative for major ity individuals, however the opposite was found. Table 322 displays these results.

PAGE 452

417 Table 322. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Token Status, Med = Subtle Discrimination, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) -4.822.79-1.72.09.951.12 .85 .40 b(MX) 6.181.553.99.00-.291.00 -.29 .77 b(YM.X) -.15.12-1.20.23-.18.08 -2.14 .03 b(YX.M) -3.912.89-1.35.18.901.11 .81 .42 Overt Race-based Discrimination Token status in the workplace was investigated as a predictor in a model where overt race-b ased discrimination medi ates the relationship between token status and five dependent variables (job satisfaction, organizational commitment, intention to turnover, physical health, and psychological health). Minority status was investigated as a moderator in the relationship between the predictor and the mediator. First, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between token status and job satisfaction. Support for overall mediation was found as the paths between token st atus and perceived discrimination ( B = 13.52, t = 5.18, p < .01) and between the mediator and job satisfaction ( B = -.10, t = -4.87, p < .01) were jointly significant. A si gnificant interaction was found be tween the predictor (token status) and the moderator ( B = -14.75, t = -5.22, p < .01). Thus, minority status is a moderator in the current model. Additionally, th e indirect effect is significant for white

PAGE 453

418 participants (Indirect effect = -1.39, z = -3.51, p < .01), but not for non-white participants. Results are shown in Table 323. Table 323. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Overt Discrimination, DV = Job Satisfaction, Mod = Race) Predictor B SE t p Overt race discrimination Constant 11.34.46 24.88.00 TOKEN STATUS 13.522.61 5.18.00 Minority Status 3.46.75 4.60.00 TOKEN STATUS Minority -14.752.82 -5.22.00 Job Satisfaction Constant 12.09.31 39.03.00 Overt race discrimination -.10.02 -4.87.00 TOKEN STATUS -.381.16 -.33.74 Minority Status .29.33 .87.38 TOKEN STATUS Minority .461.25 .36.72 Minority Status Indirect Effect SE z p Whites -1.39.40 -3.51.00 Non-whites .13.12 1.09.28 The results for the separate tests of medi ation indicated support for the mediation model in white participants, but not in non-wh ite participants. Moreover, the paths are not in hypothesized directions. The relationshi p between token stat us and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals, however the opposite was found. Table 324 displays the results for the simple tests of mediation.

PAGE 454

419 Table 324. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Token Status, Med = Overt Discrimination, DV = Job Satisfaction) Whites Non-whites B SE t p B SE t p b(YX) -1.781.16-1.53.13.20.47 .42 .67 b(MX) 13.522.295.91.00-1.231.22 -1.01 .32 b(YM.X) -.14.03-4.26.00-.08.03 -2.77 .01 b(YX.M) .101.21.08.94.10.46 .23 .82 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between token status and organizational co mmitment. Support for overall mediation was found as the paths between t oken status and percei ved discrimination ( B = 13.60, t = 5.25, p < .01) and between the mediator and organizational commitment ( B = .26, t = -4.50, p < .01) were jointly significant. A significant interaction was found between the predictor (token status) and the moderator ( B = -14.39, t = -5.13, p < .01). Thus, minority status is a moderator in the cu rrent model. Additionally, the indirect effect was significant for white partic ipants (Indirect effect = -3.52, z = -3.38, p < .01) but not for non-white participants. Resu lts are shown in Table 325.

PAGE 455

420 Table 325. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Overt Discrimination, DV = Organizational Commitment, Mod = Race) Predictor B SE t p Overt race discrimination Constant 11.25.45 24.78.00 TOKEN STATUS 13.602.59 5.25.00 Minority Status 3.43.75 4.55.00 TOKEN STATUS Minority -14.392.80 -5.13.00 Organizational Constant 32.72.84 39.08.00 Overt race discrimination -.26.06 -4.50.00 TOKEN STATUS -3.553.12 -1.13.26 Minority Status 2.33.90 2.59.01 TOKEN STATUS Minority 2.663.37 .79.43 Minority Status Indirect Effect SE z p Whites -3.521.04 -3.38.00 Non-whites .20.29 .71.48 The results for the separate tests of mediation indicated full support of the mediation model for white participants only. Moreover, the paths are not in hypothesized directions. The relationship between token status and perceive d discrimination was predicted to be positive for minority individua ls and negative for ma jority individuals, however the opposite was found. Results are displayed in Table 326.

PAGE 456

421 Table 326. Results of Simple Mediation for Whites/N on-whites (PREDICTOR = Token Status, Med = Overt Discrimination, DV = Organizational Commitment) Whites Non-whites B SE t p B SE t p b(YX) -7.083.14-2.25.03-.691.25 -.55 .58 b(MX) 13.602.236.09.00-.791.23 -.64 .52 b(YM.X) -.28.09-3.03.00-.25.07 -3.36 .00 b(YX.M) -3.293.33-.98.32-.881.22 -.72 .47 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between token status and intention to turnov er. Support for overall mediation was found as the paths between t oken status and percei ved discrimination ( B = 13.41, t = 5.09, p < .01) and between the mediator and intention to turnover ( B = .06, t = 5.92, p < .01) were jointly significant. A signifi cant interaction was found between the predictor (token status) and the moderator ( B = -14.35, t = -5.04, p < .01). Thus, minority status is a moderator in the current mode l. Additionally, the indirect effects were significant for white particip ants (Indirect effect = .76, z = 3.83, p < .01) but not for nonwhite participants. Results are displayed in Table 327.

PAGE 457

422 Table 327. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Overt Discrimination, DV = Intention to Turnover, Mod = Race) Predictor B SE t p Overt race discrimination Constant 11.44.46 24.88.00 TOKEN STATUS 13.412.64 5.09.00 Minority Status 3.31.76 4.34.00 TOKEN STATUS Minority -14.352.85 -5.04.00 Turnover Constant 2.09.14 14.72.00 Overt race discrimination .06.01 5.92.00 TOKEN STATUS .21.53 .39.70 Minority Status -.26.15 -1.71.09 TOKEN STATUS Minority -.27.57 -.46.64 Minority Status Indirect Effect SE z p Whites .76.20 3.83.00 Non-whites -.05.06 -.85.40 The results for the separate tests of mediation indicated full support of the mediation model for white participants, but not for non-white participants. Moreover, the paths are not in hypothesized directions. Th e relationship between token status and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals, howeve r the opposite was found. Table 328 displays the results for the simple tests of mediation.

PAGE 458

423 Table 328. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Token Status, Med = Overt Discrimination, DV = Intention to Turnover) Whites Non-whites B SE t p B SE t p b(YX) .97.541.79.07-.11.22 -.52 .61 b(MX) 13.412.355.71.00-.941.21 -.77 .44 b(YM.X) .08.015.28.00.04.01 3.26 .00 b(YX.M) -.05.55-.09.93-.07.21 -.34 .73 Next, perceived overt race-ba sed discrimination was invest igated as a mediator in the relationship between token status and physical health. Support for overall mediation was found as the paths between token st atus and perceived discrimination ( B = 12.17, t = 4.38, p < .01) and between the mediator and physical health ( B = -.29, t = -6.61, p < .01) were jointly significant. A si gnificant interaction was found be tween the predictor (token status) and the moderator ( B = -13.92, t = -4.64, p < .01). Thus, minority status is a moderator in the current model. Additionally, th e indirect effect is significant for white participants (Indirect effect = -3.56, z = -3.62, p < .01) but not for non-white participants. Results are shown in Table 329.

PAGE 459

424 Table 329. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Overt Discrimination, DV = Physical Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 11.33.47 24.09.00 TOKEN STATUS 12.172.78 4.38.00 Minority Status 3.53.80 4.41.00 TOKEN STATUS Minority -13.923.00 -4.64.00 Physical Health Constant 69.58.64 108.66.00 Overt race discrimination -.29.04 -6.61.00 TOKEN STATUS -1.532.41 -.63.53 Minority Status -.72.70 -1.02.30 TOKEN STATUS Minority 3.262.62 1.25.21 Minority Status Indirect Effect SE z p Whites -3.56.98 -3.62.00 Non-whites .51.34 1.49.14 The mediation model was fully supporte d for white participants, but was not supported for non-white participants. More over, the paths are not in hypothesized directions. The relationship between token status and perceive d discrimination was predicted to be positive for minority individua ls and negative for ma jority individuals, however the opposite was found. Results are displayed in Table 330.

PAGE 460

425 Table 330. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Token Status, Med = Overt Discrimination, DV = Physical Health) Whites Non-whites B SE t p B SE t p b(YX) -5.102.33-2.19.032.241.09 2.06 .04 b(MX) 12.172.514.85.00-1.751.26 -1.39 .17 b(YM.X) -.27.06-4.38.00-.31.06 -4.85 .00 b(YX.M) -1.812.35-.77.441.701.03 1.65 .10 Finally, perceived overt race-based discrimi nation was investigated as a mediator in the relationship between token status and psychological hea lth. Support for overall mediation was found as the paths between t oken status and percei ved discrimination ( B = 13.35, t = 5.02, p < .01) and between the mediator and psychological health ( B = -.19, t = -3.60, p < .01) were jointly significant. A signi ficant interaction wa s found between the predictor (token status) and the moderator ( B = -14.45, t = -4.99, p < .01). Thus, minority status is a moderator in the current model. The indirect effect is significant for white participants (Indirect effect = -2.47, z = -2.88, p < .01) but not for non-white participants. Table 331 displays the results for the moderated mediation.

PAGE 461

426 Table 331. Results of Moderated Mediation (PRED ICTOR = Token Status, Med = Overt Discrimination, DV = Psychol ogical Health, Mod = Race) Predictor B SE t p Overt race discrimination Constant 11.50.47 24.31.00 TOKEN STATUS 13.352.66 5.02.00 Minority Status 3.43.80 4.28.00 TOKEN STATUS Minority -14.452.89 -4.99.00 Psychological Health Constant 52.67.76 69.17.00 Overt race discrimination -.19.05 -3.60.00 TOKEN STATUS -2.492.78 -.90.37 Minority Status -.45.83 -.55.59 TOKEN STATUS Minority 3.403.02 1.13.26 Minority Status Indirect Effect SE z p Whites -2.47.86 -2.88.00 Non-whites .20.23 .90.37 The results for the separate tests of medi ation exhibited a lack of support for the mediation model in either white or non-white participants. Moreover, the paths are not in hypothesized directions. The relationship between token status and perceived discrimination was predicted to be positive for minority individuals and negative for majority individuals, however the opposite was found. Results are shown in Table 332.

PAGE 462

427 Table 332. Results of Simple Mediation for Whites and Non-whites (PREDICTOR = Token Status, Med = Overt Discrimination, DV = Psychological Health) Whites Non-whites B SE t p B SE t p b(YX) -4.972.79-1.78.081.121.13 .99 .33 b(MX) 13.352.395.59.00-1.101.27 -.87 .39 b(YM.X) -.15.08-1.93.06-.21.07 -3.19 .00 b(YX.M) -2.942.97-.99.32.881.10 .80 .43

PAGE 463

428 Table 333. Summary of Significant Results for Pe rceived Equal Employment Opportunity Path Overall model Interaction Majority Indirect Effect Minority Indirect Effect Majority Simple Mediation Minority Simple Mediation Subtle sex Job Satisfaction X X X X X Subtle sex Org Commitment X X X Subtle sex Intent to Turnover X X X X X Subtle sex Physical health X X X X Subtle sex Psych health X X X X Overt sex Job Satisfaction X X X X X Overt sex Org Commitment X X X X X Overt sex Intent to Turnover X X X X X Overt sex Physical health X X X X X Overt sex Psych health X X X X Subtle race Job Satisfaction X X X X X Subtle race Org Commitment X X X Subtle race Intent to Turnover X X X X X Subtle race Physical health X X X X X Subtle race Psych health Overt race Job Satisfaction X X X X X Overt race Org Commitment X X X X X Overt race Intent to Turnover X X X X X Overt race Physical health X X X X X Overt race Psych health X X X X X indicates a significant finding at p < .05

PAGE 464

429 Table 334. Summary of Significant Resu lts for Minority Segmentation Path Overall model Interaction Majority Indirect Effect Minority Indirect Effect Majority Simple Mediation Minority Simple Mediation Subtle sex Job Satisfaction X X X X X Subtle sex Org Commitment X X X X X Subtle sex Intent to Turnover X X X X X Subtle sex Physical health X X X X Subtle sex Psych health X X X X Overt sex Job Satisfaction X X X X X Overt sex Org Commitment X X X X X Overt sex Intent to Turnover X X X X X Overt sex Physical health X X X X X Overt sex Psych health X X X X Subtle race Job Satisfaction X X X Subtle race Org Commitment X X X Subtle race Intent to Turnover X X X Subtle race Physical health X X X Subtle race Psych health Overt race Job Satisfaction X X X X Overt race Org Commitment X X X X Overt race Intent to Turnover X X X X X Overt race Physical health X X X X X Overt race Psych health X X X X X indicates a significant finding at p < .05

PAGE 465

430 Table 335. Summary of Significant Resu lts for Diversity Climate Path Overall model Interaction Majority Indirect Effect Minority Indirect Effect Majority Simple Mediation Minority Simple Mediation Subtle sex Job Satisfaction X X X Subtle sex Org Commitment Subtle sex Intent to Turnover X X X X X Subtle sex Physical health X X X X Subtle sex Psych health X X X Overt sex Job Satisfaction X X X X Overt sex Org Commitment X X X Overt sex Intent to Turnover X X X X X Overt sex Physical health X X X X X Overt sex Psych health X X X X Subtle race Job Satisfaction X X X X Subtle race Org Commitment X Subtle race Intent to Turnover X X X X X X Subtle race Physical health X X X X X X Subtle race Psych health X Overt race Job Satisfaction X X X X X Overt race Org Commitment X Overt race Intent to Turnover X X X X X Overt race Physical health X X X X X X Overt race Psych health X X X X X X indicates a significant finding at p < .05

PAGE 466

431 Table 336. Summary of Significant Results for Instrumental Social Support Path Overall model Interaction Majority Indirect Effect Minority Indirect Effect Majority Simple Mediation Minority Simple Mediation Subtle sex Job Satisfaction X X X Subtle sex Org Commitment X Subtle sex Intent to Turnover X X X X Subtle sex Physical health X X X Subtle sex Psych health X X X Overt sex Job Satisfaction X X Overt sex Org Commitment X Overt sex Intent to Turnover X X X X X Overt sex Physical health X X X X X Overt sex Psych health X X X X Subtle race Job Satisfaction X X X X Subtle race Org Commitment X X X X Subtle race Intent to Turnover X X X X X X Subtle race Physical health X X X Subtle race Psych health Overt race Job Satisfaction X X X Overt race Org Commitment X Overt race Intent to Turnover X X X X Overt race Physical health X X X X X Overt race Psych health X X X X indicates a significant finding at p < .05

PAGE 467

432 Table 337. Summary of Significant Results for Emotional Social Support Path Overall model Interaction Majority Indirect Effect Minority Indirect Effect Majority Simple Mediation Minority Simple Mediation Subtle sex Job Satisfaction X Subtle sex Org Commitment Subtle sex Intent to Turnover X X X X Subtle sex Physical health X X Subtle sex Psych health X X X Overt sex Job Satisfaction X Overt sex Org Commitment X X X X Overt sex Intent to Turnover X X X X X Overt sex Physical health X X X X X Overt sex Psych health X X X X Subtle race Job Satisfaction X X X Subtle race Org Commitment X X X Subtle race Intent to Turnover X X X X Subtle race Physical health X X Subtle race Psych health Overt race Job Satisfaction X X X Overt race Org Commitment X X X X Overt race Intent to Turnover X X X X Overt race Physical health X X X X X Overt race Psych health X X X X indicates a significant finding at p < .05

PAGE 468

433 Table 338. Summary of Significant Re sults for Token Status Path Overall model Interaction Majority Indirect Effect Minority Indirect Effect Majority Simple Mediation Minority Simple Mediation Subtle sex Job Satisfaction X X Subtle sex Org Commitment X X Subtle sex Intent to Turnover X X Subtle sex Physical health X X X Subtle sex Psych health X X X Overt sex Job Satisfaction X X X Overt sex Org Commitment X X X Overt sex Intent to Turnover X X X Overt sex Physical health X X X Overt sex Psych health X X X Subtle race Job Satisfaction X X X X Subtle race Org Commitment X X X X Subtle race Intent to Turnover X X X X Subtle race Physical health X X X X Subtle race Psych health X X Overt race Job Satisfaction X X X X Overt race Org Commitment X X X X Overt race Intent to Turnover X X X X Overt race Physical health X X X X Overt race Psych health X X X X indicates a significant finding at p < .05 Discussion Summary of Results for Equal Employment Opportunity Hypothesis 1 predicted that percei ved discrimination would mediate the relationship between equal employment oppor tunity and outcomes. However, minority status would not moderate the relationship between EEO and perceived discrimination.

PAGE 469

434 Both minority and majority group members w ould report a negative relationship between EEO and perceived discrimination, a pos itive relationship between EEO and job satisfaction, organizational commitment, physi cal health and psychol ogical health, and a negative relationship with turnover intention. Taken together, there is strong support for the mediating role of perceived discrimination in the relationship between perceived equal employ ment opportunity and workplace outcomes. The relationship is cons istent regardless of whether one is considering subtle or overt discrimination an d/or sex-based or race -based discrimination. All twenty models examined rev ealed support for this finding. This relationship does not appear to be different for Men and Women, or whites and nonwhites. In most of the twenty models examined, the indirect effects and/or path coefficients were significant for men and women or whites and non-whites. There were exceptions to this in seven of the twenty models. In six of these cases, all paths or indirects were significant except for one path or one indirect that approached significance. For example, in the model examining perceived subtle race-based discrimination as mediator in the rela tionship between EEO and organizational commitment, the path coefficient in the simple test of mediation fo r white participants approached significance whereas all other paths and the indirect were significant. This is likely a power issue that would be resolved with a larger sample. Minority status does not a ppear to moderate the re lationship between EEO and perceived discrimination, as none of the tw enty models found a si gnificant interaction between EEO and minority status in predicting perceived di scrimination. Thus, perceived

PAGE 470

435 strength of a company’s EEO policy is an impor tant consideration in lessening feelings of unfair treatment across categories of employees. Summary of Results for Minority Segmentation Hypothesis 2 predicted that percei ved discrimination would mediate the relationship between minority segmentati on and outcomes. Minority status would moderate the relationship between minority se gmentation and perceived discrimination. It was predicted that minority individuals w ould report a positive relationship between minority segmentation and perceived discri mination, a negative relationship between minority segmentation and job satisfaction, or ganizational commitment, physical health and psychological health, and a positive rela tionship with turnover intention. Majority individuals, on the other ha nd, would report a negative relationship between minority segmentation and perceived discrimination, a positive relationship between minority segmentation and job satisfaction, organiza tional commitment, physical health and psychological health, and a negative re lationship with turnover intention. Taken together, the results for minority segmentation indicate that there is an overall effect of minority segmentation on outcomes through perceived discrimination. The models for subtle sex discriminati on, overt sex discrimination and overt race discrimination all demonstrated support for mediation, suggesting that minority segmentation is related to increased perceive d discrimination, which in turn leads to poorer outcomes. The five models that inves tigated subtle race discrimination did not indicate full support for the mediation model. Ho wever, in all cases all of the paths were significant except one that approached si gnificance. Thus, minority segmentation does relate to perceived discrimination acro ss groups of employees, and differentially

PAGE 471

436 grouping the workforce by gender or ethnic ity can negatively impact outcomes. However, there were differences between white/nonwhite particip ants and men-women. The results for men and women seemed to indicate support for the overall mediation model, but there was no support for mi nority status as a moderator, the indirect effects were significant for both groups in ev ery case, and the path coefficients support mediation for both groups in nearly every case. Thus, the rela tionship of minority segmentation on outcomes through perceived di scrimination does not appear to differ based on gender. The results for white and non-white particip ants were less straightforward. In two of the ten models investigated, minority st atus was found to be a moderator in the relationship between minority segmentation an d perceived discrimination. In six of the remaining models, this interaction effect was marginally significant. Moreover, the indirect effects support a possible difference between white and non-white individuals. In eight of the ten models investigated, tests of non-white individuals exhibited a significant indirect effect, whereas tests of white indivi duals did not. However, the simple tests of mediation reveal a different scenario as bot h groups displayed support for the model in most cases. Based on these findings it is difficult to conclude that the effect of minority segmentation on outcomes through perceived di scrimination differs between white and non-white individuals. However, the pattern of results suggests there may be ethnic differences. It is possible that a larger sample of employees would better reveal the group differences. Finally, the directions of the path coeffici ents are different than expected. It was hypothesized that minority group members woul d exhibit a positive relationship between

PAGE 472

437 minority segmentation and perceived discrimina tion, as minorities are more likely to be treated unfairly as they are gr ouped together within an orga nization. Conversely, majority group members were expected to exhibit a negative relationship between minority segmentation and perceived discrimination, as they are likely to benefit from ethnic grouping. Next, it was predicted that minority group members would report a poorer job satisfaction, organizational commitment, physi cal health, and psychological health with increasing minority segmentation. Also, they would experience a positive relationship between minority segmentation and intention to turnover. The rationale was that minority individuals will realize they are being grouped into an area with less prestige and influence, and this will negatively impact work attitudes and health. The opposite directions were predicted for majority group members as they will realize that they are surrounded by others with power and influen ce, and this will positively affect work attitudes and health. However, path directions were identi cal across models and between men/women and whites/non-whites. In each case, greater minority segmentation was associated with more perceived discrimi nation. Thus, as a workplace became more grouped by minority status, individuals felt more unfair treatment, regardless of minority status. It seems that majority group member s may not receive more benefits as the workplace becomes segmented; people from all groups are at a disadvantage. Also, across models there was a positive relati onship between minority segmentation and intention to turnover, and a ne gative relationship with the ot her work attitude and health outcomes. Thus, as a workplace became more segmented, individuals reported less job satisfaction, organizational commitment, physi cal health, and psychol ogical health, and a greater intention to turnover. This is somewh at puzzling as one would expect a worker to

PAGE 473

438 experience more positive outcomes as they are surrounded by more similar others. However, it is possible that an organizati on which funnels minority group members into certain areas has other negative environm ental characteristics which make it an undesirable place to work. Employees may not experience unfair treatment based on race or sex (because they are surrounded by sim ilar others) but they may experience unfair treatment for other reasons. For example, management may unfairly treat all employees below a certain level or in certain types of positions. Or perhaps a company which archaically sorts employees based on physical characteristics also has archaic technology, ancient buildings, and outdated policies. The wording of the minority segmentati on scale may account for the difference in findings between men and women, and white s and non-whites. The scale refers to majority and minority group members without de fining what is meant by “minority”. It is possible that respondents associat ed race or ethnicity with minor ity status more often than gender when responding to these questions. Thus, when ethnic minorities were answering questions about minority segm entation, they were thinking about themselves whereas ethnic majority individuals were thinking about others. Whom men or women were thinking of depended on their ethnicity. Re gardless of the group comparisons, there seemed to be overall support for the model where perceived discrimination mediated the relationship between minority segmentation and outcomes. There were no differences between men and women, although both groups e xhibited significant indirect effects. There were differences between whites and non-whites, where the indirect effects were significant for non-white participants only. The i ndirect effects appear significant for both men and women because both of these groups are composed of roughly equal ethnic

PAGE 474

439 groups. The group differences in this model ar e ethnic differences. N on-white individuals report a relationship between minority segm entation and discrimination because they were likely referring to their ethnic group wh en responding to items. White individuals, on the other hand, were specifically envisioni ng their group as “majority group members” when responding to the items. Summary of Results for Diversity Climate Hypothesis 3 predicted that perceive d discrimination would mediate the relationship between diversity climate and out comes. Minority status would moderate the relationship between diversity climate and perc eived discrimination. It was predicted that minority individuals would re port a negative relationship be tween diversity climate and perceived discrimination, a positive relati onship between diversity climate and job satisfaction, organizational commitment, physi cal health, and psychol ogical health, and a negative relationship with intent to turnove r. Majority individuals, on the other hand, would report a positive relationship betw een diversity climate and perceived discrimination, and negative relationship betw een diversity climate and job satisfaction, organizational commitment, physical health and psychological he alth, and a positive relationship with tu rnover intention. Taken together, there was overall su pport for the mediation model where perceived discrimination mediated the rela tionship between diversity climate and the outcomes. Of the twenty models examined, sixteen displayed full support for the model and one model with all paths significant except a single pa th approaching significance. The three remaining models a ll involved race-based discrimi nation. However, there does appear to be a general relationship of di versity climate on work outcomes and health

PAGE 475

440 through perceived discrimination. Specificall y, a more positive di versity climate is associated with less perceived discrimination, wh ich in turn is relate d to better outcomes. There appears to be a difference between sex-based and raced-based discrimination in the model. In the ten m odels that included sex-based discrimination, there was no support for minority status as a moderator in the relationship between diversity climate and perceived discriminati on. Moreover, none of the indirect effects differed between men and women. However, so me of path coefficient results differed between men and women. For example, there was not full support of the model for men with regards to the physical and psychologica l health outcomes with subtle perceived discrimination as the moderator, and with psyc hological health as th e outcome with overt perceived discrimination as the moderator. In these cases, just one of the paths did not reach significance, while all others were sign ificant. These isolated cases may reveal gender differences with regards to a specifi c outcome, but as a whole, men and women do not appear to differ on the mediation model. On the other hand, all ten of the mo dels comparing white to non-white participants demonstrated support for minority status as a moderato r in the relationship between diversity climate and the perceived discrimination. This means the relationship between diversity climate and perceived r ace-based discrimination differs by ethnic group. However, this difference was not refl ected in the indirect effects or path coefficients of the simple tests mediati on, as they were jointly significant or nonsignificant in most cases. It is possible to have evidence of minority status as a moderator but no significant differences in the media tion models. The modera tor is only indicating the presence of an interacti on effect between the moderator variable and the predictor on

PAGE 476

441 the mediator, whereas the indi rect effects are also taki ng into account the dependent variables. The interaction effects in each mode l are negative, indicati ng that the slope of the relationship between diversity climate a nd perceived discrimination are steeper for minority group members compared to majo rity group members. Finally, the path coefficients did not match hypothesized dire ctions. It was hypothesized that path directions would differ between minority a nd majority group members. First, it was predicted that minority group members would experience a negative relationship between diversity climate and perceived discrimi nation. Because minority group members are likely beneficiaries of a positive diversity climate, their perceptions of unfair treatment would decrease as the diversity climate beco mes stronger. Additionally, as the diversity climate becomes stronger, outcomes would become more positive. Conversely, majority group members were predicted to experience a positive relationship between diversity climate and perceived discrimination. Since majority group members are likely to lose power and influence and these benefits ar e dispersed among minority group members, they would experience more pe rceptions of unfair treatment as the diversity climate strengthens. Moreover, their outcomes w ould become more negative with a stronger diversity climate. However, the results in dicated no differences in path directions between majority and minority group member s. Across all models, there was a negative relationship between diversity climate and perceived discrimination. Apparently all individuals perceived less unfair treatment as the diversity clim ate improved. It is possible that a strengthened di versity climate improves fairne ss for everyone, and the loss of benefits in either unnoticed or nonexist ent in majority group members. Additionally, results indicated a positive relationship betw een diversity climate and job satisfaction,

PAGE 477

442 organizational commitment, physical healt h, and psychological health, as well as a negative relationship with intention to tur nover. All workers perceived more positive outcomes as diversity climate improved. Agai n, a strong diversity climate may foster benefits for all employees regard less of gender or ethnicity. Summary of Results for In strumental Social Support Hypothesis 4 predicted that perceive d discrimination would mediate the relationship between instrumental social support and outcomes for minority individuals, but not for majority individuals. Minority st atus would not moderate the relationship between instrumental social support and perc eived discrimination. Mi nority individuals would report a negative relati onship between instrumental social support and perceived discrimination, a positive relationship betw een instrumental social support and job satisfaction, organizational commitment, physi cal health and psyc hological health, and negative relationship with intent to turnove r. Majority group members, on the other hand, would report no relationship between instrumental soci al support and perceived discrimination. However, they would report re lationships between instrumental social support and outcomes in the same dir ections and minority group members. Taken together, there was support for the overall model. In eighteen of the twenty models examined, perceived discrimina tion mediated the relationship between instrumental social support and the outcom es. In one other model, all paths were significant save one that appro ached significance. As a whole, there seems to be support that instrumental social suppor t plays a role in affecting pe rceived discrimination in the workplace, across gender or ethnic groups. Howe ver, there were differences between men and women and between white and non-white respondents.

PAGE 478

443 First, there was no support for the mode rating role of minority status in the relationship between instrumental social s upport and perceived sex-based discrimination. However, the results differ for subtle and ove rt discrimination. The indirect effects and simple mediation tests for subtle sex-based discrimination revealed differences between men and women. In many of these cases, the mediation model was supported for women, but not for men. But for overt sex-based disc rimination, the indirect effects and simple mediation follow-up tests were more sim ilar between men and women. This likely indicates a difference in type of perceived discrimination in the workplace. Since this pattern of findings was very similar for emoti onal social support, reasons for these results will be discussed in a later section. It was hypothesized that th ere would be no mediation for majority group members, as majority group members tend to attribute a lack of social support to other factors than discrimi nation. This was somewhat supported, but apparently the type of discrimination is important. For race-based discrimination, on the ot her hand, there was some evidence of the moderating role of minority stat us in the relationship betwee n instrumental social support and perceived discrimination. The interaction e ffect was significant in three of the ten models examined, and marginal in an additiona l four models. This lends some support for ethnic differences in this part icular relationship. The indirect effects and tests of simple mediation were also different between wh ite and non-white individuals in many cases, lending further support for group differences. In most instances where there were group differences, the model was significant for nonwhite individuals a nd non-significant for white individuals. This supports the hypothesis for instrumental soci al support. Majority

PAGE 479

444 group members (e.g. white individuals) will attrib ute an absence of social support at work to other factors than unfair treatment. Across the instrumental social suppo rt models, the paths conformed to hypothesized directions. It was predicted that there would be a negative relationship between instrumental social support and perc eived discrimination, a negative relationship between perceived discrimination and outco mes, and a positive relationship between instrumental social support and outcomes. Paths were supported for both majority and minority group members. Although a lack of me diation was expected for majority group members, one would still expect to s ee the individual path s in the hypothesized directions. Summary of Results for Emotional Social Support Hypothesis 5 predicted that perceive d discrimination would mediate the relationship between emotional social support and outcomes fo r minority individuals, but not for majority individuals. Minority status would not mode rate the relationship between emotional social support and perceived discri mination. Minority indi viduals would report a negative relationship between emotional so cial support and perceived discrimination, a positive relationship between emotional social support and job satisfaction, organizational commitment, physical health and psychological he alth, and negative relationship with intent to turnover. Majo rity group members, on the other hand, would report no relationship between emotional so cial support and perceived discrimination. However, they would report relationships between emotional social support and outcomes in the same directions and minority group members.

PAGE 480

445 Taken together, there appears to be ov erall support for the mediation model, where perceived discriminati on mediates the relationship between emotional social support and outcomes. This model was fully s upported in 13 of the 20 models examined, and marginally supported in an additional five models. Thus, strong emotional support can be an important organizational factor in reducing perceived discrimination and improving work attitudes and outcomes. Results differed for the emotional soci al support predictor depending on what type of discrimination one is considering. For sex-based discrimination, there did not appear to differences between men and wo men. Minority status wa s not found to be a moderator in any of the ten models examine d. The majority of the indirect effects and tests of simple mediation did not differ be tween men and women. In cases where there were differences, one group was marginally si gnificant while the ot her was significant. Taken together, there does not appear to be strong support for a gender differences. For race-based discrimination, there was some support for the moderating role of minority status in the relationship between emotional social support and perceived discrimination. There was support for the mode rating role of minority status in the relationship between emotiona l social support and perceive d discrimination for work attitude outcomes (i.e. job sa tisfaction, organizational commitment, intention to turnover) but not health outcomes. However, there were more marginally significant than significant interaction effects as a whole. Furt her, only half of the indirect effects and tests of simple mediation revealed group differe nces in mediating effects. Taken together, it appears there some effect of ethnicity, but it is difficu lt to draw strong conclusions based on the mixed results.

PAGE 481

446 Across the emotional social support models the paths conformed to hypothesized directions. It was predicted th at there would be a negative relationship between emotional social support and perceived discriminati on, a negative relationship between perceived discrimination and outcomes, and a positive relationship between emotional social support and outcomes. Paths were supporte d for both majority and minority group members. Although a lack of mediation was expected for majority group members, one would still expect to see the individu al paths in the hypothesized directions. Summary of Results for Token Status In hypothesis six, it was predicted that perceived discrimination would mediate the relationship between token status and outcomes. Minority status would moderate the relationship between token status and perc eived discrimination. Minority individuals would report a positive relationship between tokenism and perceived discrimination, a negative relationship between tokenism and job satisfaction, organizational commitment, physical health and psychologica l health, in addition to a negative relationship with intention to turnover. Majority group indi viduals, on the othe r hand, would report a negative relationship between tokenism and perceived discrimination, a positive relationship between tokenism and job satis faction, organizational commitment, physical health and psychological health, in addition to a positive relationship with intent to turnover. The findings for token status differ considerably based on whether one is considering sex-based discrimination or race-based discrimination. For sex-based discrimination there was no evidence for th e overall mediation model. However, there was evidence for the moderating role of minor ity status in the relationship between token

PAGE 482

447 status and perceived discrimination. Moreover, in each of the comparisons of indirect effects and tests of simple mediation, there were significant differences between men and women. Specifically, the models were si gnificant for women, but not for men. The findings were very different within race-based discrimination models. Here, the overall model was supported in every ca se. That is, perceived discrimination mediated the role between token status and outcomes. Additionally, th e interaction effect between token status and minority status was si gnificant in each model. There were also significant differences between white and non-wh ite respondents in the majority of the indirect effects and tests of simple mediation. In each case, the model was significant for white individuals rather th an non-white individuals. The differences between sex-based and r ace-based discrimination illustrate how token status differentially a ffects different groups. Regardless of the model examined, token status exhibited a positiv e relationship with perceive d discrimination, meaning the predicted levels of perceived discriminati on were higher for tokens than non-tokens. Also, minority status moderated the relations hip between token stat us and both types of discrimination, indicating the “t ype” of token matters. First, the indirect effects were significant for women, but not men. Thus, toke n status plays an important role in reporting discrimination, and in turn, affecting outcomes for women. The literature review highlighted studies which found high le vels of perceived discrimination and poor workplace outcomes among female tokens. On th e other hand, male tokens tend to stand out in a positive way, often receiving benefits for their token status. Thus, token or not, factors other than token status are likely to impact percei ved discrimination and outcomes for men. However, female token status is likely to play a large role in the specified

PAGE 483

448 mediation model. Interestingly, the situation was reversed when race was considered: the mediation models were significant for major ity group members (i.e. white individuals) but not minority group members. In this case, the situation for nonwhite individuals may be similar to men. Token or not, men are li kely to be treated the same (i.e. well). Similarly, non-white individuals may be treate d the same (i.e. poorly ), token or not. Nonwhite individuals tend to perceive higher leve ls of unfair treatment in general compared to white individuals. Perhaps token status does little to change this. A minority token may perceive unfair treatment from their cowork ers, in addition to other forces in the workplace. A minority non-token ca n still perceive unfair treat ment from supervisors and the organization. However, tokenism does play an important role for white individuals. White non-tokens are surrounded by like others and are unlikely to perceived unfair treatment. White tokens, on the other hand, are surrounded by nonwhite individuals. These individuals may view themselves in an inferior position a nd, in turn, perceive unfair treatment from supervisors or the or ganization. They may also perceive unfair treatment based on their race from their different-race coworkers. Subtle vs. Overt Perceived Discrimination Two types of discrimination were investig ated in the current study: subtle and overt. Because few studies have focused on pe rceived discrimination at work, it was of interest to investigate possible differences between forms of unf air treatment. For the EEO, minority segmentation and diversity cl imate models, there appeared to be no pattern of differences between subtle and overt perceived discrimination. However, as highlighted in previous discussions of resu lts, there were some differences for the two types of social support.

PAGE 484

449 First, both subtle and overt sex-based discrimination were found to mediate the relationship between instrumental social s upport and all outcomes. Minority status did not moderate the relationship between pred ictor and perceived discrimination in each case. However, there were differences in i ndirect effects. For the subtle sex-based models, the indirect effect was significant for women but not men. For the overt sexbased models, the effects were either si gnificant or not for both groups. There were similar findings for subtle and overt race-base d discrimination. Here, the indirect effects were significant for non-white pa rticipants and non-significant for white participants for subtle race discrimination, but either jointly si gnificant or non-significant for overt racebased discrimination. This pattern of findings was nearly identical for emotional social support. Because this pattern of results was no t mirrored in models involving the other four organizational antecede nts, these findings likely illuminate how social support operates in the workplace more so than differences between subtle and overt discrimination. Perhaps reported levels of so cial support were more strongly linked to overt discrimination for both groups, because lack of support is a form of overt discrimination. For example, one of the items on the overt race-based discrimination scale was “At work, I feel that others excl ude me from their activ ities because of my race/ethnicity.” Being included in work activitie s is a form of instrumental social support. Indirect effects for overt di scrimination were similar for minority and majority group members in social support models because of the strong conceptual overlap between the two constructs.

PAGE 485

450 Subtle discrimination, on the hand, is more like interpersonal mistreatment, and is likely less strongly related to social s upport than overt discrimination. Perhaps the reported relationship between social support and subtle di scrimination is stronger for minority group members because of attributions When minority individuals lack social support, they are more likely to view minority status as a cause, whereas majority group members are more likely to perceive other reasons. Thus, majority group members are failing to report a relationship between social support and subtle discrimination because subtle discrimination (i.e. interpersonal mist reatment) likely has other causes in their minds. Sex-based vs. Race-based Perceived Discrimination An examination of models pertaining to each of the six organizational antecedents does not reveal outstanding differences betw een race-based and sex-based discrimination. Patterns of differences seemed to lie within the models for an individual antecedent. Specifically, the models for minority segmen tation revealed differences by type of minority. The sex-based discrimination mode ls all supported the overall model where discrimination mediated the relationship between minority segmentation and outcomes. However, minority status was not found to be a moderator in any of the outcomes and the indirect effects were significant for both me n and women in every case. The race-based discrimination models, on the hand, demonstr ated that minority status moderated the relationship between minority segmentation an d discrimination either marginally or significantly in eight of ten m odels. Additionally, the indirect effects were significant for non-white individuals and non-significant for wh ite individuals in most cases. Possible explanations for these patterns of findings we re discussed in a previous section covering

PAGE 486

451 the results for minority segmentation. Taken to gether, the type of discrimination (racebased or sex-based) does not seem to be as important as the antecedents and other elements entered into the mediation models. Conclusions A total of 120 moderated mediation mode ls were investigat ed in the current paper. The majority of the models demonstr ated that perceived workplace discrimination mediates the relationship between workplac e conditions and outcomes. Taken together, poor workplace conditions related to greater perceived discrimination which related to negative consequences for individuals from a ll groups. While not consistent with a priori hypotheses, these findings are novel and suggest that a lack of commitment to workplace diversity adversely affects all employees, minority or not. However, patterns of moderation and indirect effect s reveal that relationships may be stronger for minority group members, suggesting that these wor kplace factors adversely affect minority individuals to a gr eater degree. The bulk of previous research has focu sed on individual differences which relate to the perception of discrimination. The current study is novel in that it demonstrates the importance of the environment in precipit ating feelings of unf air treatment. The perception of discrimination is the logical precu rsor to discrimination claims and as such organizations may want to place importan ce on creating a fair workplace for all individuals. Emphasizing a commitment to EEO policy, reducing minority segmentation, strengthening diversity climate, creating opportunities for instrumental and emotional social support, and being aware of employees with token status may help enhance the fairness of an organization and improve workpl ace attitudes and health for all employees.

PAGE 487

452 Limitations The main limitation of the full demonstrati on is likely the untes ted nature of many of the measures. Both measures of per ceived discrimination, the perceived Equal Employment Opportunity (EEO) measure, th e minority segmentation measure, and the token status measure are relativ ely new and do not have extens ive reliability and validity information. Although a pilot test was conducted to evaluate the quality of all measures and many analyses resulted in many signifi cant findings, it would be worthwhile to gather more data using the newer scales. The measure of minority segmentation has some specific limitations. Each item in this scale refers to “minorities” or “maj ority group members”. However, unlike other scales used in the study, it does not define wh at a minority or majority member is. Thus, we do not know specifically to whom responde nts were referring when they answered questions about these groups. It could be problematic if some respondents defined a “minority” as a woman, while other intended a “minority” to be a non-white individual, and still others referred to any group who happened to comprise a small percentage of their workplace. A layer of complexity is added since the current study compared a priori-defined minority to majority gr oup members based on their perceptions of undefined minority groups. As previously addr essed, this may account for differences in results between men/wome n and whites/non-whites. Another consideration is the number of tests conducted. One-hundred and twenty tests of moderated mediation and two hundred and forty tests of simple mediation were conducted. Descriptive statistics such as co rrelations among measures and significance testing between groups on demographic char acteristics were also conducted. This may

PAGE 488

453 raise concerns about the Type I error rate. However, the number of significant findings far exceeds 5%. The consistency in findings betw een similar tests and within each type of organizational antecedent lessens fears about spurious results. Finally, there were many instances of ma rginal significance. In several of these cases, there was evidence from related an alyses that more power might result in significant findings. As such, the study coul d have benefited from a larger sample. Limitations of the coworker study shoul d also be mentioned. Most importantly, the sample size was very small. Less than 10% of the participants in the full demonstration had a matched coworker. It becomes difficult to conduct analyses, let alone draw many conclusions from the results. Moreover, participants were not always similar to coworkers in terms of gender, et hnicity, or position with in the organization. Participants were asked to send the coworker survey to a similar other in terms of demographic and employment characteristics. This was not the case in many instances. Coworkers were asked to report on the envir onment, and when these individuals held a different position or perhaps were physica lly located in a different area from the coworker, perceptions may differ. Further, perceptions are likely to differ even more when the coworker is not similar in gender or ethnicity. Taken together, the results of the coworker survey should be interpreted with caution. Future Directions The study investigated group differences two ways: comparing men to women and whites to non-whites. Every person appear s in both analyses but some individuals shift minority status between the two. A natura l next step is to compare true majority group members (i.e. white men) to single-mi nority group members (i.e. white women and

PAGE 489

454 non-white men) and double-minority group memb ers (i.e. non-white women). It would also be interesting to break the non-white minority group into specific ethnic groups, such as black and Hispanic individuals. The per ceptions of black indi viduals may very well differ from other ethnic minorities, as some gr oups can “hide” their ethnicity and are less likely to receive differential treatment ba sed on group membership. For example, many Hispanic individuals have light skin and an absence of an accent, and therefore may not be generally perceived as an ethnic minority. Finally, the nature of perceived discrimina tion at work needs more attention in the literature. Perceptions of differential treatment can cost companies millions when they lead to a lawsuit. Yet, we understand little about what leads to pe rceived discrimination, as well as the composition of perceived discri mination itself. More st udies are needed to investigate the difference between subtle and overt discrimination. Additionally, the source of discrimination at wo rk is important. The measures of perceived discrimination in the current study addressed discriminati on from interpersonal sources as well as institutional sources. Interper sonal sources of discriminati on could include coworkers, supervisors, or other people at work. Institutional sources of discrimination come from policies and practices enacted by the organization as a w hole. Depending on the source, feelings of differential treatment may vary by group. This study was unable to tease apart differences among source of discrimination, but fu ture studies should pay attention to this issue.

PAGE 490

About the Author Nicole Ellis Jagusztyn was born and ra ised in Ft. Lauderdale, FL. After graduating from Nova High School, she attended th e University of Flor ida in Gainesville, FL. In 2004, she graduated summa cum laude w ith a Bachelors of Science in Psychology and a minor in Sociology. In Fall of 2004, sh e began her graduate program in Industrial Organizational Psychology at the University of South Florida. During her tenure at USF, she became involved in the emerging field of Occupational Health Psychology, and was one of the first students to concentrate in th e area. Her extensive coursework in the area allowed her to earn a graduate certificate in Safety Mana gement. At the end of her doctoral program, she began her career in ap plied research as a Research Analyst at Hillsborough Community College. In her spare time, Nicole enjoys cook ing, the outdoors, spending time with her dog and cat, and traveling as much as possible.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004642
035
(OCoLC)
040
FHM
c FHM
049
FHMM
090
XX9999 (Online)
1 100
Jagusztyn, Nicole.
0 245
Perceived workplace discrimination as a mediator of the relationship between work environment and employee outcomes :
b does minority status matter?
h [electronic resource] /
by Nicole Jagusztyn.
260
[Tampa, Fla] :
University of South Florida,
2010.
500
Title from PDF of title page.
Document formatted into pages; contains X pages.
502
Dissertation (PHD)--University of South Florida, 2010.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
3 520
ABSTRACT: The purpose of the current study was to explore the role of six organizational factors (Equal Employment Opportunity, minority segmentation, diversity climate, instrumental social support, emotional social support, and token status) in the perception of discrimination in the workplace by minorities and majority-group members. Five outcomes (job satisfaction, organizational commitment, intention to turnover, physical health, and psychological health) were investigated in response to perceived discrimination. Moderated mediation was used to test hypothesis where perceived discrimination mediated the relationship between organizational antecedents and outcomes; minority status served as the moderators. Support for the mediating role of perceived discrimination was found in the relationship between each organizational antecedent and outcome. In each case, poorer environmental conditions related to increased perceived discrimination which in turn related to more negative workplace attitudes and health outcomes. Implications for workplace design are discussed.
590
Advisor: Kristen Salomon, Ph.D.
653
Ethnic differences
Minority
Health
Stress
Job attitudes
Environment
690
Dissertations, Academic
z USF
x Psychology
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.4642