USF Libraries
USF Digital Collections

An empirical examination of job stress and management of emotionally-based behavior

MISSING IMAGE

Material Information

Title:
An empirical examination of job stress and management of emotionally-based behavior frontline social service personnel perspective
Physical Description:
Book
Language:
English
Creator:
Sams, Doreen
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Emotional intelligence
Emotional labor
Perceived customer demands
Job stress
Dissertations, Academic -- Business Administration -- Doctoral -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Frontline service personnel (FSP) play an invaluable role in the marketing mix by directly influencing the customers perception of both the service organization as well as the service quality during the face-to-face delivery service encounter (Ashforth and Humphrey 1993). The purpose of this dissertation is to examine how sources of job stress for FSP such as perceived customer demands, role ambiguity, role conflict, role overload, and emotional labor and various mediators such as job autonomy, emotional intelligence, and emotion-focused coping directly and indirectly influence job stress and outcome variables such as attitudes, behaviors, physical consequences, emotional exhaustion, job performance, and intentions.Research from frontline social service personnels (FSSP) perspective was collected in three phases: 1) eight personal in-depth interviews to determine the fit of the constructs in the model in a social service environment, 2) a cognitive response survey gathered from 86 FSSP to identify appropriate line items, and 3) a survey questionnaire gathered from 533 members of the National Association of Social Workers-Florida. Psychometrically sound scales developed and purified in the study demonstrated reliability and validity. These scales were then used to examine the structural model. Structural equation modeling, correlations, and regression analyses were used to examine relationships in the model. Results of the study indicated that self-management of emotionally-based behavior was significant in the creation and reduction of job stress.Findings suggest that the influence of emotional-based behavior plays a significant role in job performance at the social service encounter and indirectly influences intention to switch and intention to leave.
Thesis:
Thesis (Ph.D.)--University of South Florida, 2005.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Doreen Sams.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 287 pages.
General Note:
Includes vita.

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:
aleph - 001709554
oclc - 69180414
usfldc doi - E14-SFE0001347
usfldc handle - e14.1347
System ID:
SFS0025667:00001


This item is only available as the following downloads:


Full Text
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 Ka
controlfield tag 001 001709554
003 fts
005 20060614112238.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 060523s2005 flua sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001347
035
(OCoLC)69180414
SFE0001347
040
FHM
c FHM
049
FHMM
090
HF71 (Online)
1 100
Sams, Doreen.
3 245
An empirical examination of job stress and management of emotionally-based behavior
h [electronic resource] :
b frontline social service personnel perspective /
by Doreen Sams.
260
[Tampa, Fla.] :
University of South Florida,
2005.
502
Thesis (Ph.D.)--University of South Florida, 2005.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 287 pages.
Includes vita.
520
ABSTRACT: Frontline service personnel (FSP) play an invaluable role in the marketing mix by directly influencing the customers perception of both the service organization as well as the service quality during the face-to-face delivery service encounter (Ashforth and Humphrey 1993). The purpose of this dissertation is to examine how sources of job stress for FSP such as perceived customer demands, role ambiguity, role conflict, role overload, and emotional labor and various mediators such as job autonomy, emotional intelligence, and emotion-focused coping directly and indirectly influence job stress and outcome variables such as attitudes, behaviors, physical consequences, emotional exhaustion, job performance, and intentions.Research from frontline social service personnels (FSSP) perspective was collected in three phases: 1) eight personal in-depth interviews to determine the fit of the constructs in the model in a social service environment, 2) a cognitive response survey gathered from 86 FSSP to identify appropriate line items, and 3) a survey questionnaire gathered from 533 members of the National Association of Social Workers-Florida. Psychometrically sound scales developed and purified in the study demonstrated reliability and validity. These scales were then used to examine the structural model. Structural equation modeling, correlations, and regression analyses were used to examine relationships in the model. Results of the study indicated that self-management of emotionally-based behavior was significant in the creation and reduction of job stress.Findings suggest that the influence of emotional-based behavior plays a significant role in job performance at the social service encounter and indirectly influences intention to switch and intention to leave.
590
Adviser: David Ortinau, Ph.D.
Co-adviser:
653
Emotional intelligence.
Emotional labor.
Perceived customer demands.
Job stress.
0 690
Dissertations, Academic
z USF
x Business Administration
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1347



PAGE 1

An Empirical Examination of Job Stress and Management of Emotionally-Based Behavior: Frontline Social Service Personnel Perspective by Doreen Sams A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Marketing College of Business Administration University of South Florida Major Professor: David Ortinau, Ph.D. Karin Braunsberger, Ph.D. Miriam Stamps, Ph.D. James Stock, Ph.D. Dana Zeidler, Ph.D. Date of Approval: September 16, 2005 Keywords: emotional intelligence, emotiona l labor, perceived customer demands, job stress Copyright 2005, Doreen E. Sams

PAGE 2

Dedication This dissertation and my Ph.D. are dedicated to the one I love my devoted and loving husband and friend, Phil. Without his love and support the journey would not have been possible. I am truly blessed that God gave him to me to be my cheerleader, my dearest friend, my confidant, my light in the darkness, my love, and my True North. There are no words to express my gratitude and thanks to him for his willingness to make the myriad of sacrifices of time and money needed to pursue my Ph.D. Thank you, Phil, for always being there for me through all of lif es journeys, not just the Ph.D. process. Your support has made my burden lighter and has helped me to be better prepared for the next journey as an Assistant Marketi ng Professor. Thank you for everything! I also want to thank some very specia l little people who have made life worth living and provided welcomed distraction from the dissertation process. Thanks, Justin, Lorelei, Olivia, DJ, and Little Phil. Each one of your successes in life has brightened the path to my dissertation. A sp ecial thanks to my daughter, Shayne, for taking care of the house and running errands for me and especially for the times she helped to care for my aging mother so I could focus on the journey to my Ph.D. Thanks also to my sister Linda, and my nephew, Jim, as well as my ma ny friends who have encouraged me as I pursued the Ph.D. degree.

PAGE 3

Acknowledgements I believe a quote from Hillary Clintons book, Living History it takes a village best describes the Ph.D. process. Therefore, there are several people at the University of South Florida who contributed significantly to my successful journey through the Ph.D. program that I would like to thank. My deep est gratitude goes to each one of them: my Chair, Dr. David Ortinau, who worked closel y with me throughout th e entire process. For the countless hours reviewing and editing my work, for the sharing of his expertise as a researcher, and for keeping my focus on painting the trunk of the tree whenever I tried to paint the branches before the picture of the trunk was fully painted, thank you. I want to also thank my Committee Me mbers; Dr. Karin Braunsberger, Dr. Miriam Stamps, Dr. James Stock, and Dr. Dana Zeidler, for his or her valuable role in this dissertation process. In addition, I would like to thank my colleagues; Cynthia Cano and Fernando Jaramillo, for their support and friendship throughout the doctoral process. Thanks go to the students to whom I was priv ileged to teach during my time at University of South Florida. They have enriched my life as well as my career. A special thanks goes to two of my former st udents and research assistants Beth Livesay and Valerie Kimball. Without their w illingness to make contacts wi th officials of potential participant organizations, as well as perform other research duties, the task would have been overwhelming. A special thanks to Wendy Jennings for sharing her knowledge of deadlines and dissertation policie s and procedures and for editing the final paper. Thanks everyone!

PAGE 4

i TABLE OF CONTENTS LIST OF TABLES............................................................................................................vii LIST OF FIGURES.........................................................................................................viii ABSTRACT .....................................................................................................................i x CHAPTER 1 Introduction........................................................................................... 1 Job Stress and Service Deliver y Encounter Performance.................................. 3 Emotionally-Based Displayed Behavior 5 Frontline Social Service Personnels Perspective.................................. 6 Previous Models..................................................................................... 9 Shortcomings of Current Approaches to Job Stress 10 Research Objectives........................................................................................... 13 Self-Management of Emotionally -Based Displayed Behavior.............. 14 Emotional Intelligence............................................................... 14 Emotional Labor........................................................................ 14 Emotion-Focused Coping.......................................................... 15 Proposed Theoretical Framework....................................................................... 16 Definition of Constructs......................................................................... 16 Highlights of Proposed Model............................................................... 19 Research Questions................................................................................ 22 Anticipated Contributions of the Investigation.................................................. 23 Theoretical Contributions...................................................................... 23 Measurement Contributions................................................................... 25 Contributions for Practitioners............................................................... 26 Chapter Summary............................................................................................... 27 CHAPTER 2 Literature Review and Hypotheses...................................................... 29 Affective Event Theory........................................................................ 29 Role Theory.......................................................................................... 30 Sources of Job Stress............................................................................ 32 Traditional Sources of Job Stress............................................... 33 Role Ambiguity.............................................................. 33 Role Conflict.................................................................. 34 Role Overload................................................................ 36 Other Sources of Job Stress and Mediating Relationships...................... 38 Perceived Customer Demands................................................... 38 Mediating Role of Job Autonomy.............................................. 38

PAGE 5

ii Emotional Labor........................................................................ 39 Mediating Role of Em otional Intelligence................................. 41 Mediating Role of Emotion-Focused Coping............................ 42 Job Autonomy and Emotion-Focused Coping........................... 43 Emotional Intelligence and Emotion-Focused Coping.............. 43 The Impact of Emotion-Focused Coping............................................... 44 Impact on Emotional Exhaustion............................................... 44 Impact on Attitude Toward the Job........................................... 46 Impact on Overall Organizational Commitment and Overall Job Satisfaction..................................................................... 47 Impact on Physical Consequences............................................. 48 Interrelationships of Outcome Variables............................................... 49 Impact of Job-Related Outcome Variables on Job Performance........... 53 Impact of Attitude Toward the Job............................................ 54 Impact of Overall Job Satisfaction............................................. 54 Impact of Overall Organizational Commitment........................ 55 Impact of Emotional Exhaustion............................................... 56 Impact of Physical Consequences.............................................. 57 Predicting Behavioral Intentions................................................ 57 Chapter Summary ........................................................................................ 61 CHAPTER 3Methodology ........................................................................................ 63 Research Setting..................................................................................... 63 Procedures for Securing Participation..................................................... 64 Sampling Requirements for Pilot Study.................................... 64 Sampling Requirements for Final Study.................................... 65 Non-Response Bias.................................................................... 65 Construct Development......................................................................... 66 Scale Measurement Development.......................................................... 67 New Scale Development........................................................................ 69 Attitude Toward the Job............................................................. 69 Emotional Intelligence............................................................... 70 Emotional Labor........................................................................ 73 Emotion-Focused Coping.......................................................... 75 Intention to Switch Positions Within the Organization............. 77 Intention to Leave...................................................................... 78 Overall Job Satisfaction............................................................. 79 Perceived Customer Demands................................................... 81 Physical Consequences.............................................................. 82 Role Overload............................................................................ 83 Redesigned Scales................................................................................... 85 Role Ambiguity.......................................................................... 85 Role Conflict.............................................................................. 87 Job Autonomy............................................................................ 89 Social Service Encounter (Job) Stress....................................... 90

PAGE 6

iii Emotional Exhaustion................................................................ 91 Emotional E xhaustion Scale...................................................... 92 Job Performance......................................................................... 93 Overall Organizational Commitment......................................... 96 Data Analysis ........................................................................................ 97 Final Study................................................................................. 97 Mediation Testing...................................................................... 101 Chapter Summary ........................................................................................ 102 CHAPTER 4 Analysis and Results............................................................................. 103 Construct Development......................................................................... 104 In-Depth Interviews................................................................... 104 Pilot Study Cognitive Response Survey ................................ 105 Overall Response Rate............................................................... 106 Descriptive Profile of the Sampled Respondents....................... 107 Pilot Study Results............................................................................... 108 Construct Indicators................................................................... 108 Indicator Purification................................................................. 109 Data Analysis............................................................................. 113 Overall Response Rate............................................................... 114 Report of Non-Respondents................................................................... 115 Descriptive Profile of the Sampled Respondents................................... 116 Scale Measurement Purification............................................................ 117 Exploratory Factor Analyses Results...................................... 117 Measurement Model........................................................................................... 122 Model Fit Indices................................................................................... 122 Model Analysis...................................................................................... 125 Reliability................................................................................... 135 Convergent Validity................................................................... 136 Discriminant Validity................................................................. 144 Construct Validation.................................................................. 146 Hypotheses Testing............................................................................................ 146 Sources of Job Stress................................................................. 149 Mediating Relationships............................................................ 150 Job Stress Mediating Relationships........................................... 151 Proposed Mediating Influence of Job Autonomy for Perceived Customer Demands and Job Stress................................ 151 Proposed Mediating Influence of Emotional Intelligence for Emotional Labor and Job Stress.................................... 152 Emotion-Focused Coping.......................................................... 153 Test of Emotion-Focused Coping Mediation............................. 155 Emotion-Focused Coping Mediation from Job Stress to Emotional Exhaustion Frequency.................................. 155 Emotion-Focused Coping Mediation from Job Stress to Attitude Toward the Job................................................. 156

PAGE 7

iv Emotion-Focused Coping Mediation from Job Stress to Overall Job Satisfaction................................................. 158 Emotion-Focused Coping Mediation from Job Stress to Overall Organizational Commitment............................. 159

PAGE 8

v Emotion-Focused Coping Mediation from Job Stress to Physical Consequence.................................................... 160 Interrelationships of Outcome Variables................................... 163 Impact of Job-Related Variables on Job Performance............... 167 Predicting Behavioral Intentions................................................ 169 Power Analysis.......................................................................... 171 Chapter Summary............................................................................................... 171 CHAPTER 5 Discussion, Conclu sions, and Implications.......................................... 173 Research Purpose and Design................................................................ 173 Summary of Research Objectives.............................................. 173 Summary of the Research Study................................................ 175 Development and Assessment of New Scale Measurements............................. 176 Perceived Customer Demands............................................................... 177 Emotional Labor.................................................................................... 177 Emotional Intelligence........................................................................... 178 Emotion-Focused Coping...................................................................... 178 Attitude Toward the Job......................................................................... 179 Intention to Switch Positions Within the Organization......................... 179 Intention to Leave the Organization...................................................... 180 Overall Job Satisfaction............................................................. 180 Physical Consequences.......................................................................... 180 Role Overload........................................................................................ 181 Discussion of Empirical Fi ndings and Conclusions........................................... 181 Role Ambiguity, Role Conflict, and Role Overload as Antecedents to Job Stress........................................................... 181 Perceived Customer Demands as an Antecedent to Job Stress.............. 183 Emotional Labor as an An tecedent to Job Stress................................... 184 Mediating Effects of Job Aut onomy and Emotional Intelligence on Job Stress.............................................................................. 185 Constructs Direct Influen ce on Emotion-Focused Coping.................... 186 Emotion-Focused Copings Medi ating Role on Selected Outcome Constructs.................................................................................. 187 Impact of Emotion-Focused Coping Frequency........................ 188 Impact of Emotion-Focused Coping Effectiveness................... 188 Interrelationships Among th e Job Related Outcome Constructs 189 Job-Related Outcome Construc ts (Antecedents) Impact on Job Performance............................................................................... 189 Attitude Toward the Job............................................................. 190 Overall Job Satisfaction............................................................. 190 Overall Organizational Commitment......................................... 190 Emotional Exhaustion................................................................ 191 Physical Consequences.............................................................. 191 Job Performance as a Predicto r of Behavioral Intentions...................... 192 Implications for Researchers and Practitioners.................................................. 193

PAGE 9

vi Limitations of the Study..................................................................................... 195 Contributions to the Research............................................................................ 196 Scale Measurement and methodology Contributions............................ 197 Theoretical Contributions...................................................................... 198 Managerial Contributions...................................................................... 199 Directions for the Future.................................................................................... 203 Concluding Comments....................................................................................... 206 REFERENCES.............................................................................................................. 208 APPENDICES............................................................................................................... 223 Appendix A Pilot Study Survey...................................................................... 224 Appendix B Personal Interview Qu estions for Key Constructs...................... 233 Appendix C Pilot Study Inter-Item Correlations............................................. 234 Appendix D Pilot Study Item-T o-Total Correlations Scale Development Analysis................................................................. 244 Appendix E Pilot Study Summary of Factor Analysis Results....................... 247 Appendix F Pilot Study Item Indicator Retention........................................... 249 Appendix G Exploratory Factor Analysis Scale Purification.......................... 260 Appendix H Exploratory Factory An alysis Indictor Retention....................... 262 ABOUT THE AUTHOR....................................................................................... End Page

PAGE 10

vii LIST OF TABLES Table 2.1 Summary of Model Hypotheses................................................................... 59 Table 3.1 Hypothesized Relationships.......................................................................... 99 Table 4.1 Response Rates by Organization.................................................................. 106 Table 4.2 Sample Demographics.................................................................................. 107 Table 4.3 Pilot Study Key Statistical Results............................................................ 113 Table 4.4 Response Rates by Organization.................................................................. 115 Table 4.5 Non-Respondent Data................................................................................... 115 Table 4.6 Sample Demographics.................................................................................. 116 Table 4.7 EFA Key Statistical Results.......................................................................... 121 Table 4.8 CFA Key Statistical Re sults (N=260) Constructs........................................ 124 Table 4.9 Scale and Construct Reliab ilities and Variances Assessments Measurement Model....................................................................................... 136 Table 4.10 Final Measurement M odels Factor Loadings.......................................... 137 Table 4.11 Degree of Interrelati onship Between Constructs........................................ 145 Table 4.12 Summary of Findings of Proposed Hypotheses.......................................... 146 Table 4.13 Sources of Job St ress Regression Model.................................................... 150 Table 4.14 Mediation Analysis of Job Autonomy Relationships................................. 152 Table 4.15 Mediation Analysis of Emo tional Intelligence Relationships.................... 153 Table 4.16 Mediation Analysis for Emo tion-Focused Coping Effectiveness............... 162 Table 4.17 Mediation Analysis for Emo tion-Focused Coping Frequencies................. 163 Table 4.18 Outcome Correlations................................................................................. 167 Table 4.19 Job Performance Regression Model........................................................... 168 Table 4.20 Intention to Leave Regression Model......................................................... 170

PAGE 11

viii LIST OF FIGURES Figure 1.1 Structural M odel: Social Service Encounter Job Stress.............................. 16 Figure 1.2 AET............................................................................................................. 24 Figure 2.1 Structural Model with Hypothesized Relationships.................................... 33 Figure 2.2 Traditional Sour ces of Job Stress................................................................ 37 Figure 2.3 Mediating Role of Job Autonomy............................................................... 38 Figure 2.4 Mediating Role of Emotional Intelligence.................................................. 41 Figure 2.5 Outcome Variables Inter-Relationships on Job Performance..................... 50 Figure 2.6 Impact of Job-Related Outc ome Variables on Job Performance................. 53

PAGE 12

ix An Empirical Examination of Job Stress and Management of Emotionally-Based Behavior: Frontline Social Service Personnel Perspective Doreen Sams ABSTRACT Frontline service personnel (FSP) play an invaluable role in the marketing mix by directly influencing the custom ers perception of both the se rvice organization as well as the service quality during the face-to-face delivery service encounter (Ashforth & Humphrey 1993). The purpose of this dissert ation is to examine how sources of job stress for FSP such as perceived customer demands, role ambiguity, role conflict, role overload, and emotional labor and various mediators such as job autonomy, emotional intelligence, and emotion-focused coping direc tly and indirectly influence job stress and outcome variables such as attitudes, beha viors, physical consequences, emotional exhaustion, job performance, and intentions. Research from frontline social serv ice personnels (FSSP) perspective was collected in three phases: 1) ei ght personal in-depth interviews to determine the fit of the constructs in the model in a social service environment, 2) a cognitive response survey gathered from 86 FSSP to identify appropriate line items, and 3) a survey questionnaire gathered from 533 members of the National As sociation of Social Workers-Florida.

PAGE 13

x Psychometrically sound scales developed and purified in the study demonstrated reliability and validity. These scales were th en used to examine the structural model. Structural equation modeling, co rrelations, and regression analys es were used to examine relationships in the model. Results of the study indicated that self-management of emotionally-based behavior was significant in the creation and re duction of job stress. Findings suggest that the influe nce of emotional-based behavior plays a significant role in job performance at the social service encounter and indirect ly influences intention to switch and intention to leave.

PAGE 14

1 CHAPTER 1 Introduction Few would argue that toda ys business environment is experiencing dramatic changes due, in part, to rapid expansion in th e service sectors. Service organizations are becoming a dominant force in creating new jobs in the United States. The U.S. Department of Labor reported rapid growth in the service indust ry from 1991 to 2000. Service-oriented employment opportunities are projected to rapidly expand increasing in size by 5.1 million between 2000 and 2010 with over ha lf of those jobs within the health, business, and social services industries (Hecker 2001). With this rapid growth, there is a renewed interest among researchers in fo cusing on the critical issues underlying customers perceptions of service providers such as service encounter performance (Chenet, Tynan, & Money, 1999). Unlike traditional product exchange encounters, the service delivery encounter plays an important role in the development of the customers impression of the service provider (McAlexander, Kaldenberg, & Koen ing, 1994). A product can be described precisely (Shostack, 1977). However, servi ce images and service realities are based largely on our five senses (i.e., tangible things), but the service itself cannot be tangible. Thus, customers rely on peripheral clues (Shostack, 1977). Today, frontline service personnel (FSP) play an invaluable role in the formulation of customers perceptions about the service organization, as well as the service qual ity being provided during the

PAGE 15

2 face-to-face delivery encounter with the cust omer (Ashforth & Humphrey, 1993). FSPs attitudes and displayed behavior s can influence the customers perceptions of not only the service being created, but also the service organizations (Bowen & Schneider, 1985). Consequently, FSP are a significant component of the marketing and service delivery mix for any service-based organization. During the service delivery encounter, FSPs job performance plays a critical role in the cu stomers formulation of service quality and their satisfaction or dissatisfaction with the encounter. Some researchers suggest that FSPs are the pulse of the service organization and the organizations image is often portrayed by the actions of just one FSP (Folkes & Patrick, 2003). Other researchers note that customer satisfaction and actual purchase behavior are two important outcomes derive d from the service encounter and are a function of the customers service quality perceptions (Athanassopoulos, 2000; Chenet et al., 1999; Sweeney, Soutar, & Johnson, 1999). In an effort to control and manage the desired consistency of servi ce quality from one encounter to the next, many service organizations try to implement specific strategies that impose emotionally-based behavior display rules on their FSP during the servic e delivery encounter. While in some situations imposed emotionally-b ased behavior display rules are beneficial to FSP, these types of rules can also result in emotional la bor, a potential source of increased social service encounter stress that negatively aff ects service encounter performances (Singh, 1998). Social service encounter stress hereinafter is denoted as job stress. Emotionallybased behavior display rules, which are ac ts used in exchange under the guidance of standards dictated by manuals, training, etc., a nd not subject to change or termination, are ritually sealed and almost inescapable (Hochschild, 1983). Emotionally-based behavior

PAGE 16

3 display rules are used in a wide spectrum of services from airline flight attendants (e.g., Deltas friendliness) to bill collectors (e.g., intimidation) (Hochschild, 1983). Given the wide use of imposed emotionally-based behavi or display rules servi ce strategies and the critical importance of better understanding th e phenomena of servic e delivery encounter performance and job stress, it is surprising that there is litt le empirical research that focuses on FSPs abilities to self-manage thei r emotionally driven behaviors during the delivery process of co-creating and executing qu ality services with customers. The next section provides a discussion on the phenom ena of job stress and service delivery encounter performances within a social services framework. Job Stress and Service Deliv ery Encounter Performance We know from previous research that job stress continues to be a concern for U.S. firms (Sauter, Murphy, Colligan, Swanson, Hurrell, Jr., Scharf, Jr., Sinclair, Grubb, Goldenhar, Alterman, Johnston, Hamilton, & Ti sdale, 1999) and that job performance is empirically linked to job st ress (e.g., Singh, Goolsby, & Rhoads, 1994). Job performance at the service encounter is di rectly correlated to service qua lity (Chenet et al., 1999). The underlying assumption is that the customers pe rception of a service encounter is a vital part of the service. For that reason, the actual service deliv ery is considered the moment of truth (Carlzon, 1987; Normann, 1983). It is at th is point that the customer formulates the perception of quality, satisfaction, and long-term loyalty. Unlike the commercial service sector where the product offering is a good or service, social marketers in social service organizations primary o ffering is a behavior change (Andreasen, 1995). Because of this uniqueness, quality is very difficult for the

PAGE 17

4 customer to assess. In many social service encounters, customers are expected to voluntarily accept, reject, modif y, or abandon a behavior for their benefit, the benefit of a group, or of society without a promise of a di rect benefit or immediat e payback in return for their behavior change (Kotler, Robert o, & Lee, 2002). In turn, the price to the customer is often perceived by the customer as too high. Price may be characterized as an inconvenience, personal embarrassment, physical danger, or possible illness to the customer (Kotler et al., 2002). Customer s often do not see immediate or personal benefits and some are unwilling or reluctant to engage in the necessary behavior change. Consequently, social service delivery performance is of paramount importance to the customers perception of service quality and the displayed behaviors of frontline social service personnel (FSSP) may be the only su rrogate indicator for quality (Shostack, 1977). Quality may symbolize, to the customer the providers reliability, dependability, and consistency. All of these qualities are impor tant when a customer is expected to trust a service provider with something as importa nt as a behavior ch ange (Petrick, 2002). Poor social service delivery encounte r performance may mean more than unhappy customers or negative word-of-mouth advertis ing. Poor social service delivery encounter performance may mean that the customers harmful behavior continues, organizational funding disappears, others in need of the behavior change stay away, and/or society continues to be harmed by the behavior. Fr om an organizational perspective, research suggests that social service organizations may lose valuable FSSP through turnover brought on by anxiety, tension, emotional exhaus tion, and/or physical consequences such as hypertension, ulcers, etc., as a direct result of unregulated job stress (Sauter et al., 1999).

PAGE 18

5 Intuitively, unregulated j ob stress during a social service delivery encounter can be counterproductive for all the participants. Therefore, identifying the sources of job stress is critical for eliminating or reducing potential negative cons equences. Job stress difficulties facing many FSSP are due, in part, to the inherently large, complex, and bureaucratically burdened aspects that t ypically characterize many social service organizational structures. Research indicates that this type of organizational structure encourages the formulation of such sources of job stress as role ambiguity, role conflict, and role overload among FSSP (Erera, 1989). A classic example of such an organizational structure is the Department of Children and Family Services (DCFS). DCFS operates a foster care program, among other programs, that is fragmented into nearly two dozen independently run system s and staffed by an overburdened workforce (Gruskin, 2003). While the organizational structure contributes to FSSP job stress, customer demands and emotional labor are two additional sources of job stress that may significantly influence FSSP performa nce capabilities (Goo lsby, 1992; Morris & Feldman, 1996). These two sources are discussed in greater detail later in this chapter. Emotionally-Based Displayed Behaviors It is critical to remember that any service encounter includes social elements that are emotional in nature (Czepiel, 1990). For example, understanding job stress impact on FSSP performance goes beyond just identifying sources and consequences of job stress. In an effort to understand the role that job stress has on FSSP performance in social service delivery encounters, we must examine the role of emotionally-based behaviors as well. In many situations, social service customers are often angry or

PAGE 19

6 frustrated and tend to make unreasonable de mands of FSSP such as requesting that the FSSP bend rules and overlook something that e liminates them from obtaining a certain aspect of the service offered. When FSSP re spond to the customer with organizationally deemed inappropriate emotionally-based be havior, such behaviors tend to lessen FSSPs objectivity and ability to offe r a level of service quality expected by the organization (Lazarus & Folkman. 1984) and can place FSSPs physical well-being at risk. Past research suggests that FSSP are more likel y to be injured by angry customers than workers in almost any other occupation in th e United States (Princeton Survey Research Association, 1997). By including emotiona lly-based display behaviors into the examination of FSSP job stress and their conseq uences, the investiga tion is expected to provide valuable insights that will guide social service managers in developing and implementing strategies, policies, and traini ng programs that will have a more positive impact on the social service delivery encounter. As mentioned earlier, the need for social services in the U.S. is rapidl y increasing and it is thus beco ming one of the largest growth sectors for employment in the service sector (e.g., Hecker, 2001). There is a definite need to understand the implications of job st ress and social service delivery performance from a broader perspectiv e than in years past. Frontline Social Servic e Personnels Perspective Although researchers have given the issue of job stress a great deal of attention and practitioners have attempted to reduce the influences of the sour ces of job stress, to date job stress continues to be proble matic. For FSSP, job stress has increased substantially for more than 20 years (Princeton Survey Research Association, 1997).

PAGE 20

7 Today, FSSP have one of the ten most stressful jobs in the U.S. Narayanan, Menon, and Spector (1999) recently demonstrated that sour ces of job stress and reactions to job stress differ across job levels and types of employme nt. Therefore, one po ssible explanation as to why job stress continues to be problematic is that relevant sources of job stress for FSSP have not been identified. Narayanan, Menon, and Spector (1999) concluded that examining typical sources of job stress (i.e., role ambiguity, role conflict, and role overload) alone masks and/or ignores other important sources of job stress. A second possible explanation is that researchers have neglected the emotional aspect of job stress. Goolsby (1992) proposed that without incorporating emotions into the study of job stress reactions, the researcher c ould only see a partial picture of the reaction process. For example, it is suggested that implementing st rategies such as imposing emotionally-based behavior display rules will bring consistency to the service encounter (e.g., Ashforth & Humphrey, 1993); however, this strategy may also result in emotional labor rendering the service encounter dysfunctional. The role of FSSPs self-management of emotionallybased display behavior at the service en counter has been neglected. Therefore, empirically examining the role of self-m anagement of emotionally-based display behavior of FSSP at the serv ice encounter is imperative to solving the job stress puzzle by identifying additional pieces of the puzzle. It is extremely important to complete the picture for FSSP. While some of the sources of job stress (e.g., role ambiguity, role conflict, and role overload) for FSSP may not differ greatly from those of FSP in comm ercial services, the personal consequences may be dramatically different. For exam ple, social service customers tend to be emotionally charged making the social service delivery enco unter inherently dangerous

PAGE 21

8 for FSSP. The encounter is so dangerous that the Occupational Safety and Health Administration (Chao & Henshaw, 2003) annually publishes and updates a safety manual specifically instructing social workers on how FSSP can avoid serious injury at the hands of the customer. In many cases, social service customers do not come willingly to the service delivery encounter, but have ha d that decision thrust upon th em by a third party (e.g., the courts). Although the behavior change may not come about willingly, it is still necessary. Due to the circumstances (e.g., necessitated by court order), many soci al service clients tend to bring emotional baggage (e.g., pre-existi ng levels of anger, anxiety, frustration, tension, embarrassment, etc.) to the service enco unter that directly in fluences the type of demands (e.g., unreasonable), as well as the em otionally-based behavior, directed toward FSSP. Given the real and ever-p resent threat of customer in flicted injuries, FSSP must be cautious in allowing emotionally-based behavior to be part of the service delivery encounter. Overall, it is cri tical to identify sources of j ob stress and to understand how FSSP can most effectively manage the result ant emotions. While job stress cannot be completely eliminated, once significant sources of job stress are identified, effective strategies and policies can be designed in an attempt to brin g job stress to a level where productivity is no longer hampered by its negative effects. Ignoring or misinterpreting the causes of job stress and their conseque nces may lead to incorrect managerial decisions. By implementing the wrong polic y or strategy, management may unwittingly increase job stress and reduce the FSSP level of job performance, thus reducing service quality and customer satisfaction. FSSP hold valu able insights into job stress. Therefore,

PAGE 22

9 to truly understand job stress and its personal and organizational consequences, we need to ask FSSP. Previous Models Since the pioneering work of Jacobson, Charters, and Lieberman (1951), there has been a stream of literature in disciplines such as management (B edeian & Armenakis, 1981), marketing (Behrman & Perreault, Jr., 1984; Singh et al., 1994; Walker, Churchill, & Ford, 1975), organizational behavior (Mil es & Perreault, Jrs ., 1976) and psychology (Netemeyer, Johnston, & Burton, 1990) exploring th e role of job stress in organizations. The most commonly studied sources of job st ress include role ambiguity (RA), role conflict (RC), and role overload (RO). Howe ver, in the Miles a nd Perreault, Jr. (1976) study, RO is considered a type of RC and not a separate construct. More recently, Singh, Goolsby, and Rhoads (1994) built a theoretical rationale for the partial mediation thesis of relationships between role stressors, bur nout, and job outcomes. Their study examines role overload as one of three separate sources of job stress that when combined with other sources of job stress (e.g., emotional labor and perceived customer demands) may cause a cumulative dysfunctional effect on outcome variables. Bedeian and Armenakis (1981) propose a theoretical model of cause-effect relationship in which they find RA and RC to have a significant direct effect on job satisfaction, with a greater indirect effect through job tension. In an industrial setting, Walker, Churchill, and Ford (1975) conducte d a meta-analysis of key components of sales force job stress and find that RA and RC correlate more highly with job performance than any other pr edictor utilized. Behrman and Perreault, Jr. (1984) find unmediated job stress on ps ychological outcomes (i.e., job satisfaction, organizational

PAGE 23

10 commitment, and turnover intention) to be statistically significant; whereas, Singh, Goolsby, and Rhoads (1994) find that unmedia ted job stress on psychological outcomes is non-significant. Netemeyer, Johnston, a nd Burton (1990) compared four job stress models and find that four of the eight relationships found in the Bedeian and Armenakis (1981) model are not supported. From the st udies above, we know that RA, RC, and RO have either a direct or indi rect effect on outcome variab les; however, the lack of consistent findings in the above studies indica tes a need for investigating the impact of other factors as well. Shortcomings of Current Approaches to Job Stress Marketing researchers have examined antecedents and consequences of job stress at the service encounter from a sales fo rce management perspective (e.g., Behrman & Perreault, Jr., 1984) and, more recently, the c onsumers perspectives (Folkes & Patrick, 2003) to the neglect of the FSSPs perspect ive. Ignoring the FSSPs perspective of sources of job stress, how emotions are self-managed, and stre ss at the service encounters only allows us to see a small portion of the job stress picture. A nother shortcoming is that services researchers examining self-man agement of emotionally -based behavior for FSSP traditionally examine only emotional labo r and its impact on personal outcome variables (e.g., emotional exhaus tion, physical illness, mental). Yet, self-management of emotionally-based behavior displays is more complex than simply adhering to organizationally formulated emotionally-based display behavior rules. FSSP with a high level of emotional intelligence may not function well in an organization that imposes emotionally-based behavior display rules. Emotionally intelligent FSSP have the ability

PAGE 24

11 to understand and effectively deal with cl ients; therefore, imposed rules lacking flexibility may prove counterproductive by incr easing job stress for the FSSP when he or she determines that the situation warrants emotional displays different from imposed display rules. Hence, it is shortsight ed to examine emotional labor without simultaneously examining emotional intelligence. Based on an extensive literature search, th ere is no meaningful empirical research reported in the services litera ture of emotional intelligence as a form of self-management of emotionally-based display behavior at the service encounter. However, emotionally intelligent FSP have the ability to perceive and self-man age emotional responses of customers (Salovey & Mayer, 1990). An emotionally intelligent FSSP may posses a greater ability to respond to displayed custom er emotions without allowing emotions to inhibit his/her rational th inking than would a FSSP me rely following a set of emotionally-based behavior display rules. Examining emotional labor and emotional intelligence together is expected to offer insight into how two very different methods for self-management of emotions (i.e., imposed emotionally-based behavior and emotional intelligence) may interact to reduce or increase job stress and its consequences. Understanding the relationship of self-man agement of emotionally-based display behavior at the social service delivery enc ounter allows management to determine if interventions such as imposed emotionally -based behavior display rules should be considered. If managerial intervention is determined to improve job performance then the next natural step is to determine where, when, and which interventions are appropriate (Bitner, Booms, & Tetreault, 1990).

PAGE 25

12 Perceived customer demands are also impor tant, yet have recei ved little empirical investigation as a source of job stress for FSSP. When a FSSP perceives that he or she is unable to meet customer demands for quality and reliability, percei ved customer demands become a source of job stress (Chung & Schneid er, 2002). The FSSPs ability to adapt in order to meet perceived customer demands has been shown to be of significant value in determining a customers dis/ satisfaction with the service (Bitner et al., 1990). In a previous research study in the services lit erature, customer demands are empirically studied as an antecedent of role conflict (Chung & Schneider, 2002). However, customer demands are a separate and distinct source of job stress stemming from outside the FSSPs organizational role, and should be examined as such (e.g., Goolsby, 1992). Another construct that is loosely em pirically linked to job stress and job performance in the marketing and services literature is job autonomy. Yet, it has not been empirically examined as a mediator of perceived customer demands and job stress. By giving FSSP the power to make important day-to-day decisions, job autonomy might reduce the amount of job stress produ ced by perceived customer demands. While many studies examine intentions to leave an organization, no studies were found where FSSP switch to different positions within the organization. Switching behavior is directly related to the cost of switching provi ders of a service (Burnham, Frels, & Mahajan, 2003). It makes sense that there are switching costs for changes in employment as well. If the costs are t oo high for the FSSP to leave the organization altogether (e.g., vested in re tirement, poor job market, etc. ), the FSSP may move to what her or she perceives as a less stressful position within the same firm The move may be temporary while waiting for the job market to improve or long-term. Switching behavior

PAGE 26

13 within an organization may be as costly to the firm as FSSP leaving the firm. Clients comfortable with current FSSP may find conf iding in new FSSP stressful and may be reluctant to work with them. Without incorporating the above constructs into a job stress model, effects may yield incomplete or misleading results and conclusions. Research Objectives One of the research objectives for this di ssertation is to determine the role of emotional intelligence and emotional labor wi thin self-management of emotionally-based display behavior in the social service delivery enc ounter from the FSSPs perspective. In addition, this dissertation examines the influence of perceived customer demands (PCD) and emotional labor (EL), two empirically neglected sources of job stress on outcome variables (e.g., attit udes, behaviors, emotional exhaus tion, physical consequences, and intentions). Due to inconsistencies in the literature, this dissertation re-examines role ambiguity (RA), role conflict (RC), and role ov erload (RO) as sources of job stress. The mediating effects of job autonomy (JA), emotion-focused coping (EFC), and emotional intelligence (EI) are also examined. Another re search objective is that of empirically examining the relationships of job stress a nd social service encounter performance on a currently unexplored outcome, intention to switch positions within the organization and/or intention to leave the organization. Social service en counter performance herein is denoted as job performance (JP). Furthermor e, this dissertation examines relationships between job stress and self-management of emotionally-based behavior on outcome variables from an interdisci plinary research orientation.

PAGE 27

14 SelfManagement of Emotionally-Based Displayed Behavior Emotional Intelligence Emotional intelligence (EI) is a form of self-management of emotionally-based displayed behavior. EI is th e ability of a person to manage his/her own emotions and to perceive and manage the emotional responses of other individuals (Salovey & Mayer, 1990). A highly emotionally intelligent indivi dual does not allow his/her emotions to inhibit his/her rational thinking. Sojka and Deeter-Schmelz (2002) propose that anecdotal evidence suggests EI to be important in situati ons requiring adaptation and the ability to cope. Thus, an FSSP who has a high level of emotional in telligent would have the ability to adapt to emotionally-based behavior during th e social service encounter. Emotional Labor In an effort to manage the percep tion of service quality, more and more organizations are using strategies that impose emotionally-based behavior display rules to manage FSP service behavior during the service delivery enc ounter. Although these types of strategies can result in positive job performance outco mes, they might also create less desirable outcomes (Ashforth & Humphr ey, 1993). The notion of emotional labor (EL) (see page 17 for definition) resulting from imposed emotionally-based behavior display rules was first proposed as a source of job stress, but not empirically examined by Hochschild (1983). Better understanding of EL in a social service delivery environment becomes critically important to the main premise of social marketing which is to improve society, not harm it (Kotler, Roberto, & Lee 2002). As a potential s ource of job stress, some researchers believe that imposed emoti onally-based behavior rules resulting in EL

PAGE 28

15 could have both negative emotional and physical consequences (Morris & Feldman, 1996; Rafaeli & Sutton; 1987). Consequently, EL can be counterproduct ive to the social goal of bettering society. Emotion-Focused Coping It has been acknowledged for years that job stress produces negative emotions (Kahn, Wolf, Quinn, Snock, & Rosenthal, 1964). More recently, Goolsby (1992) proposes that the magnitude and direction of consequences of job stress for sales force personnel are partially determined by the coping strategies available to the individual. Two types of coping strategies presented in previous studies are: 1) problem-focused coping and 2) emotion-focused coping (L azarus & Folkman, 1984). Problem-focused coping is an active attempt to alter the source of the st ress itself. FSSP have limited influence over the sources of job stress w ithin an organization, therefore, problemfocused coping is not examined in this study. Emoti on-focused coping consists of passive or reactive strategies (e.g., denial acceptance, behavioral and/or mental disengagement) used by FSSP to reduce effort to control experienced-stress coming from the work environment (Goolsby, 1992). Some researchers suggest that if coping resources are exceeded, job stress becomes dysfunctional and may result in negative personal and organization consequences (J ackson, Schwab, & Schuler, 1986; Babakus, Cravens, Johnston, & Moncrief, 1999). Emotion-focused coping is believed to reduce the magnitude of emotional distress that taxe s FSSP coping resources through modification of perceptions, attitudes, a nd/or goals (Cherniss, 1980).

PAGE 29

16 Proposed Theoretical Framework In order to empirically examine the influence of self-management of emotionally-based display behavior and job stress, the following framework is used (Figure 1.1 below). Figure 1.1: Structural Model: Social Services Encounter Job Stress + Emotion-Focused Coping PhysicalConsequences Attitude Towardthe Job Attitude/FeelingsToward OverallOrganizationalCommitment Attitude/FeelingToward OverallJob Satisfaction EmotionalExhaustion Intention toSwitch Positionsw/I theOrganization Intention to Leavethe Organization PerceivedCustomerDemands Role Ambiguity Role Conflict Role Overload Emotional Labor Job Autonomy Social ServiceEncounterStress (JobStress) EmotionalIntelligence Social ServiceEncounterPerformance(Job) ++---+++-+++--+---Previously Examined Signifcant Relationship New Relationship Under Examined Relationship Definitions of Constructs In an effort to gain an understanding of the proposed model, the following conceptual definitions have been adopted. Attitude-Toward-the Job (ATJ) is a composite of learned predispositions toward reacting positively or negatively toward a particular job.

PAGE 30

17 Emotion-Focused Coping (EFC) is the cognitive and behavioral effort (i.e., passive or reactive strategi es) put forth by Frontline Service Personnel to manage specific extern al and/or internal demands to reduce effort to control felt stress. Emotional Exhaustion (EE) is a specific stress related action that refers to a state of depleted energy cause d by excessive emotional demands made of social service employees interacting with customers (Saxton, Phillips, & Blakeney, 1991). Emotional Intelligence (EI) is an individuals learned ability to selfmonitor his/her own emotions and the feelings and emotions of customers resulting in useful info rmation for guiding his or her own thinking and actions (Salovey & Mayer, 1990). Emotional Labor (EL) is the degree of ef fort exerted by the FSP to display or resist organizationa lly based emotional behavior. Intention to Leave the Organization (ITL) represents the likelihood of a FSP to voluntarily end his or he r current employment with the organization at some time in the future (Bluedorn, 1982; Johnston & Futrell, 1989). Intention to Switch Positions within the Same Organization (ITS) represents the likeli hood of a FSP to voluntarily end a current position and take another position within the same organization sometime in the future.

PAGE 31

18 Job Autonomy (JA) is the degree of cont rol over job-related activities an employee feels he or she possess (Wang & Netemeyer, 2002). Overall Organizational Commitment (OOC) is the FSPs perception relating to relative stre ngth of identification wi th, as well as his/her dedication to, that organization. Overall Job Satisfaction (OJS) is the FSPs comp rehensive evaluative judgment (or feeling) toward his/her job. Perceived Customer Demands (PCD) are the FSPs perception of customers requirements and expectations for service quality, reliability, and deliverability. Physical Consequences (PC) are the resulting negative outcomes that directly relate to actual me ntal and medical conditions. Role Ambiguity (RA) is the FSPs perception of uncertainty about what his or her tasks are in carrying ou t his or her job that results in increased levels of job stress. Role Conflict (RC) is the degree of perceived incongruence or incompatibility of organizational expectations associated with job performance that result in in creased levels of job stress. Role Overload (RO) is the FSPs perception that there is a definite imbalance between job tasks required and time allocated to complete those tasks resulting in increased levels of job stress.

PAGE 32

19 Social Service Encounter Performance (JP) is the FSPs selfassessment of how well he or she ex ecutes job tasks, responsibilities, and assignments. Social Service Encounter Stress (JS) is a dynamic condition in which an individual is confronted with an opportunity, constraint, or demand on being/having/doing what one de sires for which resolution is perceived to have uncertainty, but which will lead to important outcomes (McGrath, 1976; Schuler, 1980). Highlights of Proposed Model The model presented in Figure 1.1 serves as a visual aid in understanding the theoretical framework that guides the current em pirical investigation. This section offers only highlights of the workings of the m odel. The models constructs and their relationships are discussed in more detail in the next chapter. As the model illustrates, three typical organizationally-bas ed sources of job stress (JS) [i.e., role ambiguity (RA), role conflict (RC), and role overload (RO)] and their relations hips are re-examined in the current study. In addition, several of the traditionally studied personal outcome consequences [i.e., organizat ional commitment (OC), overall job satisfaction (OJS), emotional exhaustion (EE), social service delivery encounter performance (JP) and intention to leave (ITL)] that have had vary ing results are also re-examined within a social service context. Two neglected constructs, perceived customer demands (PCD) and emotional labor (EL) and their proposed relationshi ps to outcome variables are empirically

PAGE 33

20 examined in the current study. There are thr ee additional constructs in the model (i.e., job autonomy, emotional intelligence, and emotion-focused coping) with proposed mediating influences on the identified emo tional, attitudinal, and physical outcome consequences that are worthy of further highlight. In many social service organizations, job -related decisions are typically made in a bureaucratic fashion by top management. In contrast, many FSSP are not normally in a position that allows them to make operational or strategic ma nagerial decisions that could reduce or eliminate sources of their job stre ss. Nevertheless, job autonomy differs as a function of job type and is, therefore, a cr itical facet of the social service work environment (e.g., Hall, 1989; Karasek, Jr., 1979). When FSSP have the latitude to make routine daily decisions, job autonomy might re duce FSSPs job stress as well as the need to engage in emotion-focused coping. Therefore, job autono my is presented in the model as a mediator of the relationship between pe rceived customer demands and job stress. Another proposed mediator presented in the model is emotional intelligence. An emotionally intelligent FSSP has the ability to manage and appropriately display emotionally-based behavior because he or sh e has the ability to perceive and understand his or her own emotions, as well as the emo tions of others (Salovey & Mayer, 1990). Therefore, FSSP with emotiona lly intelligent capa bilities are expected to display the proper emotionally-based behaviors with or without imposed emotionally-based behavior display rules. Consequently, emotional in telligence is presented in the model as a mediator of the relationship between emotional labor and job stress. In turn, high levels of emotional intelligence should influence FSSPs need to engage in emotion-focused coping strategies.

PAGE 34

21 Emotion-focused coping encompasses strate gies to reduce the level of experienced job stress, not the conditions causing the stress. When engaging in emotion-focused coping, the emphasis is on controlling emotiona l responses rather th an controlling the work environment (Narayanan et al., 1999). Gi ven that management tends to make most strategic job-related decisions, it is posited that F SSP primarily engage in emotionfocused coping rather than problem-focused coping. The model presents emotionfocused coping as a mediator between job stress and several specific outcome variables. Other researchers such as Folkman and Lazar us (1988) have examined emotion-focused coping as a mediator of various outcome vari ables. This dissertation presents emotionfocused coping in mediating roles that have received limited attention in the services literature. Emotion-focused coping is also presented as mediating emotional exhaustion and physical consequences. Although most past studies focus on ex amining the relationships between job stress and intention to leave, it is intuitive that not everyone dissatisfied with his or her job is also dissatisfied with the organizati on. As such, the current study also examines intention to switch positions within a company as an alternative to intention to leave. While some FSSP who are dissatisfied with their job will immediately leave an organization, others may intend to switch posit ions within the same organization only as a temporary solution until the timing is ri ght for them to leave the organization. Therefore, the model presents a direct rela tionship between intention to switch positions within the organization and inten tion to leave the organization.

PAGE 35

22 Research Questions The following research questions serve as the underpinnings, which guide the current research endeavor and the developm ent of the hypothesized relationships between the modelss various constructs. 1. To what extent does job autonomy mediate the relationship between customer demands and job stress? 2. To what extent does job autonom y directly influence emotionfocused coping capabilities of FSSP? 3. To what extent does emotional intelligence mediate the relationship between emotional la bor and job stress? 4. To what extent does emotional intelligence directly influence emotion-focused coping capabilities of FSSP? 5. To what extent does emotion-focused coping mediate the relationship between job stress and personal and organizational consequences? 6. To what extent does job performan ce directly influence intention to switch positions within organization? 7. To what extent does job performan ce directly influence intention to leave the organization? 8. To what extent does intention to switch positions within the organization directly influence inte ntion to leave the organization?

PAGE 36

23 Anticipated Contributions of the Investigation Anticipated contributions fall into three categories: theoretical, measurement, and practitioner oriented. This section presents the proposed contributions of this dissertation for these three areas. Theoretical Contributions Affective event theory (AET) brings atte ntion to events that may unfold in the workplace and generate emotional reactio ns (Weiss & Cropanzano, 1996). AET holds that environmental conditions in the workplace result in hassles and uplifts for the employee (Weiss & Cropanzano, 1996). For example, if an employee receives conflicting instructions (i.e., ro le conflict) from important ro le partners, they may feel hassled, frustrated, and disgruntled (an affective state). These feelings may manifest themselves in a poor attitude toward the j ob and a resultant search for other positions within the firm or outside the firm. Affect Event Theory (AET) presents emotions at five levels: within person, between persons, interpersonal intera ctions, groups, and organization wide (See Figure 1.2 below) (Ashkanasy, 2003). This dissertation contributes to AET by investiga ting the sources of job stress that drive the events that unfold in the workplace such as perceived customer demands, emotional labor, role ambiguity, role conflict, role overload, as well as the role of mediators (e.g., emotionfocused coping, emotional intelligence and j ob autonomy) in reducing the influence of affective events on FSSP. Previous resear ch has neither proposed nor empirically examined the relationships between levels (e.g.., emotional intelligence influences the experienced emotions caused by emotional labor) This dissertation contributes to theory

PAGE 37

development by examining relationships between two constructs across levels of interpersonal (emotional labor) and between persons (emotional intelligence) defining both as different means of self-management of emotionally-based behavior displays, as well as the nature of those interactions. Figure 1.2 AET 5. Organization-wideOrganizational policies; requirement for emotional labor;stress and wellbeing; emotional climate and culture4. GroupsAffective composition; emotionally intelligent groups;emotional contagion; leader-member exchange3. Interpersonal InteractionsEmotional labor; emotional exchange;displayed vs. felt emotion2. Between personsTrait affectivity, affective commitment; jobsatisfaction; burnout; emotional intelligence1. Within-personState affect; affective events;discrete emotions; mood; behaviors 5. Organization-wideOrganizational policies; requirement for emotional labor;stress and wellbeing; emotional climate and culture4. GroupsAffective composition; emotionally intelligent groups;emotional contagion; leader-member exchange3. Interpersonal InteractionsEmotional labor; emotional exchange;displayed vs. felt emotion2. Between personsTrait affectivity, affective commitment; jobsatisfaction; burnout; emotional intelligence1. Within-personState affect; affective events;discrete emotions; mood; behaviors This dissertation takes a theoretical role approach emphasizing the nature of people as social actors who learn behaviors appropriate to their position in society (Solomon, Surprenant, Czepiel, & Gutman, 1985). Role theory has particular relevance for FSSP as it provides a major link between FSSP and the organizational levels of research and scientific inquiry. It is a summation of mechanisms through which systems confront their members as individuals, as well as a building block of social systems (Kahn et al., 1964). Organizational expectation about FSSPs behaviors allows FSSP to react accordingly. For example, FSSP have expectations about performance at the service encounter and attempt to act accordingly. Customers hold expectations about FSSP service encounter behavior. Displayed behaviors that differ from behavioral 24

PAGE 38

25 expectations may negatively affect the customer-FSSP relationship (Solomon et al., 1985). Results from recent studies demonstrate that examining organizational variables offers great potential for ex tending knowledge of the role of FSSP (e.g., Chenet et al., 1999). The results of such studies are promisi ng, yet little appears to be known about the overall organizational dynamics of the FSSPs workplace from the FSSP perspective. This dissertation extends role theory by bringing together and empirically examining influences on the organizational environment, and interpersonal and intra-personal factors of perceived customer demands and emotional la bor, respectively. Also explored are the variables that mediate the influence of j ob stress as well, as typical organizational outcome consequences. Measurement Contributions Much of the work done in the area of job stress (i.e., RA, RC, RO) relies on traditional rating scales (e.g., closed-end measur es). Respondents report on job stress over an unnamed time. However, these m easures alone may not adequately capture stress-related experiences related to th e social service environment of FSSP. Where scales do not exist or are inad equate for this study, new scales are generated or existing scales are modified by this researcher through pertinent l iterature and a pilot study including both closed and open-ended measures. New and modified scales are psychometrically tested (see Chapter 3, Methodology).

PAGE 39

26 Contributions for Practitioners Because it is possible for customer satisfaction and repeat purchase to be determined solely by job performance at the service encounter (Bowen & Schneider, 1985), understanding the roles of job stress a nd self-management of emotionally-based display behaviors at the soci al service delivery encounter are extremely important. Service providers that are better equipped to ha ndle emotionally stressful situations at the service encounter should impr ove the service encounter and, hence, service quality and customer satisfaction. For service providers holding a mark et-orientation philosophy, the service encounter provides a perfect opportunity to collect information about customers and, therefore, better serv e them. FSSP who are stressed cannot accomplish this because they are distracted by their own emotional mana gement and coping. Knowledge gained in these encounters could provide a competitive advantage for a not-for-profit organization as only a limited number of not-for-profit firms embrace a market orientation philosophy (Cano, Carrillat, & Jaramillo, 2004). This dissertation offers so cial service practitioners a clearer understa nding of the sources of job stress, their consequences and the role of self-management of emotionally-based displayed behavior at the so cial service delivery encounter. With the knowledge gained from this study, social serv ice practitioners have the opportunity to make informed decisions to manage the social service delivery encounter. In addition, some of the insights gained from the current study may provide inva luable insights for similar service encounters where imposed emo tionally-based behavior display rules are employed to ensure service quality performance.

PAGE 40

27 Chapter Summary Due to conflicting results in findings from previous research, this dissertation reexamines traditional sources of job stress and their outcomes. Two additional sources of job stress, emotional labor and perceived customer demands, have received limited empirical research and are empirically exam ined in this dissertation. In previous research, emotional intelligence has primarily been examined as a strategy for improving managers emotional competencies; however, th e contribution of emo tional intelligence of FSP to service delivery performance is neglec ted. There is no empirical evidence as to emotional intelligences role on job stress. Th erefore, the inclusion of EI, a form of selfmanagement of emotionally-based behavior, is exploratory in nature and EI is empirically examined as a mediator between emotional la bor and job stress. Em otional intelligence is also examined as to its direct influen ce on emotion-focused coping. This dissertation examines job autonomy as a mediator between perceived customer demands and job stress and the direct relationship of job autonomy on FSSPs emotion-focused coping ability. The findings from this study are expected to help direct future job stress research toward considering self-management of em otionally-based display behavior and the nature of the job in FSSPs appraisal of, and the reaction to, stressful job conditions. The findings are also expected to further knowledge as to relevant sources of job stress in a social service environment and the extent to which mediators (e.g., emotion-focused coping, emotional intelligence, and job autonomy) reduce the influen ce of job stress. Measuring job stress may require more than a general agreement on events that are present in the workplace. Where scales ar e inadequate or do not exist to measure a

PAGE 41

28 construct, scales are developed through th e literature and a pilot study with FSSP and psychometrically examined. The following chapter is an in-depth literatur e review that delves into relationships of constructs offered for examination in this dissertation. The con ceptual and functional definitions of the studys key constructs are presented. The model is examined and deeper explanations of hypothesized relationships are offered

PAGE 42

29 CHAPTER 2 Literature Review and Hypotheses The model presented in Chapter One depict s a situation in which frontline social service personnel (FSSP) are conf ronted with job stress from sources such as emotional labor, perceived customer demands, role ambi guity, role conflict, and role overload. A theory of job stress is without substance unl ess emotions are incor porated into the study Goolsby (1992). Together, Affective Even t and Role theories offer researchers a framework for understanding how self-manag ement of emotionally-based behavior during the service encounter influenc es job stress and its consequences. In an attempt to better un derstand the proposed relations hips presented in Chapter One, this chapter presents an overview of th e research relevant to the proposed models sources of job stress, mediators, and consequences for frontline social service personnel and the organization. Hypotheses are formed ba sed on the literature and are presented in this chapter. A summary of the hypotheses is presented in Table 2.1 on pages 59-60. Affective Event Theory AET is a relatively new theory that purpor ts that things happen to individuals in a work setting and the individual often reac ts emotionally (Weiss & Cropanzano, 1996). AET focuses on structure, causes, and conse quences of affective experiences occurring on the job (Weiss & Cropanzano, 1996). AET hol ds that affective reactions influence

PAGE 43

30 overall feelings about ones job. Cumulativ e affective experiences contribute to the affective component of attit udes such as job satisfaction and judgment-driven behaviors (Fisher & Ashkanasy, 2000) su ch as job performance. The higher the number of similar affectiv e events, the more likely a predictable behavior will occur (Fisher & Ashkanasy, 2000) Work environments influence affective experience by making specifi c events more likely (Wei ss & Cropanzano, 1996). For instance, an emotionally charged work envir onment, such as one that may be found in a social service organization, creates an atmo sphere where perceived customer demands could result in a confronta tion between the customer and frontline social service personnel (FSSP). Emotion-focused coping (e.g., Goolsby, 1992) and self-management of emotions (e.g., Diefendorff & Gosserand, 2003) are proposed to indirectly influence such affect driven behaviors as job performance. By examining the influence of selfmanagement of emotional-based behavior, this study systematically examines elements of AET and adds value to the understanding of th e role of affective ev ents in the social service environment. Role Theory An interdisciplinary role perspective evolved from studies in anthropology, psychology, and sociology (Sarbin, 1954). Ro le theorys roots are found in social penetration theory (Altman & Taylor, 1973), soci al interaction approach es to sociological thinking (Simmel, 1908/1950; Goffman, 1967), a nd the dyadic elements of social exchange theory (Homans, 1961; Kelley & Th ibaut; 1978). Role theory is a grouping of social clues that guide and di rect an individuals behavior in a gi ven setting and is

PAGE 44

31 important in explaining patterns of social interaction and inte gration. It is relevant to marketing through its focus on social exchange within marketing encounters (Broderick, 1999). Role theory applies across a wide vari ety of contexts incl uding a dyadic focus on social exchanges between only two parties w ho have a mutually reciprocal relationship (Biddle, 1979) and a more generalized social exchange between three or more actors who benefit each other indirectly and involves de layed reciprocity in which, over time, both recipients give and receive something in a circular process (Levi-Strauss, 1969). Generalized social exchange takes place in so cial services encounters. Social service organizations provide public value to ci tizens by providing private value (e.g., money, food, etc.) to the needy. Citizens do not e xpect to receive value immediately, but do receive value, over time, in the form of such benefits as lower crime rates when human basic needs are met. Role theory holds that expectations of what is appropriate service encounter behavior for FSSP comes from multiple role members. Whether expectations are met appropriately and to what degree is determin ed by the reaction of other role members and observers. Other members of the role set, su ch as immediate supervisors, executives in the firm, and/or internal/external customer s, communicate pressure s for conformity to their expectations through organizational norms or formal training, company handbooks, supervisor directions, and/or apprenti ceships (Merton, 1957; Solomon et al., 1985). By taking a role-theoretic approac h, this study emphasizes the nature of FSSP as social actors who learn appropria te behaviors for their role in an organization on-the-job (Solomon et al., 1985). By including self-management of emotionally-based behavior and emotion-focused coping in the model, this

PAGE 45

32 study extends role theory a nd affective event theory by looking at previously unexamined relationships of emotion-focused coping with both job autonomy and emotional intelligence. This research is also expected to move towards an understanding of the processes, which may affect both the level of job stress and its outcome consequences, by building upon in terdisciplinary research to develop a systematic framework that guides empirical research. Sources of Job Stress Job stress, also referred to as role stre ss (e.g., Behrman & Perreault, Jr., 1984), is a complex and widely studied psyc hological construct. What c onstitutes job stress depends on when an environmental situation is perc eived as presenting a demand which threatens to exceed the persons capabilities and res ources for meeting it, under conditions where he or she expects a substantial differential in the rewards and costs from meeting the demand versus not meeting it (McGrath, 1976, p. 1352). Figure 2.1 below presents a visualization of the hypothesized relations hips found in the proposed model.

PAGE 46

Figure 2.1: Structural Model with Hypothesized Relationships Emotion-Focused Coping Physical Consequences Attitude Toward the Job Attitude/Feelings Toward Overall Organizational Commitment Attitude/Feeling Toward Overall Job Satisfaction Emotional Exhaustion Intention to Switch Positions w/I the Organization Intention to Leave the Organization Job Autonomy Emotional Intelligence Social Service Encounter Performance (Job) H H5H6H7(a-b)H8(a-b)H10(a-b)H9(a-b)H11(a-b)H12(a-b)H16H18H17 H15(a-c)Ho1(a-c)H21H23H22H20H19Ho13(a-b)H1 Role Ambiguity Role Conflict Role Overload Social Service Encounter Stress (Job Stress) H2H3 Perceived Customer Demands motional Labor E In the following sections, the overall model is broken down to enhance the readers understanding of th e hypothesized relationships. Traditional Sources of Job Stress Role ambiguity. Role ambiguity is the frontline social service personnels perception of uncertainty about what tasks are involved in carrying out his or her job. There appears to be sufficient evidence to suppor t the assertion that ro le ambiguity is an inherent characteristic of many jobs, including that of frontline service employees (e.g., Singh, 2000). However, some studies suggest this relationship does not exist for other job types, such as clerical workers, poli ce sergeants, and university professors (Lord, 1996; Narayanan et al., 1999). These conflic ting findings suggest that role ambiguity may be a function of job type and may not be typical to all organizations. For example, 33

PAGE 47

34 Narayanan et al. (1999) measur ed sources of job stress by asking respondents from three occupations to describe a conc rete event that he or she perceives to be stressful which occurred at work in the mont h prior to the study. Naraya nan et al. (1999) also asked respondents how stressful the event was (i.e., not very to very much) and why the incident is perceived to be a problem for them and to report their accompanying emotional reaction to the incident. In the Na rayanan et al. (1999) study, role ambiguity is rarely identified as a source of job stress acr oss three occupations. Whereas, in a recent study, Bettencourt and Brown (2003) found role am biguity to be a significant source of job stress that has an indirect negativ e influence on service delivery (i.e., job performance) of frontline service personnel (i.e., both nonprofessional service providers and higher-level service providers) significan t at p<.001. Role ambiguity for social service organizations such as the Department of Children and Family Services (DCFS) often comes from fragmented programs like the foster care program. The fragmented foster care program lacks uniform performance standards which creates an atmosphere of uncertainty for FSSP (Bridges & Lauer, 2003). Not knowing what is expected creates ambiguity, which in turn, may increase job stress for frontline service personnel (Kahn et al., 1964). Therefore, the study of role ambiguity as a source of job stress is relevant for FSSP. Role conflict. Role conflict is a factor in the work environment that creates psychological conflict for frontline serv ice personnel (Kahn et al., 1964). The relationship between role conflic t and job stress is clearly de fined through past empirical research. For example, in 1978, Cooper, Mal linger, and Kahn found that in a sample of

PAGE 48

35 dentists, role conflict contribu ted significantly (p< .01) to job stress. In the case of dentists, role conflict stems from what a den tist believes society perceives him or her to be (i.e., inflictor of pain) and what the dentist believes he or she should be (i.e., healer). In a recent study of 514 human service workers, the relationship of role conflict and job stress is path analyzed using ordinary-leas t squares regression anal ysis and is found to have a positive significant relationshi p at p <.001 (Lait & Wallace, 2002). Role conflict is operationally defined as a multidimensional construct. However, exactly what constitutes those dimensions has been a subject for controversy. Role theory distinguishes between four types of role conflict: 1) intrasende r, 2) intersender, 3) personrole, and 4) overload (Shenkar & Zeira, 1991). Intrasender conflict is defined as the inability to satisfy all expectations of a single member of a role set. For example, a supervisor requests that an FSSP working for the Department of Children and Family Services (DCFS) obtain records on an at-ris k child which are unava ilable through normal channels. At the same time, DCFS prohibi ts the FSSP from obtaini ng the records outside of the normal channel. Intersender conflict is the inability to satisfy all expectations from multiple members of the role set. For in stance, an FSSP at the DCFS must answer to external customer wants and needs, satisfy government regulations special interest groups, and managements expectations. The e xpectations of all of these contingencies often conflict. Person-role conflict exists wh en a persons values are inconsistent with role requirements. An illustration of person -role conflict is when an FSSP, in an attempt to advance in his or her job, circumvents company policies and regulations by obtaining needed documents through unauthorized cha nnels. Although role overload has been studied as a dimension of role conflict, re search of a sample of 770 customer service

PAGE 49

36 representatives conducted by Singh, Goolsby, and Rhoads (1994) demonstrates that role overload and role conflict are tw o separate constructs that co ntribute independently to job stress. Therefore, for the purpose of this st udy, role conflict consists of three dimensions: 1) intrasender, 2) intersender, and 3) person-role conflict. Role overload Role overload is the perception that there is an imbalance between job tasks requirements and time allocated to comp lete those tasks. It has been shown that role overload is a sign ificant source of job st ress for customer service representatives in the commercial service sector (Singh, G oolsby, & Rhoads, 1994). Role overload is believed to be a significant concern for fr ontline service personnel in social service organizations. A case in point would be social service pe rsonnel at Department of Children and Family Services in Miami, Florid a. These personnel have been portrayed as victims of an overburdened workforce (Grusk in, 2003). According to Gruskin despite the overtime, the backlog sits at just under 1,700 cases 13% higher than when the backlog unit was established two years ago ( p. 1). In turn, some South Florida social workers entrusted to make life-and death de cisions for the states abused and neglected children routinely clock 20 to 40 hours of overtime a week (p. 1).

PAGE 50

37 Figure 2.2: Traditional Sources of Job Stress Role Ambiguity Role Overload Role Conflict Job Stress H1 (+)H2 (+)H3 (+) The three typical sources of job stress illustrated in Figure 2.2 above are believed to be relevant in a social service context and, therefore, are examined. All three sources of job stress are expected to significantly create job stress. Hence, the following hypothesized relationships are offered: H 1 : Role ambiguity has a significant positive impact in creating FSSP job stress. H 2 : Role conflict has a significant positive impact in creating FSSP job stress. H 3 : Role overload has a significant positive impact in creating FSSP job stress.

PAGE 51

38 Other Sources of Job Stress and Mediating Relationships Perceived customer demands. Perceived customer demands are the FSSPs perceptions of requirements and expectations of the customer. Based on the centrality of the role of FSSP and the customer as co-producers of the service offering (Chung & Schneider, 2002), perceived customer demands are proposed to influence various job outcomes in a social service organization. Job stress may occur when FSP perceive that customer demands are incongruent (i.e., not consistent) with what FSP believe management expects them to do (Chung & Schneider, 2002). Furthermore, Li and Calantone (1998) and Wang and Netemeyer (2002) examine customer demandingness in relationship to various outcome variables including job performance, but neither study examines its role as a source of job stress. The current study examines the role of perceived customer demands as a source of job stress for social service organizations. Mediating Role of Job Autonomy Figure 2.3: Mediating Role of Job Autonomy PerceivedCustomerDemands Job Stress Job Autonomy H4 Job autonomy demonstrates to FSSP the managers belief in him or her as worthy of trust to resolve problems and meet perceived customer demands. When an employee

PAGE 52

39 takes personal responsibility for performance outcomes, he or she gains a sense of pride of ownership in the job (Tyagi, 1985), whic h promotes a learning atmosphere (Parker, Wall, & Jackson, 1997). In an atmosphere c onducive to learning, employees often seek knowledge outside of their immediate job res ponsibilities and increased knowledge is expected to lead to the production of a quality product (P arker et al., 1997). Wang and Netemeyer (2002) found that for a sample of 147 sales force personnel consisting of real estate agents, job aut onomy motivates individuals to lear n and develop skills needed to perform successfully on the job. As illustrate d in Figure 2.3 above, it is proposed that job autonomy arms FSSP with the knowledge and authority needed to meet perceived customer demands and problems effectively a nd efficiently. Therefore, job autonomy is viewed as mediating the impact of perceived customer demands that create job stress. H 4 : Job autonomy mediates the relati onship between perceived customer demands and job stress such that job autonomy reduces the impact of perceived customer demands on job stress. Emotional Labor Negative emotions short-circuit a persons ongoing cognitive processes by changing a persons motivations and/or goals (Lerner & Keltner, 2001). Frontline service personnel in a social service organization ri sk coming face-to-face daily with angry and often violent customers (Princeton Survey Research Association, 1997). Because emotions are contagious and injury at the ha nds of customers is a real and ever present threat for FSSP, these service providers cannot afford to display inappropriate

PAGE 53

40 emotionally-based behavior at the servi ce counter (Verbeke, 1997). Consequently, controlling emotionally-based behavior disp layed by FSSP is important to the efficient and effective delivery of quali ty service and safety of FSSP (Lazarus & Folkman, 1984). Ashforth and Humphrey (1993) propose th at one way an organization can assure consistency of the service delivery encounter is to impose emotionally-based behavior display rules (i.e., prescribed method of e xpression of on-the-job emotions) on FSSP. Although emotionally-based behavior disp lay rules can benefit the organization and FSSP, they can also have negative consequences such as emotional labor. Emotional labor is the effort put forth to display emo tionally-based behavior deemed appropriate for a given situation such as the service delivery encounter (Morris & Feldman, 1996). Emotional labor has received attention in th e literature since Hochsc hilds (1983) seminal book. Practitioners such as Vitello-Cicciu (2003) acknowledge the role of emotional labor in a service environment. Management academics such as Ashforth and Humphrey (1993), as well as academics from organizational behavior (e.g., Schaubroeck & Jones, 2000), demonstrate an interest in emotional labor and its conseque nces. Ashforth and Humphrey (1993) propose that the act of di splaying an emotion is emotional labor and that the amount of effort put into the acting determines whether there are negative personal consequences such as job stress. In other words, the more effort involved in performing an emotionally-based behavior, the greater the le vel of emotional labor. Schaubroeck and Jones (2000) examine poten tial physical consequences of emotional labor when negative emotions are suppressed, but find no significant relationship between emotional labor and physical consequences Marketing academics Rogers, Clow, and Kash (1994) examine emotional labor by meas uring the display of empathy of frontline

PAGE 54

41 service personnel and find that as genuine empathy for the customer increases, job stress declines. Mediating Role of Emotional Intelligence Figure 2.4: Mediating Role of Emotional Intelligence Emotional Labor EmotionalIntelligence Job Stress H5 In contrast to emotional labor which is in response to a potential affective event, emotional intelligence is not an affective event or a response to an affective event. Emotional intelligence is a learned ability to self-monitor ones own emotions and the emotions of others. Emotional intelligence equips an individual to effectively manage emotionally-based behavior. Emotional intelligence is not a new concept. Its roots go back to the concept of social intelligences, first introduced by Thorndike (1920) and subsequently extended by Gardner (1983). As illustrated in Figure 2.4 above, the current study proposes that, based on the concept that emotional intelligence equals emotional competency, an emotionally intelligent FSSP senses, knows, and displays emotionally-based behaviors that are appropriate at the service encounter eliminating the need to act out an imposed

PAGE 55

42 emotionally-based behavior (Goleman, 1998). Emotional intelligence is proposed to mediate the relationship of emotional labor and job stress by reducing the impact of emotional labor that creates job stress. H 5 : Emotional intelligence significantly mediates the relationship between emotional labor and job stress such that it reduces the impact of emotional labor on job stress. Mediating Role of Emotion-Focused Coping An emotion-focused coping style involves a cognitive and behavioral effort to manage demands in an effort to reduce felt stress such as distancing oneself from stress through avoidance, seeking social support, selective attention, etc. (Lazarus & Folkman, 1984). Although emotion-focused coping reduces job stress, the negative consequences of emotion-focused coping may outweigh the benefit. An individual may give up on organizational goals which over the long-run affect the individual s and organizations well-being (Strutton & Lumpkin, 1994). Once the FSSP gives up on organizational goals, job performance may decline and the FSSP may leave or be asked to leave his or her position. It stands to reason that reducing the need fo r emotion-focused coping could be advantageous for both FSSP and the organi zation. It is proposed that job autonomy and emotional intelligence may reduce the ne ed for FSSP to engage in emotion-focused coping.

PAGE 56

43 Job Autonomy and Emotion-Focused Coping Job autonomy can reduce the need to enga ge in emotion-focused coping strategies by proactively reducing job stress (Singh, 1993) For example, Singh (1993) found that boundary-spanning personnel in situations where job autonomy exists experience less job stress. Chebat and Kollias (2000) examine the relationship between job stress and job autonomy and discover that when job stress is reduced through th e influence of job autonomy, frontline service managers are more willing to adapt to demands of the service delivery encounter. Therefore, it is proposed that when job autonomy is present, the need to engage in emotion-focused coping is re duced. Although explorat ory in nature, it is reasonable to hypothesize a relationship be tween job autonomy and emotion-focused coping based on previous research findings. H 6 : The stronger the presence of j ob autonomy, the less likely FSSP will engage in emotion-focused coping. Emotional Intelligence and Emotion-Focused Coping FSSP need to be self-directed, to pos sess the ability to adapt to changing circumstances, to hold social competencies and to be creative (Goleman, 1995). The emotionally intelligent FSSP has the capabi lity to recognize and understand his or her emotions and the emotions of others in the workplace. This permits him or her to be able to adapt to changing circumst ances and to display the prope r emotional behavior. To date, no other studies have been found that empirically examine th e role of emotion-

PAGE 57

44 focused coping and emotional in telligence. Therefore, this study contributes to the literature by examining the followi ng exploratory relationship: H 7a : There is a significant relationship between emotional intelligence and emotion-focused coping effectiveness. H 7b : There is a significant relationship between emotional intelligence and emotion-focused coping frequency. The Impact of Emotion-Focused Coping Because the coping process arises during the service encounter and changes the relationship between job stress and outcom es, emotion-focused coping is proposed to significantly mediate the rela tionships between job stress and the following outcome variables: 1) emotional exhaustion, 2) overa ll organizational commit ment, 3) overall job satisfaction, 4) attitude toward the job, a nd 5) physical consequences. Hypotheses are presented following each of the next four (4) subsections: 1) Emotional Exhaustion, 2) Attitude Toward the Job, and 3) Overall Organization Commitment and Overall Job Satisfaction, and 4) Physical Consequences. Impact on Emotional Exhaustion Although human responses to stress prepare a person for action in case of danger, job stress in the form of work overload experienced every day may dr ain an individuals energy allowing a state of emotional exhaustion to ensue (Houkes, Janseen, DeJonge &

PAGE 58

45 Bakker, 2003). Emotional exhaustion is a st ress-related reaction caused by excessive emotional demands made of FSSP that, in turn, deplete the FSSPs emotional energy (Saxton, Phillips, & Blakeney, 1991). The emoti onal cost often manifests itself as a sense of futility, loss of self-esteem, acute anxiety, confusion and indecision, systems of hypertension, and psychosomatic disorders (Kahn et al., 1964). Emotional exhaustion is shown to occur wh ere job stress stems from interpersonal interaction, and not in monotonous jobs, but in jobs like those held by FSSP with high arousal. As such, emotional exhaustion is of ten experienced by personne l in jobs such as sales force, social service, teachers, human se rvices, health care, and criminal justice (e.g., Babakus et al., 1999; Savicki & Cooley, 1994; Jackson et al., 1986; Cordes, Dougherty, & Blum, 1997; Leiter, 1991; Maslach & Jackson 1 979). In other words, the greater the job stress the more likelihood of emotional exhaustion. Based on previous empirical research, the following hypotheses are offered: H 8a : Emotion-focused coping effectiv eness significantly mediates the positive relationship between job stress and emotional exhaustion such that it reduces the impact of job stress. H 8b : Emotion-focused coping frequency significantly mediates the positive relationship between job stress and emotional exhaustion such that it reduces the impact of job stress.

PAGE 59

46 Impact on Attitude Toward the Job Attitude toward the job is loosely defi ned in most studies as a work-related attitude. Saks and Cronshaw (1990) defined it as a general feeling of favorableness and liking toward a job (p. 21). In contrast, Kidwell, Jr. and Bennett (2001) and Lee, Carswell, and Allen (2000) defined attitude toward the job as a combination of job satisfaction and job involvement; whereas, Aghazadeh (1999) opera tionalized attitude toward the job as a component of overall job satisfaction. Because th e literature does not offer a clear definition of att itude toward the job or defini tions presented in previous research include other attitudina l constructs, attitude toward the job is defined based on a social psychological perspective that attitude s are learned. Therefore, for the purpose of this study, attitude toward the job is viewed as a composite of learned predispositions toward reacting positively or negatively towa rd a particular job. As defined, attitude toward the job is a negative or positiv e judgment based on knowledge gained through experience accumulated on the job. Based on an understanding of attitudes, coping, and job stress, emotion-focused coping is proposed to reduce the negative influence of job stress on attitude toward the job. H 9a : Emotion-focused coping effectiv eness significantly mediates the negative relationship between job st ress and attitude toward the job such that it reduces the impact of job stress.

PAGE 60

47 H 9b : Emotion-focused coping freque ncy significantly mediates the negative relationship between job st ress and attitude toward the job such that it reduces the impact of job stress. Impact on Overall Organizational Comm itment and Overall Job Satisfaction Previous studies establish the presence of inverse relationships between job stress and both overall job satisfaction and overall organizational commitment. In the sales force literature, several studies including Busch and Bush (1978), Bagozzi (1980), Teas (1983) and, more recently, Singh, Verbeke, a nd Rhoads (1996) verify the findings of Churchill Jr., Ford, and Walker (1976) that job stress negatively influences job satisfactions. Moncrief, Babakus, Craven s, and Johnstons (1 997) study of 188 boundary spanning sales force personnel demonstrates th e negative influence of overall job stress on overall organizational commitment ( -1.80, t = -2.59). Another recent study in marketing also demonstrates an inverse re lationship between sources of job stress ( -.15, t = -3.7) and organizational commitment (Si ngh, 1998). Based on research findings in the coping literature, emotion-focuse d coping is expected to reduce the influence of job stress on these two outcome variables. Hen ce, the following hypotheses are offered: H 10a : Emotion-focused coping effectiv eness significantly mediates the negative relationship between job st ress and overall job satisfaction such that it reduces the impact of job stress.

PAGE 61

48 H 10b : Emotion-focused coping freque ncy significantly mediates the negative relationship between job st ress and overall job satisfaction such that it reduces the impact of job stress. H 11a : Emotion-focused coping effectiven ess significantly mediates the negative relationship between job stress and overall organizational commitment such that it reduces the impact of job stress. H 11b : Emotion-focused coping frequenc y significantly mediates the negative relationship between job stress and overall organizational commitment such that it reduces the impact of job stress. Impact on Physical Consequences Work-related matters, including job stre ss, are linked to various dysfunctional physical consequences includi ng high blood pressure, ulcers, etc. (Kimes, 1977). For example, job stress is shown to result in ne gative physical consequences such as increased heart disease (Sales, 1969). Sales experime nted on 73 undergraduate students by drawing blood before and after the st ress treatment was administered to examine changes in cholesterol levels with findings of an a pproximate five percent increase from the participants mean cholesterol level (p <.10) for participants experiencing job stress. The Center for Womens Healthcare (1998) also reports that long-term effects of job stress take aim at an individual s physical well-being result ing in diseases such as gastrointestinal disease, diabetes, and even the common cold. Due to the exploratory

PAGE 62

49 nature of the proposed mediating relationship of emotion-focused coping on job stress and physical consequences, the following hypothesized relationships are offered: H 12a : Emotion-focused coping effectiv eness significantly mediates the positive relationship between job stress and physical consequences such that it reduces the impact of job stress. H 12b : Emotion-focused coping freque ncy significantly mediates the positive relationship between job stress and physical consequences such that it reduces the impact of job stress. Interrelationships of Outcome Variables Empirical and theoretical wo rks addressing the interrelati onships of the constructs examined in this study indicate that many of the proposed outcome variables are highly interrelated to one another. Previous research of cust omer service representatives demonstrates a significant negative influen ce of emotional exhaustion on job satisfaction (p< .01) (Singh et al., 1994). The Babakus et al. (1999) study of sale s force attitudes and behaviors also demonstrates the significant negative influence of emotional exhaustion on both overall organizational commitment (t = -2.3 0) and overall job sati sfaction (t = -4.04) significant at p <.05. These findings indicate a significa nt influence of emotional exhaustion on both overall organizational commitment and overall job satisfaction. However, there are cases where research findings conflict. Bateman and Strasser (1984) in a longitudinal study of employees of a nursing department found organizational

PAGE 63

50 commitment to be an antecedent of job satisfaction. But, in a meta-analysis by Brown and Peterson (1993), overall job satisfaction is found to directly influence organizational commitment. Differences in findings may be explained by differences in the operationalization of a construct (i.e., unidimensional or multidimensional), conceptual definitions, or whether the study is based on a sound theoretical base. Because of the lack of consistency in findings of interrelatedness among outcome variables, these relationships are revisited in our study (see Figure 2.5 below). Figure 2.5: Outcome Variables Interrelationships on Job Performance PhysicalConsequences Attitude Towardthe Job Attitude/FeelingsToward OverallOrganizationalCommitment Attitude/FeelingToward OverallJob Satisfaction EmotionalExhaustion Ho13cHo13bHo13aHo13dHo14aHo14bHo14cH15aH15bH15c It is expected that, as an attitudinal outcome variable, attitude toward the job should follow the direction of other attitudinal variables in the model such as overall job satisfaction and overall organizational commitment. Burke and Deszca (1986) assessing

PAGE 64

51 frequency of physical consequences and emotional exhaustion find that emotional exhaustion influences several physical conseq uences including headaches and chest pain. Based on a lack of empirical examination of relationships between both attitude toward the job and physical consequences with othe r outcome variables, Hypotheses 13 and 14 are exploratory in nature. HO 13 : Attitude toward the job (ATJ) is not significantly correlated with other outcome variables in the model. HO 13a : Attitude toward the job (ATJ) is not significantly correlated with overall organizational commitment (OOC). HO 13b : Attitude toward the job (ATJ) is not significantly correlated with overall job satisfaction (OJS). HO 13c : Attitude toward the job (ATJ) is not significantly correlated with physical consequences (PC). HO 13d : Attitude toward the job (ATJ) is not significantly correlated with emotional exhaustion (EE).

PAGE 65

52 HO 14 : Physical consequences (PC) ar e not significantly correlated with outcome variables of overall organizational commitment (OOC), overall job satisfaction (OJS), or emotional exhaustion (EE). HO 14a : Physical consequences (PC) ar e not significantly correlated with overall organizational commitment (OOC). HO 14b : Physical consequences (PC) ar e not significantly correlated with overall job satisfaction (OJS). HO 14c : Physical consequences (PC) ar e not significantly correlated with emotional exhaustion (EE). H 15 : Overall organizational co mmitment (OOC), overall job satisfaction (OJS), and emo tional exhaustion (EE) are significantly correlated with each other. H 15a : Overall organizational commit ment (OOC) is significantly correlated to overall j ob satisfaction (OJS). H 15b : Overall organizational commit ment (OOC) is significantly correlated to emotional exhaustion (EE).

PAGE 66

53 H 15c : Overall job satisfaction (OJS) is significantly correlated to emotional exhaustion (EE). Relationships between attitudinal variables (i.e., attitude toward the job, job satisfaction, and organizational commitment) and job performance are presented in the following section (see Figure 2.6 below). Impact of Job-Related Outcome Variables on Job Performance Figure 2.6: Impact of Job Related Outcome Variables on Job Performance Attitude/FeelingsToward OverallJob Satisfaction PhysicalConsequences Attitude TowardThe Job Attitude/FeelingsToward OverallOrganizationalCommitment EmotionalExhaustion Job Performance H20H16H17H18 H19

PAGE 67

54 Impact of Attitude Toward the Job Attitude toward the job is yet untested in its relationship to job performance. Based on past research of other attitude variables such overall organizational commitment, attitude toward the job is anticipated to positively influence job performance. In other words, FSSP holding a positive attitude toward the job perform better on the job. H 16 : Attitude toward the job has a sign ificant positive influence on FSSPs job performance. Impact of Overall Job Satisfaction There has been a plethora of research on the relationship between job satisfaction and job performance, but with contradict ory findings. Bagozzi (1980), Behrman and Perreault, Jr. (1984), and Hamp ton, Dubinksy, and Skinner (1986) empirically show that job performance had a positive influence on job satisfaction, debunking the theory that satisfied workers are more productive. In contrast, Fisher (2003) proposes that the way job satisfaction and job performance are conceptualized and operationali zed may contribute to findings that are contrary to the common-sense theory that job satisfaction leads to job performance (p. 754). Fisher examines the relationship of job satisfaction and job performance using alternative conceptualizations with significan t findings of a positive relationship between job satisfaction and job performance. A meta-analysis by Judge, Thoresen, Bono, and Patton (2001) presents findings that correlations between satisfaction-performance are

PAGE 68

55 stronger positively in high-complexity jobs. Frontline social service jobs are considered complex based on the diversity and extent of knowledge and skills needed to deal with multiple contingencies. In the current study, FSSP are asked to express their perception of job satisfaction and job performance rather than relying on the external judgments of supervisors or human resource reports. As a resu lt, it is expected that job satisfaction will positively influence job performance. H 17 : Overall job satisfaction has a significant positive influence on job performance. Impact of Overall Organizational Commitment Previous research is unclear as to the relationship between organizational commitment and job performance. However, Brown and Petersons (1993) meta-analysis demonstrates that job performance positivel y influences organizational commitment. Other studies demonstrate small, but significant positive relationships between organizational commitment and job performan ce. For example, one sales force study by Bashaw and Grant (1994) demonstrates that organizational commitm ent is significantly related positively to job pe rformance (p < .041). In a recent study by Bettencourt and Brown (2003), researchers found customer-oriented frontline service personnel service delivery behaviors to be ne gatively affected by job stress through organizational commitment. The studies presented a bove cast a shadow of doubt on the true directionality of the relationship. However, based on the notion that a person with a high level of organizational commitment may be motivated to high levels of performance in

PAGE 69

56 order to help the organization achieve its objectives, the following relationship is hypothesized. H 18 : Overall organizational commitment has a significant positive influence on job performance. Impact of Emotional Exhaustion Cordes and Dougherty (1993) state th at individuals in boundary spanning positions, such as FSSP, tend to experience hi gher levels of emotional exhaustion than many other positions. Some researchers argue that emotional exhaustion is a strong predictor of job performance only th rough organizational commitment and job satisfaction (Lee & Ashforth, 1993). However, others such as Wright and Cropanzano (1998) find that emotional exhaustion significan tly negatively influences job performance (p < .05). The Cropanzano, Rupp, and Byrne (2003) study of both public and private sector organizations including human services manufacturing, and f itness organizations found that emotional exhaustion significantly negatively influences job performance (p < .01). In keeping with findings of past empirical studies, the following relationship between emotional exhaustion and job performance is offered: H 19 : Emotional exhaustion has a signifi cant negative influence on job performance.

PAGE 70

57 Impact of Physical Consequences In her research of wellness in the workplace, Madsen (2003) proposes that improving the physical wellness of an individual influences hi s or her job performance. In other words, the better a FSSPs health, the more likely job performance will improve. Physical consequences are proposed to influence job performance based on the hypothesis that poor health of any kind may cause bad working conditions (DeJonge, Dormann, Janssen, Dollard, Landeweerd, & Nijhuis, 2001). H 20 : Physical consequences have a si gnificant negative influence on job performance. Predicting Behavioral Intentions Research shows that inte ntion to leave a job plays a central role in turnover models (Johnston & Futrell, 1989; Lee & Mowday, 1987; Sage r, Futrell, & Varadarajan 1989; Sager & Menon, 1994). Intention to leav e is generally viewed as the employees intent to end his or her current employme nt. Studies show that stressful working conditions are actually associated with inte ntions by workers to quit sales force jobs (Moncrief et al., 1997). Findings that job stress indirectly in fluences intention to leave also holds for customer service positions (Singh et al., 1994). The possibility exists that FSSP who are invested in an organization, but find that job stress negatively influences their job perf ormance, will have a propensity to switch positions within the organization rather than leave the organization. It is proposed that switching positions within a firm may also be a precursor to intention to leave an

PAGE 71

58 organization. Individuals performing poor ly on the job who attr ibute their reduced performance to job stress may switch positio ns temporarily while waiting for another position to open up in a different organiza tion. Although no specific empirical research is found on intention to switch positions within an organization, it stands to reason, based on knowledge of intention to l eave, that there should be a negative relationship between job performance and intenti on to switch positions. H 21 : Job performance has a significant negative influence on intention to switch positions within the organization. H 22 : Intention to switch positions within an organization and intention to leave an organization are correlated. H 23 : Job performance has a significant negative influence on intention to leave the organization. Table 2.1 below recaps the testable hypot heses for this study. The table also visually presents the proposed dire ction for each hypothesized relationship.

PAGE 72

59 Table 2.1 Summary of Model Hypotheses Testable Hypotheses Relationships H 1 : Role ambiguity (RA) has a significan t positive impact in creating FSSP job stress (JS). RA JS + H 2 : Role conflict (RC) has a significant positive impact in creating FSSP job stress (JS). RC JS + H 3 : Role overload has a significant po sitive impact in creating FSSP job stress (JS). RO JS + H 4 : Job autonomy (JA) mediates the relationship between perceived customer demands (PCD) and job stress (JS) such that job autonomy (JA) reduces the impact of job stress. PCD JA JS H 5 : Emotional intelligence (EI) significantly mediates the relationship between emotional labor (EL) and job stress (JS) such that it reduces the impact of emotional labor (EL) on job stress (JS). EL EI JS H 6 : The stronger the presence of job autonomy (JA), the less likely FSSP will engage in emotion-focused coping (EFC). JA EFC H 7a : There is a significant relationship between emotional intelligence (EI) and emotion-focused coping effectiveness (EFCE). EI EFCE H 7b : There is a significant relationship between emotional intelligence (EI) and emotion-focused coping frequency (EFCF). EI EFCF H 8a : Emotion-focused coping effectiveness (EFCE) significantly mediates the positive relationship between job stress (JS) and emotional exhaustion (EE) such that it reduces the impact of job stress (JS). JS EFCE EE H 8b : Emotion-focused coping frequency (EFCF) significantly mediates the positive relationship between job stress (JS) and emotional exhaustion (EE) such that it reduces the impact of job stress (JS). JS EFCF EE H 9a : Emotion-focused coping effectiveness (EFCE) significantly mediates the negative relationship between job stress (JS) and attitude toward the job (ATJ) such that it reduces the impact of job stress (JS). JS EFCE ATJ + H 9b : Emotion-focused coping frequency (EFCF) significantly mediates the negative relationship between job stress (JS) and attitude toward the job (ATJ) such that it reduces the impact of job stress (JS). JS EFCF ATJ + H 10a : Emotion-focused coping effectiveness (EFCE) significantly mediates the negative relationship between job stress (JS) and overall job satisfaction (OJS) such that it reduces the impact of job stress (JS). JS EFCE OJS + H 10b : Emotion-focused coping frequency (EFCF) significantly mediates the negative relationship between job stress (JS) and overall job satisfaction (OJS) such that it reduces the impact of job stress (JS). JS EFCF OJS + H 11a : Emotion-focused coping effectiveness (EFCE) significantly mediates the negative relationship between job stre ss (JS) and overall organizational commitment (OOC) such that it reduces the impact of job stress (JS). JS EFCE OOC + H 11b : Emotion-focused coping frequency (EFCF) significantly mediates the negative relationship between job stre ss (JS) and overall organizational commitment (OOC) such that it reduces the impact of job stress (JS). JS EFCF OOC + H 12a : Emotion-focused coping effectiveness (EFCE) significantly mediates the positive relationship between job stress (JS) and physical consequences (PC) such that it reduces the impact of job stress (JS). JS EFCE PC H 12b : Emotion-focused coping frequency (EFCF) significantly mediates the positive relationship between job stress (JS) and physical consequences JS EFCF PC

PAGE 73

60 (PC) such that it reduces the impact of job stress (JS). HO 13 : Attitude toward the job (ATJ) is not significantly correlated with other outcome variables in the model. HO 13a : Attitude toward the job (ATJ) is not significantly correlated with overall organizational commitment (OOC). ATJ OOC HO 13b : Attitude toward the job (ATJ) is not significantly correlated with overall job satisfaction (OJS). ATJ OJS HO 13c : Attitude toward the job (ATJ) is not significantly correlated with physical consequences (PC). ATJ PC HO 13d : Attitude toward the job (ATJ) is no t significantly correlated with emotional exhaustion (EE). ATJ EE HO 14 : Physical consequences are not signi ficantly correlated with outcome variable (OV) of overall organiza tional commitment (OOC), overall job satisfaction (OJS), or emotional exhaustion (EE). HO 14a : Physical consequences (PC) are not significantly correlated with overall organizational commitment (OOC). PC OOC HO 14b : Physical consequences (PC) are not significantly correlated with overall job satisfaction (OJS). PC OJS HO 14c : Physical consequences (PC) are not significantly correlated with emotional exhaustion (EE). PC EE H 15 : Overall organizational commitment (OOC), overall job satisfaction (OJS), and emotional exhaustion (EE) are significantly correlated with each other. H 15a : Overall organizational commitment (OOC) is significantly correlated with overall job satisfaction (OJS). OOC OJS H 15b : Overall organizational commitment (OOC) is significantly correlated with emotional exhaustion (EE). OOC EE H 15c : Overall job satisfaction (OJS) is sign ificantly correlated with emotional exhaustion (EE). OJS EE H 16 : Attitude toward the job (ATJ) has a significant positive influence on FSSPs job performance (JP). ATJ JP + H 17 : Overall job satisfaction (OJS) has a significant positive influence on job performance (JP). OJS JP + H 18 : Overall organizational commitment (OOC) has a significant positive influence on job performance (JP). OOC JP + H 19 : Emotional exhaustion (EE) has a significant negative influence on job performance (JP). EE JP H 20 : Physical consequences (PC) have a significant negative influence on job performance (JP). PC JP H 21 : Job performance (JP) has a significant negative influence on intention to switch positions within the organization (ITS). JP ITS H 22 : Intention to switch positions (ITS) within an organization and intention to leave an organization (ITL) are correlated. ITS ITL + H 23 : Job performance (JP) has a significant negative influence on intention to leave the organization (ITL). JP ITL

PAGE 74

61 Chapter Summary Chapter Two consists of a comprehensive l iterature review of relevant constructs and theories. The preceding literature re view connects previous research with hypothesized relationships found in the current investigati on and highlights neglected areas of research. The work presented in this chapter lays the foundation for the remainder of the study by adding depth to our understanding of relati onships presented in this study. The literature review above is a mean s of moving the read er to a clearer understanding of relationships within the ove rall model. This is accomplished through the examination of sources of job stress th at have been neglected in the marketing literature (i.e., perceived cust omer demands and emotional labor) as well as typically examined sources of job stress (i.e., role ambiguity, role conflict, role overload), neglected mediating relationships (job au tonomy, emotional intelligence, emotionfocused coping), various means of managing em otionally-based behavior (i.e., emotional intelligence and emotional labor), coping st rategies (i.e., emotion focused coping), various neglected outcome variables (i.e., phys ical consequences, attitude toward the job), new outcome variables (i.e., intention to switch positions within the organization), and typically examined outcomes (i.e., ove rall organizational co mmitment, overall job satisfaction, emotional exhaustion and inte ntion to leave the organization). The next chapter focuses on the methodology used to test a holistic model that investigates antecedents and consequences of job stress for FSSP. The chapter begins with a description of the research setting and sample characteristics. Next, an

PAGE 75

62 explanation of the measures used and data collection procedures are presented. The chapter concludes with a description of the analytical techniques.

PAGE 76

63 CHAPTER 3 Methodology This chapter describes the methodology used to test relationships involved in the social service encounter deliver y stress model presented in prev ious chapters for frontline social service personnel. Fi rst, the research setting a nd sample characteristics are presented. Second, a detailed explanation of the measures used and data collection procedures is presented. Fina lly, the justification of the c hosen analytical technique is presented. Research Setting The target population consists of fron tline service personnel employed by social service organizations serving clie nts in the State of Florida. Social service organizations such as the Suncoast Center for Community Mental Health and the Family Service Centers, Inc. were used in the scale deve lopment phase of the study. Members of the National Association of Social Workers of Fl orida and a sample of FSSP from a social service organization in the Tampa Bay area we re used to test the hypotheses. Sampled FSSP serve customers face-to-face on a daily basis in emotionally charged work environments where the emotions he or sh e are expected to display are dictated by organizational norms (written or implied). Frontline social servi ce personnel are from social service organizations varying in size and offerings which are anticipated to produce a representative sample of the target population.

PAGE 77

64 Procedures for Securing Participation The current study employs corporate buy-in as the method of securing subject participation. A letter of introduction including a brief overview of the study was presented to decision makers of social service organizations identified through the College of Public and Allied He alth, University of South Florida, Tampa. The letter was the first step in requestin g permission to survey FSSP of these social service organizations. To enhance the organizations buy-in participation, a follow-up telephone call to the key decision maker of each organi zation occurred within one week after the delivery of the letter of introduction. Disc ussions focused on the need for the study, the benefits to the organizations, and soliciting as sistance in obtaining potential participants names and contact information. In a furt her attempt to encourage organizational participation, the letter of introduction informed decision makers of participating organizations that they will be given an executive report of the research findings including a summary of the pertinent statisti cal analysis and manage rial implications. The report includes an outline stating the benefits and drawbacks of various means of self-management of emo tionally-based behavior. Sampling Requirements for Pilot Study The current study consists of actually two studies. A pilot study was conducted that focused on collecting data for assessing cons truct validity and reliability issues. The participant sample for the p ilot study was randomly drawn fr om the list of FSSP provided by the participating organizations. A survey instrument, letter of introduction, participation instructions, a nd a self-addressed stamped enve lope were sent to each

PAGE 78

65 selected participant. To en courage participation, the survey packet also included a prize entry ticket. Sampling Requirements for Final Study Based on the construct development and re finement results from the pilot study, the final survey instrument was developed. Th e final survey instrument was delivered to 2,500 FSSP from the participati ng organizations. Based on the rule of thumb of 10 observations per measured construct, 270 us able surveys were acceptable (27 measured constructs x 10 observations). Three hundred usable surveys obtaining sufficient power of >.80 for summated scales was not an i ssue (MacCallum, Browne, & Sugawara, 1996). Each survey packet included a cover letter on University of South Florida Department of Marketing letter head explaining to the respon dent the importance of the research, the benefits of being part of this study, instructions on how to complete and return the survey, a return stamped self-addr essed envelope, and a prize ticket for the drawing of one of several gift certificates. All participants were asked to return the original survey and prize ticket by the return date designated in the letter. Non-Response Bias Non-response bias is associat ed with the failure to obtain a 100% response rate from the sample and to differences in key at titudes, behaviors, as well as psychographics and demographic characteristics between re spondents and non-respondents (Mangione, 1995). Efforts to improve the response rate include: 1) offering a monetary incentive (e.g., chance at a drawing), 2) ge nerating a positive response towa rd the research itself, 3)

PAGE 79

66 decision maker of the organiza tion publicizing the study prior to requesting participation, and 4) a postcard or email to non-respondents. Participants received a postcard two week s after the initial mailing thanking those who had responded and politely reminding thos e who had not responded to do so. Based on the low probability of obtaining a 100% response rate, a shorter version of the survey instrument, including constructs that are critical to the st udy, was sent to a second small group of non-respondents (Mangione, 1995). Re sponses from previous non-respondents were compared to responses of known responde nts using one-way analysis of variance (ANOVA) and t-tests to determ ine if differences in respons es for critical variables between the two groups (respondents verse non-re spondents) were statistically significant (Lambert & Harrington, 1990). Four weeks after the pre-selected return deadline date, a drawing takes place for the awarding of the gift certificates. On e winning ticket is drawn for every 200 usable surveys returned to the researcher. The part icipants with the winning tickets were called or emailed to request delivery instructions of his or her gift ce rtificate. Student volunteers not associated with the research project were solicited from a Junior/Senior level class in the College of Business at th e University of South Florida, Tampa, to extract the winning tickets. Construct Development Since all of the key constructs in the main study were borrowed from the social psychology and marketing disciplines and applie d to an untested unique social service provider environment, it was necessary to as sess how well the existing structures fit the

PAGE 80

67 new environmental setting. To this end, a pilot study was conduc ted using a direct cognitive response approach (Ortinau & Bren singer, 1992). In this approach, the items relating to each construct, based on the input from service provider experts and relevant literature, were given to 361 FSSP randomly drawn from lists of FSSP provided by the participating organizations. Using a unique se t of seven-point scale descriptors with the endpoints described as = Not At All a Factor to = Definitely a Factor, participants were asked to express to what extent each item indicator is related to the described construct (see Appendix A). On this study s instrument, each construct and its assumed indicators are displayed together to ensure that the participants are well-focused on one construct at a time. For each construct, participants were given the opportunity to provide additional item indicators that they felt were relevant to the co nstructs structure. The sole purpose of the pilot study direct cogni tive approach was to determine which line item indicators provided the most appropriate representation for each latent construct from the respondents perspective. The value of the pilot study results lie in determining which manifest indicators to use in the actual scale measurement phase of the study. To evaluate and refine each constr ucts structure and address validity and reliability issues, the study data was submitted to an exploratory factor analysis procedure. In addition, the item indicator s were evaluated for item difficulty, discriminatory power, as well as internal consistencies. Scale Measurement Development Based on the construct structure integr ity found from doing direct cognitive structural analysis in the pilo t study, the task at hand is one of applying solid sets of scale

PAGE 81

68 point descriptors to the resu lting construct indicators. In this section, the scale measurements to be used in the main study are presented. It should be noted that the specific item indicators displayed for each cons truct might end up being slightly different for the final measurement due to the results of the pilot study. For this dissertation, the item indicators presented are those that have b een identified by other researchers in past research endeavors. In contrast, the actua l scale point descriptors presented here are either new or modified from those measurement descriptors used by the original researchers of the constructs. All of the scales in th e main study are designed as subjective self-reporting scales completed by th e FSSP participants incl uded in the study. The use of self-report scales is theoretically sound because many of the consequences of job stress are psychological in nature and involve attitudes and emotions known only by the person surveyed (Spector & Jex, 1998). This section presents an overview of the sc ales used in this study. Ten new scales (i.e., attitude toward the j ob, emotional intelligence, emo tional labor, emotion-focused coping, intention to switch pos itions within the organiza tion, intention to leave the organization, overall job satisfaction, perceived customer demands, physical consequences, and role overload) consisting of 13 dimension are presented first. Six modified scales (i.e., role ambiguity, role c onflict, job autonomy, social service encounter (job stress), emotional exhaustion, and job pe rformance) consisting of 13 dimensions are presented in the order that they are found in the model. The scale measurement for organizational commitment is presented last and is taken from the existing literature.

PAGE 82

69 New Scale Development The following line items for each scale are those which initially went into the pilot study; however, line items might be different from the final survey line items. Therefore, the line items that appear be low may change during the actual scale purification process. Attitude-Toward-the-Job Attitude-toward-the-job is operationalized as a unidimensional composite construct. Attitude-toward-the-job is conceptualized as a composite of learned predispositions toward reacting positively or negatively toward a particular job. An eight-item semantic differential scale for the purpose of measuri ng attitude-toward-thejob is developed based on the work of Nels on and Bowen (2000). It is expected that higher scores are representative of a more favorable attitude toward the job. In the following scale, represents items that must be reversed for data analysis.

PAGE 83

70 Attitude Toward the Job Scale In this section, there are a number of incomplete opinion statements that I would like you to complete by circling any number from to 7 that best indicates your opinion or feeling Please notice that each listed statement has it own unique set of descriptors for completing that statement. Please circle only ONE answer for each statement. 1. Maintaining a positive attitude toward customers is Very Easy 1 2 3 4 5 6 7 Very Difficult 2. Being enthusiastic about my job is . Very Easy 1 2 3 4 5 6 7 Very Difficult 3. My job tends to be . Very Unpleasant 1 2 3 4 5 6 7 Very Pleasant 4. My job is usually . .(R) Very Interesting 1 2 3 4 5 6 7 Extremely Boring 5. Feedback from my superiors is . Definitely What I Expected 1 2 3 4 5 6 7 Definitely Not What I Expected 6. Overall, communications with my boss are . Definitely Less Than What I Expected 1 2 3 4 5 6 7 Definitely More Than What I Expected 7. Overall working conditions are . .(R) Very Healthy 1 2 3 4 5 6 7 Very Unhealthy 8. Overall I see my job as Very Emotionally Draining 1 2 3 4 5 6 7 Very Emotional ly U p liftin g Emotional Intelligence Upon examination of several published emo tional intelligence (EI) measures and based on our conceptualization of the construct, it was determined that EI could be operationalized as a multi-dimensional c onstruct consisting of two dimensions (recognition of emotions and regulation of emotions). Recognition of emotions is defined as perceiving and identifying emoti ons (i.e., the ability to recognize how you and those around you are feeling). Regulation of emotions is defined as managing emotions or the ability which allows you to manage em otions in yourself and in others (Mayer, Salovey, & Caruso. 2000).

PAGE 84

71 In the current study, EI was examined as an ability-based construct. However, most of the available EI ability-based s cales such as the Multifactor Emotional Intelligence Scale (MEIS) and the Mayer, Sa lovey, and Caruso Emotional Intelligence Test (MSCEIT) (2000) were lengthy and require a licen sed psychologist to supervise their use and analysis. Scales deve loped by Akers and Porter (2003), although operationalized as ability-based, have no repor ted reliability or validity. Based on the constraints presented above, a subjective se lf-report scale was deve loped to measure an ability-based emotional intelligence based on the works of several researchers including Mayer and Salovey (1993) and Mayer, Salovey, and Caruso (2000). The EI scale consists of 19 semantic differential scale it ems. It was expected that higher scores represent a higher level of emotional intellig ence. In the following scale, represents items that must be reversed for data analysis.

PAGE 85

72 Emotional Intelligence Perception Scale In this section, there are a number of incomplete opinion statements that I would like you to complete by circling any number from to 7 that best indicates your opinion or feeling Please notice that each listed statement has its own unique set of descriptors for completing that statement. Please circle only ONE answer for each statement. 1. Recognizing emotions that I experience in a particular situation is . Extremely Easy 1 2 3 4 5 6 7 Extremely Difficult 2. My ability to figure out the reasons behind my different emotions is . Extremely Low 1 2 3 4 5 6 7 Extremely High 3. Differentiating between emotions I experience is . Extremely Difficult 1 2 3 4 5 6 7 Extremely Easy 4. I think about the emotions behind my actions . Definitely After the Action 1 2 3 4 5 6 7 Definitely Before the Action 5. When it comes to how my feelings are affecting me, I am . Totally Aware 1 2 3 4 5 6 7 Completely Oblivious 6. When it comes to other peoples feelings at work, acknowledging their feelings is . Very Difficult to Do 1 2 3 4 5 6 7 Very Easy to Do 7. Examining the feelings, thoughts, and actions of others is . .(R) Very Easy to Do 1 2 3 4 5 6 7 Very Difficult to Do 8. Ones ability to understand why other people feel the way they do is . Extremely Useful 1 2 3 4 5 6 7 Completely Useless 9. Observing how others react to me helps me better understand my own behavior . Definitely Disagree 1 2 3 4 5 6 7 Definitely Agree Emotional Intelligence Management Scale 1. When I get frustrated or angry, considering my options is . Extremely Easy 1 2 3 4 5 6 7 Extremely Difficult 2. Generally when I feel angry, I am . Completely Out-ofControl 1 2 3 4 5 6 7 Completely Under Control 3. On most things I try, I expect to . Completely Fail 1 2 3 4 5 6 7 Totally Succeed 4. When taking on challenges where there is a strong chance that I may fail, I feel . Totally Energized 1 2 3 4 5 6 7 Completely Defeated 5. My ability to control my emotions is . Very Ineffective 1 2 3 4 5 6 7 Very Effective 6. I seek out activities that make me . Very Cheerful 1 2 3 4 5 6 7 Very Sad 7. For me, arranging events others enjoy is A Complete Bother 1 2 3 4 5 6 7 Absolute Joy 8. Helping other people feel better when they are down is Of the Utmost Importance 1 2 3 4 5 6 7 A Complete Waste of Time

PAGE 86

73 9. My handling of conflict is . Very Strong 1 2 3 4 5 6 7 Very Weak 10. When it comes to getting involved in other peoples problems, I am Very Reluctant 1 2 3 4 5 6 7 Always Eager Emotional Labor Emotional labor has previously been examined through two different foci (i.e., job and emotion). Job focus examines the frequency of interaction with customers and/or job expectations to express certain emoti ons (e.g., Wharton and Erickson, 1993). An emotion focus examines the amount of effort required to perform organizationally imposed emotionally-based behaviors (Broth eridge and Lee, 2003; Hochschild, 1983). In the current study, emotional labor was operationalized as a multidimensional construct and measured as the perceived effort involved in acting out emotionally-based behavior display rules (Steinberg, 1999) and the perception of FSSP as to the frequency with which he or she believes that displays of emotionally-based be haviors are necessary. It is expected that the mo re often FSSP display emotionally -based behavior, the greater the degree of emotional labor present. Ther efore, the frequency dimension was measured with six scale items asking FSSP how often he or she must display an imposed emotionally-based behavior based on the Brotheridge and Lee (2003) study.

PAGE 87

74 Display of Emotional Labor Scale For each of the following behavior statements, please check the box that best describes the extent to which you are required or expected to perform the behavior during a social service encounter Please mark only ONE response for each statement. During any service encounter . . Never Rarely Occasionally Regularly A Great Deal of the Time Almost Always Always 1. . I am required to hold back expressing my true feelings 1 2 3 4 5 6 7 2. . I am required to pretend to have emotions that I dont really have. 1 2 3 4 5 6 7 3. . I am required to hide my true feelings about a situation. 1 2 3 4 5 6 7 4. . I am expected to make an effort to actually feel the emotions that management believes I should feel. 1 2 3 4 5 6 7 5. . I am expected to try to actually experience the emotions that I must show on the job. 1 2 3 4 5 6 7 6. . I am expected to really try to feel the emotions I have to show as part of my job. 1 2 3 4 5 6 7 Emotional effort is consistent with em otional labor theory that draws heavily on Goffmans (1959, 1969) dramaturgical persp ective. Goffman (1959, 1969) argues that labor is viewed as an actor performing on stag e for a frequently discriminating audience. Various degrees of acting require various am ounts of effort. For example, FSSP surface acting carefully regulates the situation or appr aisal that precedes the emotion and present verbal and non-verbal cues such as fa cial expression, gestur es, and voice tone (Hochschild, 1983). Deep acting is an attempt to actually experience or feel the imposed emotionally-based behavior. The individua l may conform to an imposed emotionallybased behavioral-display rule by simulating the emotion that is not actually felt with less effort than a FSSP involved in deep acting (H ochschild, 1983). The current measure of

PAGE 88

75 emotional effort is based on the Kruml and Geddes (2000) scale development study. The measure is a five item composite scale of surface and deep acting. Higher scores are expected to show a greater level of emotional labor. Emotional Labor Effort Scale For each statement below, circle the number that best approximates the amount of effort it takes you to perform the behavior des cribed in that statement You may choose any number between 1 and . The higher the number, the more the effort required. For example, circling a would mean the stated behavior would take no effort on your part and a 7 would mean it would take you an extreme amount of effort to do the behavior. Please circle only ONE answer for each statement. No Effort Extreme Amount of Effort 1. Fake the emotions I show customers. 1 2 3 4 5 6 7 2. Talk myself out of feeling what I really feel when helping customers. 1 2 3 4 5 6 7 3. Summon up feelings I need to show to customers. 1 2 3 4 5 6 7 4. Change my actual feelings to match those that I must express to customers. 1 2 3 4 5 6 7 5. Attempt to create certain emotions in myself that present the image my organization desires. 1 2 3 4 5 6 7 Emotion-Focused Coping Sources of job stress may not be as critical as are their magnitude vis--vis an individuals coping ability (J ackson et al., 1986). Widespread acceptance of transactional theory of coping (Lazarus, 1966; Lazarus & Folkman, 1984) views coping as a dynamic process. Unlike trait measures, measure of transactional coping allows for situationspecific effects during stages of a stressful ev ent. The emotion-focus coping subscale of

PAGE 89

76 the Occupational Stress Indication (OSI) Vers ion 2 is a subjective self-report scale developed to measure transactional coping that is used across a wide range of occupations and for which its theoretical development is fully reported (Cooper, Mallinger, & Kahn, 1988; Robertson & C ooper, 1990). The assumption of equal applicability of scale items across situati ons made the comparison of scale scores reasonable. Dewe and ODriscoll (2002) examined this subscale consisting of four items and reported a .65 coefficient alpha. Ther efore, a new eight-item multidimensional emotion-focused coping scale was developed for this study (i.e., 4-item extent of use and 4-item effectiveness). This scale asked FSSP to indicate the extent to which he or she actually used the strategy and their belief as to the effectiveness of each strategy in reducing job stress. Scores on the Extent of Use subscale reflect a behavioral assessment, not an attitudinal level, and are indicative of thoughts or ac tions used to deal with a problem (Stone, Greenberg, Kennedy-Moore, & Newman, 1991). High scores were expected to reflect higher le vels of emotion-focused copi ng. The effectiveness subscale was also developed based on the OSI emoti on-focused coping subscale and asked FSSP if he or she perceived that th e strategies were effective in coping with job stress. Higher scores on the Effectiveness s ubscale were expected to re flect the strength of the FSSPs belief in the effectiveness of these strategies in reducing job stress.

PAGE 90

77 Use of Emotion-Focused Coping For each of the following activities, please check the box that best describes how often, during a normal workweek, do you engage in that activity to help you cope with job stress Please check only ONE response for each activity. Never Rarely Occasionally Regularly A Great Deal of the Time Almost Always All The Time 1. Turn to hobbies/pastimes. 1 2 3 4 5 6 7 2. Talk to understanding friends. 1 2 3 4 5 6 7 3. Expand interests/activities outside of work. 1 2 3 4 5 6 7 4. Seek social support. 1 2 3 4 5 6 7 Emotion-Focused Coping Effectiveness Scale For each activity listed below, circle the number that best reflects to what extent you believe that activity or method either helps or hinders yo ur ability to cope with job stress Please circle only ONE answer for each activity. Definitely Hinders Hinders Somewhat Hinders Neither Hinders nor helps Somewhat Helps Helps Definitely Helps 1. Resort to hobbies/pastimes. 1 2 3 4 5 6 7 2. Talk to understanding friends. 1 2 3 4 5 6 7 3. Expand interests/activities outside of work. 1 2 3 4 5 6 7 4. Seek social support. 1 2 3 4 5 6 7 Intention to Switch Positions Within the Organization No usable scales were found to measure intention to switch positions within an organization. Intention to switch positions within an organization was conceptualized and measured as a subjective judgment of the FSSPs likelihood to switch positions

PAGE 91

78 within the firm within a designated timef rame. The intention to switch scale was developed specifically for this study based on relevant research on intentions such as intention to leave (e.g., Sager & Menon, 1994). Scale items are carefully developed to avoid redundancy and drifting away from the or iginal conceptual defi nition of intentions (Drolet & Morrison, 2001). Intention to sw itch positions within the organization is a subjective measure as to the likelihood that FSSP will switch positions within the organization in a 12 month period. It is expe cted that higher scores represent greater intention to switch positions within the organi zation. In the following scale, represents items that must be reversed for data analysis. Intention to Switch Positions Within the Organization Scale For each of the following behavioral scenarios please circle the number that best reflects how likely or unlikely you would do that described behavior in the next 12 months Please circle only ONE answer for each behavioral scenario. Very Unlikely Unlikely Somewhat Unlikely Neither Unlikely nor Likely Somewhat Likel y Likely Very Likely 1. I will actively look for another position within this organization. 1 2 3 4 5 6 7 2. I plan to switch positions within this organization. 1 2 3 4 5 6 7 3. I plan to keep my current position within the organization. 1 2 3 4 5 6 7 Intention to Leave As discussed in previous chapters, resear ch has shown that intention to leave a job plays a central role in turnover models (J ohnston & Futrell, 1989; Lee & Mowday, 1987; Sager et al., 1989; Sager & Menon, 1994). The conceptualizati on that inten tion to leave is generally viewed as the employees intent to end his or her current employment is

PAGE 92

79 congruent with the Donnelly, Jr. and Ivancevichs (1975) and the Singh, Verbeke, and Rhoads (1996) conceptualization of inten tion to leave. The FSSPs likelihood to leave the organization was obtained by simply aski ng the FSSP to make a subjective judgment on their likelihood of leaving the organiza tion within a designated timeframe (12 months). Three scale items were developed specifically for this study based on the research of Singh, Verbeke, and Rhoads (1996) It was expected that higher scores represent a greater intention to leave the organizations. In th e following scale, represents items that must be reversed for data analysis. Intention to Leave the Organization Scale For each of the following behavioral scenarios please circle the number that best reflects how likely or unlikely you would do that described behavior in the next 12 months Please circle only ONE answer for each behavioral scenario. Very Unlikely Unlikely Somewhat Unlikely Neither Unlikely nor Likely Somewhat Likel y Likely Very Likely 1. In the next 12 months, I will actively look for a job with another company. 1 2 3 4 5 6 7 2. In the next 12 months, I plan to switch companies. 1 2 3 4 5 6 7 3. In the next 12 months, I plan to work for my current employer. 1 2 3 4 5 6 7 Overall Job Satisfaction In order to measure overall job satisf action, Rice, Gentile and McFarlins (1991) measure ( = .83) of global job satisfaction (i.e., ove rall job satisfaction) revised from the Quinn and Shepard (1974) scale was used as a foundation to develop a semantic differential scale of overall job satisfaction for this study. The measure examines the

PAGE 93

80 FSSPs general affective reaction to his or her job without re ferencing any specific facet of job satisfaction. The Rice, Gentile, and McFarlin (1991) scale was used in a study by Lam, Baum, and Pine (2001) for a sample of 171 Chinese restaurant managerial employees in which the researchers report an al pha of .77. It is expected that the higher the score, the greater the le vel of job satisfaction. Ba sed on the Rice, Gentile, and McFarlin (1991) study, a subjec tive self-report measure consisting of six semantic differential scale items was developed to meas ure overall job satisfaction. Higher scores are expected to reveal higher le vels of overall job satisfaction. In the following scale, represents items that must be reversed for data analysis. Overall Job Satisfaction Scale In this section, there are a number of incomplete opinion statements that I would like you to complete by circling any number from to 7 that best indicates your opinion or feeling Please notice that each listed statement has it own unique set of descriptors for completing that statement. Please circle only ONE answer for each statement. 1. If you had to decide all over again, knowing what you know now, would you take the job if it were offered to you? . Not A Chance 1 2 3 4 5 6 7 Absolutely 2. If a friend should apply for a job like yours with your employer, how likely are you to recommend the job to him or her? . Very Likely 1 2 3 4 5 6 7 Not Likely At All 3. Compared to your ideal job, your current job is . Very Different Than Expected 1 2 3 4 5 6 7 Exactly As Expected 4. To what extent does your current job match your expectations when you took it? Definitely Worse than I Expected 1 2 3 4 5 6 7 Definitely Better Than I Expected 5. What is your overall satisfaction/dissatisfaction with your current job? Very Satisfied 1 2 3 4 5 6 7 Very Dissatisfied 6. Your overall feeling about your job would be Definitely Hate It 1 2 3 4 5 6 7 Definitely Love It

PAGE 94

81 Perceived Customer Demands The perceived customer demand scale used in this study was developed specifically for this study. The perception of customer demands was previously examined by Li and Calantone (1998) and Wang and Netemeyer (2002) as customer demandingness with scales revised from the Wheel wright and Clark (1992) study. The focus of the current study is the FSSPs per ception of the intensity of the customers demands compared to the FSSPs perception of managements expectations. Therefore, a subjective self-report semantic differential s cale containing six item s based on the Li and Calantone (1998) and the Wang and Netemeye r (2002) study was developed to measure perceived customer demands. Higher scores ar e expected to represent higher levels of perceived customer demands. In the following scale, represents items that must be reversed for data analysis. Perceived Customer Demand Scale For each of the following statements, please circle the number that best expre sses your perception of customer demands. Please circle only ONE answer for each statement. Statements Not at all Demanding Extremely Demanding a. In terms of service, the customers I serve are 1 2 3 4 5 6 7 b. In terms of quality, the customers I serve are 1 2 3 4 5 6 7 c. In terms of reliability, the customers I serve are 1 2 3 4 5 6 7 d. Customers expectations for service are 1 2 3 4 5 6 7 e. Customers expectations that the services offe red will meet is or her needs are 1 2 3 4 5 6 7 f. Customers expectations for delivery level of service quality are 1 2 3 4 5 6 7

PAGE 95

82 Physical Consequences A subjective self-report P hysical Symptom Inventory ( PSI) of FSSPs health is the most popular method of measuring physic al consequences (e.g., headache or upset stomach) of job stress that cannot be measured directly (Spector & Jex, 1998). The scale assesses physical health symptoms related to or aff ecting the body (separate from the mind) believed by stress researchers to be related to psychological distress. The PSI was derived from the Spector (1987) physical sy mptoms scale. Participants respond to during the past 30 days did you have? with (No, Yes, but I did not see doctor, or Yes and I saw doctor). The PSI is a causal i ndictor scale; therefore, coefficient alpha is not relevant. A list of 18 physical symptoms from the Spector (1987) PSI was used to develop a subjective self-report frequency meas ure of physical consequences of job stress for FSSP. Higher scores on the physical conseq uence scale are expected to show a higher occurrence of negative physical consequences.

PAGE 96

83 Physical Consequences Scale For each of the following listed physical symptoms, please check the box that best describes how often you experience that symptom when yo u are feeling stress from your job Please mark only ONE response for each physical symptom. Never Rarely Occasionally Regularly A Good Deal of the Time Almost Always Always 1. A backache 1 2 3 4 5 6 7 2. A skin rash 1 2 3 4 5 6 7 3. Acid indigestion or heartburn 1 2 3 4 5 6 7 4. An infection 1 2 3 4 5 6 7 5. An upset stomach or nausea 1 2 3 4 5 6 7 6. Chest pain 1 2 3 4 5 6 7 7. Constipation 1 2 3 4 5 6 7 8. Diarrhea 1 2 3 4 5 6 7 9. Dizziness 1 2 3 4 5 6 7 10. Eye strain 1 2 3 4 5 6 7 11. Fever 1 2 3 4 5 6 7 12. Headache 1 2 3 4 5 6 7 13. Heart pounding when not exercising 1 2 3 4 5 6 7 14. Loss of appetite 1 2 3 4 5 6 15. Shortness of breath 1 2 3 4 5 6 7 16. Stomach cramps (not menstrual) 1 2 3 4 5 6 7 17. Tiredness or fatigue 1 2 3 4 5 6 7 18. Trouble sleeping 1 2 3 4 5 6 7 Role Overload Role overload was operationalized as a unidimensional construct and was measured with a scale developed based on the Spector and Jexs (1998) Quantitative Workload Inventory (QWI). The QWI was derived from the Spector (1987) and the Spector, Dwyer, and Jex (1988) workload scales and presented a coefficient alpha of .82

PAGE 97

84 in the Spector and Jex (1998) study. Role overload is conceptualized in terms of quantitative overload (i.e., how much work there is) and not qualitatively (i.e., work difficulty). It has been established that FSSP in social service organizations perform many standard duties that although varying in diffi cultly, are routine in nature. It has also been established that FSSP often have a greater volume of work than can be completed in a normal workday (Bridges & Lauer, 2003). The QWI (1998) is a five-item five-point scale that lacks uniform descriptors that do not lend well to multivariate analysis. Keeping job context in mind, this study develo ped a five-item interval scale to measure quantitative work overload establishing consiste ncy in the descriptors (Hair, Jr., Bush, & Ortinau, 2003). The scale descriptors for th e current scale reflect times per day. Higher scores are expected to represent highe r levels of role overload (Spector & Jex, 1998). Role Overload Scale For each of the following work situations please check the box that best approximates how often in a typical workday each event occurs Please check only ONE answer for each work situation. Never 0% Rarely 1-20% Occasionally 21-40% Regularly 4160% A Great Deal of the Time 5180% Almost Always 81-99% Always 100% 1. My job requires me to work very fast. 1 2 3 4 5 6 7 2. My job requires me to work very hard. 1 2 3 4 5 6 7 3. My job leaves me with little time to get things done. 1 2 3 4 5 6 7 4. There is a great deal of work to be done. 1 2 3 4 5 6 7 5. I have to do more work than I can do well. 1 2 3 4 5 6 7

PAGE 98

85 Redesigned Scales Role Ambiguity Singh and Rhoads (1991) developed a multifaceted multidimensional role ambiguity subjective self-report instru ment known as the MULTIRAM (company flexibility .70, work .84, promotion .75; boss support .86, demand .86; customer interaction .78, objection .81, presentation .81; ethical external .90, internal .83; other managers, coworkers, family range from .87 to .88) scale to measure role ambiguity of boundary spanning employees. Dimensions of the MULTIRAM were found to be distinct with intercorrelations between .49 and .69. Convergent validity was found in that the dimensions of role ambiguity in the MULTIRAM had large, positive, and significant correlations with the Rizzo, House, & Lirtzman (1970) role ambiguity scale. Th e MULTIRAM scale was used extensively in boundary spanning context by researchers such as Johlke and Dunhan (2000 and 2001) as well as Singh (2000). Singh (2000) examined role ambiguity relative to performance, quality, and burnout of frontline employees in service organizations by using two subscales ( company with an of .83 and .81 and customer with an of .87 and .91 over two samples within the study) of the (1991) MU LITRAM scale. In the current study, large bureaucratic organizations where FSSP have many bosses, the boss sub-scale may confound the results as FSSP would have to de termine which boss the scale is referring to. In the social service environment, customers play a key role in producing the service offering. Therefore, it is important to understand the customers contribution in the creation of role ambiguity and job stress. As suggested by Singh and Rhoads (1991), when the purpose of the study is to focus on ro le ambiguities internal to the organization,

PAGE 99

86 it is more preferable to be context specific, not use all dimensions of role ambiguity, and utilize only relevant dimensions usin g the MULTIRAM scale. Therefore, twodimensions of role ambiguity were meas ured with two-subscales of the MULTIRAM scale. In order to examine a ll constructs at the same intensity and to increase the chance of uncovering greater variability, the eight item customer MULTIRAM subscale and the nine item company MULTRIAM subscale were measured with seven-points instead of the original five-points. It is expected that higher scores represent higher levels of ambiguity. Role Ambiguity Scale For each of the following statements, please circle the number that best expresses the level of certainty or uncertainty you have concerning the following activity. Please circle only ONE answer for each statement. Very Certain Certain Somewhat Certain Neither Certain nor Uncertain Somewhat Uncertain Uncertain Very Uncertain 1. How I am expected to interact with my customers. 7 6 5 4 3 2 1 2. How much service I should provide my customers is . 7 6 5 4 3 2 1 3. How I should behave (with customers) while on the job is . 7 6 5 4 3 2 1 4. How I am expected to handle my customers objections is . 7 6 5 4 3 2 1 5. How I am expected to handle unusual problems and situations is . 7 6 5 4 3 2 1 6. How I am expected to deal with customers criticism is . 7 6 5 4 3 2 1 7. Which specific company streng ths I should present to customers is . 7 6 5 4 3 2 1 8. Which specific service benefits I am expected to highlight for customers is . 7 6 5 4 3 2 1 9. The actions required in meeting customer needs are . 7 6 5 4 3 2 1 10. How I am expected to handle non-routine activities on the job is . 7 6 5 4 3 2 1

PAGE 100

87 11. The amount of work I am expected to do is . 7 6 5 4 3 2 1 12. Which tasks I should give priority is . 7 6 5 4 3 2 1 13. How much work I am expected to do is . 7 6 5 4 3 2 1 14. How I should handle my free time on the job is . 7 6 5 4 3 2 1 15. What I can do to get promoted is . 7 6 5 4 3 2 1 16. How vulnerable to job termination I am is . 7 6 5 4 3 2 1 17. What the critical factor is in getting promoted is . 7 6 5 4 3 2 1 Role Conflict Role conflict was examined as a multidimensional construct (i.e., intrapersonal, interpersonal, and person-role). These dime nsions were originally labeled intrarole, interrole, and intersender c onflict by Rizzo, House, a nd Lirtzman (RHL) (1970). Although some researchers question the cont ent validity of the original RHL citing susceptibility to wording bias and factor structure (King & Kin g, 1990; Smith, Tisak, & Schmieder, 1993), the scale is used extensiv ely by other researchers such as Jex and Elacqua (1999), Netemeyer, Burton, and J ohnston (1995), Schaubroeck, Ganster, Sime, and Ditman (1993) with coefficient alpha valu es range from .71 to .87. The current study utilized the original RHL s ubjective self-report (1970) role conflict scal e. For the purpose of measurement consistency, the five -point RHL intersender conflict (3-item), intrasender conflict (2-item), and person-role conflict (3-item) sub-scales were modified to seven points with endpoints of 1 = str ongly disagree and 7 = strongly agree. One additional item was developed for the intrasende r conflict subscale. It is expected that higher scores on each of the subscales represent higher levels of the dimension being assessed.

PAGE 101

88 Person-Role Conflict Scale For each statement listed below, circle the number that best expresses the extent to which you either agree or disagree with that statement Please circle only ONE answer for each statement. Strongly Disagree Disagree Slightly Disagree Neither Agree Nor Disagree Slightly Agree Agree Strongly Agree 1. I have to do things that I believe should be done differently. 1 2 3 4 5 6 7 2. I have to break company policy rules in order to carry out an assignment. 1 2 3 4 5 6 7 3. I work on unnecessary things. 1 2 3 4 5 6 7 Intersender Role Conflict Scale For each statement listed below, circle the number that best expresses the extent to which you either agree or disagree with that statement Please circle only ONE answer for each statement. Strongly Disagree Disagree Slightly Disagree Neither Agree Nor Disagree Slightly Agree Agree Strongly Agree 4. I receive incompatible requests from two or more people. 1 2 3 4 5 6 7 5. I do things that are apt to be accepted by one person and not accepted by others. 1 2 3 4 5 6 7 6. I work with two or more groups who operate quite differently. 1 2 3 4 5 6 7 Intrasender Role Conflict Scale For each statement listed below, circle the number that best expresses the extent to which you either agree or disagree with that statement Please circle only ONE answer for each statement. Strongly Disagree Disagree Slightly Disagree Neither Agree Nor Disagree Slightly Agree Agree Strongly Agree 7. I receive assignments without the manpower to complete them. 1 2 3 4 5 6 7 8. I receive assignments without adequate resources and material to execute them. 1 2 3 4 5 6 7 9. I receive assignments for which I am not adequately trained. 1 2 3 4 5 6 7

PAGE 102

89 Job Autonomy Job autonomy was operationalized as a unidimensional construct and was measured with a three-item subscale from th e revised Job Diagnostic Survey (JDS) (Idask & Drasgow, 1987). The JDS was originally developed by Hackman and Oldham (1975) in conjunction with the theory of job characteristics and is one of the most widely used and revised scales to measure the nature of jobs. For example, across several studies, the estimated portion of true va riance accounted for by the autonomy subscale of the original JDS is 63% (e.g., Munz, Huelsman, Konol d, & McKinney, 1996, Steel & Rentsch, 1997). The job autonomy subscale of the JDS is well-tested (e.g., Wang & Netemeyer, 2002 = .77). However, Idaszak and Drasgow (1987) analyzed and revised the JDS by removing reversal items and significantly improved meas urement properties (e.g., coefficient scores on all subscales including j ob autonomy above .80 as opposed to .77). Spector, Jex, and Chen (1995) measured job autonomy using the Idaszak and Dras gow (1987) revised subscale of the JDS and reported an alpha of .85. Idaszak and Drasgow (1987) believe that the phrasing of the original reversed it ems may be responsible for artifact factors. The current study modifies the Idaszak and Drasgow (1987) job autonomy subscale scale descriptors from agree/disagree to frequenc y. A high score on the job autonomy scale suggests the presence of a hi gher level of job autonomy.

PAGE 103

90 Job Autonomy Modified Job Diagnostic Survey Scale For each of the following be havior statements, please check the box that best describes the extent to which your job allows you to do the behavior described in that statement Please mark only ONE answer for each statement. Never Rarely Occasionally Regularly A Great Deal of the Time Almost Always Always 1. . decide how to go about doing my work. 1 2 3 4 5 6 7 2. . use my personal initiative or judgment in carrying out the work. 1 2 3 4 5 6 7 3. . have the opportunity for independence and freedom in how I do my work. 1 2 3 4 5 6 7 Social Service Encounter (Job) Stress Social service encounter stress was op erationalized as a multidimensional (time stress and anxiety) construct. The Jamal and Baba (1992) nine-item five-point strongly agree/strongly disagree scale was adapted fr om the Parker and Decotiis (1983) 13-item scale to measure job stress. In the Jamal a nd Baba (1992) study, alpha for the nine-items was .83. For purposes of our study, nine items from the Jamal and Baba (1992) plus one additional item from the Parker and Decotiis (1983) scale ( = .74), are believed to accurately demonstrate both time pressure a nd anxiety of job stress for FSSP. The additional item from the Parker and Decotiss (1983) study is I frequently get the feeling I am married to the company. One scale item is modified to read Too many frontline social service personnel get burned out by job demands to replace nurses in the original scale item. For consistency in analysis, a seven-item agree/disagree scale was used. Higher scores are expected to reveal higher levels of social service encounter (job) stress

PAGE 104

91 of FSSP. In the following scale, represen ts items that must be reversed for data analysis. Job Stress Scale For each of the listed statements below, please circle the number that best expresses the extent to which you either agree or disagree with that statement Please circle only ONE answer for each statement. Strongly Disagree Disagree Slightly Disagree Neither Agree Nor Slightly A g ree Agree Strongly Agree 1. I often feel fidgety or nervousness as a result of my job. 1 2 3 4 5 6 7 2. My job irritates me more than it should. 1 2 3 4 5 6 7 3. On the job, there are lots of times when my job drives me right up a wall. 1 2 3 4 5 6 7 4. Sometimes when I think about my job, I get a tight feeling in my chest. 1 2 3 4 5 6 7 5. I do not feel guilty when I take time off from the job. 1 2 3 4 5 6 7 6. I have too much work to do and too little time to do it in. 1 2 3 4 5 6 7 7. Very few frontline social service personnel in my company get burned out because of job demands. 1 2 3 4 5 6 7 8. I sometimes dread the telephone ringing at home because the call might be job-related. 1 2 3 4 5 6 7 9. I feel like I never have a day off. 1 2 3 4 5 6 7 10. I frequently get the feeling I am married to the company. 1 2 3 4 5 6 7 Emotional Exhaustion Emotional exhaustion was measured using a modified version of the Maslach and Jackson (1981) frequency of emotional exhaustio n scale. This scale is used extensively (e.g., sales Babakus et al., 1999 and human services Houkes et al., 2003). Cronbach coefficient alphas for frequency were reporte d as .89 and .86 for intensity. Higher scores suggest higher levels of emotional exhaus tion. The scale consis ts of a nine-item

PAGE 105

92 frequency and a nine-item intensity scale th at asks respondents to evaluate how often they feel exhausted from their work, as well as the perceived intens ity of the emotional exhaustion. The Maslach and Jackson (1981) scale was modified by changing item descriptors from A few times a year, monthly, a few times a month, every week, a few times a week, everyday to reflect a range c onsisting of equal values (e.g., 1 day a week to 7 days a week). The scale offered an option for individuals who believe that emotional exhaustion is not a factor (i.e., never) for them It was expected that higher scores show that the more often FSSP report feeling emoti onally exhausted from their job, the greater the condition of emotional exha ustion. In the following intensity of emotional exhaustion scale, represents items that must be reversed for data analysis. Emotional Exhaustion Scale Frequency of Emotional Exhaustion For each of the following situations, circle the number that best reflects how many days a week you experience that feeling Please circle only ONE answer for each situation. Never 1 Day a Week 2 Days a Wee k 3 Days a Week 4 Days a Week 5 Days a Week 6 Days a Week 7 Days a Week 1. I feel emotionally drained from my work. 0 1 2 3 4 5 6 7 2. I feel used up at the end of the workday. 0 1 2 3 4 5 6 7 3. When I get up in the morning to face another day on the job I feel tired. 0 1 2 3 4 5 6 7 4. Working with people all day is really a strain for me. 0 1 2 3 4 5 6 7 5. I feel burned out from my work. 0 1 2 3 4 5 6 7 6. I feel frustrated with my job. 0 1 2 3 4 5 6 7 7. I feel I am working too hard on my job. 0 1 2 3 4 5 6 7 8. Working directly with people really puts a strain on me. 0 1 2 3 4 5 6 7 9. I feel like Im at the end of my rope with my job. 0 1 2 3 4 5 6 7

PAGE 106

93 Intensity of Emotional Exhaustion For each of the following listed questions, please circle the number from to that best describes the intensity level of your feelings regarding that question If you feel a question does not pertain to you, please circle the N/A [not applicable] response category. Please circle only ONE response for each listed question. 1. How emotionally draining is your work? N/A Very Mild Drained 1 2 3 4 5 6 7 Extremely Drained 2. How do you feel at the end of the workday? N/A Exhausted 1 2 3 4 5 6 7 Very Energized 3. How do you feel in the morning knowing you have to face another day on the job? N/A Very Excited 1 2 3 4 5 6 7 Not Excited At All 4. How do you feel after working with people all day? N/A Very Energized 1 2 3 4 5 6 7 Worn Out 5. How burned out do you feel from your work? N/A Extremely Burned Out 1 2 3 4 5 6 7 Somewhat Burned Out 6. How frustrated are you on the job? N/A Very Little Frustrated 1 2 3 4 5 6 7 Extremely Frustrated 7. How hard do you feel you must work on the job? N/A Not Very 1 2 3 4 5 6 7 Extremely Hard 8. How does working directly with people make you feel? N/A Very Stressed 1 2 3 4 5 6 7 Not At All Stressed 9. How does your job make you feel? N/A Very Energized 1 2 3 4 5 6 7 Exhausted Job Performance Job performance is often measured on tw o separate dimensions (i.e., performance quality and performance productivity). Perfor mance quality is unique to service delivery and, therefore, performance quality is measured in this study. Performance quality is the process of interacting with customers and is thought of as emotionally laborious. Because performance quality is intangible, it is difficult for management to quantify increases or decreases in perf ormance quality. Changes are less likely to be visible to management than would be a decr ease in performance productivity.

PAGE 107

94 Churchill, Ford, Hartley, and Walker, Jr., (1985), report that subjective selfreported performance measures are less restrict ive in range and have less error than several other purportedly object ive measures. Schneider, Ganster, Sime, and Ditman (1996) accumulated a large amount of empiri cal evidence suggesting that self-report by employees of performance have validity and are significantly correlated with judgments made by observers external to the organization (e.g., customers) (p. 697), especially if self-reports ar e obtained anonymously. The multidimensional (i.e., building trust = .89, promptness = .89, reliability = .77, and individualize attention = .93) PERFQ scale developed by Singh (2000) to examine job performance for boundary spanning employees was used to measure soci al service encounter performance. PERFQ (2000) was modified by changes to scale desc riptors and set up. The original PERFQ (2000) asked respondents to compare his or he rself relative to other employees in the same organization with desc riptors ranging from 1 = botto m 20% to 7 = top 5%. The modified PERFQ scale involves four subscales (i.e., building customer trust, promptness, reliability, individualized atte ntion) asking FSSP to reflect on his or her job performance with descriptors ranging from 1= truly terrible to 7= truly exceptional. It is expected that higher scores reflect a higher level of job performance.

PAGE 108

95 Building Customer Trust Scale For each of the following job related activities listed below, please circle the number that best expresses your judgment of how well you normally perform that activity Please circle only ONE response for e ach a ctivity. Truly Terrible Poor Fair Average Good Excellent Outstanding 1 Taking the initiative to help your customers even when it is not part of your responsibility. 1 2 3 4 5 6 7 2. Taking the time to help your customers at the expense of not meeting daily productivity goals. 1 2 3 4 5 6 7 3 Developing customer trust/ confidence in your service provided. 1 2 3 4 5 6 7 Promptness Scale 4. Responding promptly to customer requests, despite your busy workload. 1 2 3 4 5 6 7 5. Following up on promises make to your customers. 1 2 3 4 5 6 7 6. Overall, consistently providing prompt service to all of your customers. 1 2 3 4 5 6 7 Reliability Scale 7. Consistently resolving customer concerns the first time. 1 2 3 4 5 6 7 8. Consistently demonstrating emotionally-based behavior deem ed appropriate by the organization. 1 2 3 4 5 6 7 9. Providing accurate or correct information to the customer. 1 2 3 4 5 6 7 10. Even though it is not your responsibility, making sure ot her departments follow through with your customers requests. 1 2 3 4 5 6 7 11. Telling the customer the straight facts rather telling them what they want to hear. 1 2 3 4 5 6 7 12. Performing your job dependably/ accurately. 1 2 3 4 5 6 7 Individualized Attention Scale 13. Servicing the account with the customers best interest in mind. 1 2 3 4 5 6 7 14. Listening attentively to identify and understand the conce rns of customers. 1 2 3 4 5 6 7 15. Working out solutions to each customers questions or concerns. 1 2 3 4 5 6 7 16. Overall, providing individualized attention to each custom ers concerns. 1 2 3 4 5 6 7

PAGE 109

96 Overall Organizational Commitment There are many scales developed to m easure organizational commitment based on varying definitions of organizational commitment. For the current study, o rganizational commitment is defined as the FSSPs perception relating to the relative strength of identification with, as well as his/her dedica tion to, the organization. The focus of this study is on attitudinal or affective commitment (i.e., emotional attachment to, identification with, and involvement in the organization) and not normative (i.e., pressures on an employee to remain with an organization resulting from socialization in an organization), nor from continuance commitment (i.e., associated with the perceived costs of leaving an organization). Theref ore, the shortened version of the Mowday, Steers, and Porter (1979) Or ganizational Commitment Scale (OCQ) was used in this study to measure overall organizational commit ment. The shortened version of the OCQ is used in a variety of work settings as evidenced in the Cohen (1993) meta-analysis of organizational commitment with coefficien t alphas ranging from .74 to .92. As such, overall organizational commitment was opera tionalized with the OCQ nine-item sevenpoint Likert-type scale. In the following scale, represents items that must be reversed for data analysis.

PAGE 110

97 Overall Organizational Commitment Scale For each statement listed below, please circle the number that best expresses the extent to which you either agree or disagree with that statement Please circle only ONE answer for each statement Strongly Disagree Disagree Slightly Disagree Neither Agree Nor Disagree Slightly Agree Agree Strongly Agree 1. I am willing to put in a great deal of effort beyond that normally expected in order to help this organization be successful. 1 2 3 4 5 6 7 2. I talk up this organization to my friends as a great organization to work for. 1 2 3 4 5 6 7 3. I would accept almost any t ype of job assignment in order to keep working for this organization. 1 2 3 4 5 6 7 4. I find that my values are very different than the organizations values. 1 2 3 4 5 6 7 5. I am embarrassed to tell others that I am part of this organization. 1 2 3 4 5 6 7 6. This organization really inspires the very best in me in the way of job performance. 1 2 3 4 5 6 7 7. I am extremely glad that I chose this organization to work for over others I was considering at the time I joined. 1 2 3 4 5 6 7 8. I do not care about the fate of this organization. 1 2 3 4 5 6 7 9. For me, this is the best of all possible organization for which to work. 1 2 3 4 5 6 7 Data Analysis Final Study Relationships between the latent vari ables themselves and the amount of unexplained variance make up the structural model. Descriptive data statistics, correlations, and covariance matrix were developed using SPSS 13 th Edition. Model parameters of the structural model were estimated using LISREL 8.3 employing a covariance matrix to determine that the hypothesized relationships between latent variables and between latent va riables and the manifest indicators are consistent with the empirical data. In order to determine the in fluence of one variable on another, regression

PAGE 111

98 analysis and simple correlation analysis of hypothesized interrelationships was conducted (Hair, Jr., Anderson, Tatham, & Black, 1998). Reliabilities and dimensionality of scales were examined using exploratory and confirmatory factor analysis. The model was evaluated using various f it statistics. Alt hough Chi-square [H 0 : = (0)] is a test of perfect fit and of a perfect model, it is considered too sensitive when large sample sizes, such as those found in this study, are present (n> 200) and cannot be considered alone (Hair, Jr. et al., 1998). Othe r indices such as root mean square error of approximation (RMSEA) were analyzed. RM SEA is considered one of the most informative fit indices showing how well th e model with unknown but optimally chosen parameter values would fit the population cova riance matrix if it we re available (Hair, Jr. et al., 1998, p. 85). Another appropriate index employed in this study is the Expected Cross-Validation Index (ECVI). This index focuses on overall error (i.e., discrepancy between population covariance matrix and model fitted to the sample). The ECVI measures overall model fit by assessing whethe r there is discrepancy between the fitted covariance matrix for the analyzed sample and the expected covariance matrix that would be obtained in another sample of equal si ze. The goodness-of-fit (G FI) index was also examined. GFI shows how closely the m odel comes to perfectly reproducing the observed covariance matrix and is a good indica tor of the relevant amount of variances and covariances accounted for by the model. The non-normed fit index (NNFI) and the comparative fit index (CFI) were also relied upon to assess the fit allowing for a relative comparison of the proposed model and the null m odel. It is apparent from the discussion above that various indicators of fit have weaknesses and strengths that depend on sample size, estimation procedure, model complexity, violation of the underlying

PAGE 112

99 assumptions of multivariate normality, and variable independence, or any combination thereof (Diamantopoulos & Siguaw, 2000). Th erefore, multiple indi cators are relied on to evaluate the model. Because sample size plays an important ro le in model testing, power analysis was conducted to determine the validity of results and statistical power of the analysis (i.e., the probability of correctly rejecting an in correct model) (Byrne 1998). Large samples raise the question as to whether a statistically significant chi-square estimates is accurate, if serious specification error are present, or high power exists. For the purpose of determining power associated with the mode l, the approach in which RMSEA, sample size, and degrees of freedom are examined was used. A table of power for various sample sizes is found in MacCallum et al. (1996) study and was utilized as a guideline. The following table (3.1) pr esents hypothesized rela tionship and measurement analysis methodology used in this study. Table 3.1 Hypothesized Relationships Testable Hypotheses Methodology H 1 : Role ambiguity (RA) has a significant positive impact in creating FSSP job stress (JS). Regression Analysis H 2 : Role conflict (RC) has a significant positive impact in creating FSSP job stress (JS). Regression Analysis H 3 : Role overload has a significant po sitive impact in creating FSSP job stress (JS). Regression Analysis H 4 : Job autonomy (JA) mediates the relationship between perceived customer demands (PCD) and job stress (JS) such that job autonomy (JA) reduces the impact of perceived customer demands (PCD) on job stress (JS). Hierarchal Regression Analysis H 5 : Emotional intelligence (EI) significantly mediates the relationship between emotional labor (EL) and jo b stress (JS) such that it reduces the impact of emotional labor (EL) on job stress (JS). Hierarchal Regression Analysis H 6 : The stronger the pres ence of job autonomy (JA), the less likely FSSP will engage in emotion-focused coping (EFC). One-Way ANOVA H 7a : There is a significant relationship between emotional intelligence (EI) and emotion-focused coping effectiveness (EFCE). Regression Analysis

PAGE 113

100 H 7b : There is a significant relationship between emotional intelligence (EI) and emotion-focused coping frequency (EFCF). Regression Analysis H 8a : Emotion-focused coping effectiveness (EFCE) significantly mediates the positive relationship between job stress (JS) and emotional exhaustion (EE) such that it reduces the impact of job stress (JS). Hierarchal Regression Analysis H 8b : Emotion-focused coping frequency (EFCF) significantly mediates the positive relationship between job stress (JS) and emotional exhaustion (EE) such that it reduces the impact of job stress (JS). Hierarchal Regression Analysis H 9a : Emotion-focused coping effectiveness (EFCE) significantly mediates the negative relationship between job stress (JS) and attitude toward the job (ATJ) such that it reduces the impact of job stress (JS). Hierarchal Regression Analysis H 9b : Emotion-focused coping frequency (EFCF) significantly mediates the negative relationship between job stress (JS) and attitude toward the job (ATJ) such that it reduces the impact of job stress (JS). Hierarchal Regression Analysis H 10a : Emotion-focused coping effectiveness (EFCE) significantly mediates the negative relationship between job stress (JS) and overall job satisfaction (OJS) such that it redu ces the impact of job stress (JS). Hierarchal Regression Analysis H 10b : Emotion-focused coping frequency (EFCF) significantly mediates the negative relationship between job stress (JS) and overall job satisfaction (OJS) such that it redu ces the impact of job stress (JS). Hierarchal Regression Analysis H 11a : Emotion-focused coping effectiveness (EFCE) significantly mediates the negative relationship between job stress (JS) and overall organizational commitment (OOC) such that it reduces the impact of job stress (JS). Hierarchal Regression Analysis H 11b : Emotion-focused coping frequency (EFCF) significantly mediates the negative relationship between job stress (JS) and overall organizational commitment (OOC) such that it reduces the impact of job stress (JS). Hierarchal Regression Analysis H 12a : Emotion-focused coping effectiveness (EFCE) significantly mediates the positive relationship between job stress (JS) and physical consequences (PC) such that it re duces the impact of job stress (JS). Hierarchal Regression Analysis H 12b : Emotion-focused coping frequency (EFCF) significantly mediates the positive relationship between job stress (JS) and physical consequences (PC) such that it re duces the impact of job stress (JS). Hierarchal Regression Analysis HO 13 : Attitude toward the job (ATJ) is no t significantly correlated with other outcome variables in the model. Correlation Analysis HO 13a : Attitude toward the job (ATJ) is not significantly correlated with overall organizational commitment (OOC). Correlation Analysis HO 13b : Attitude toward the job (ATJ) is no t significantly correlated with overall job satisfaction (OJS). Correlation Analysis HO 13c : Attitude toward the job (ATJ) is not significantly correlated with physical consequences (PC). Correlation Analysis HO 13c : Attitude toward the job (ATJ) is no t significantly correlated with emotional exhaustion (EE). Correlation Analysis HO 14: Physical consequences are not signi ficantly correlated with outcome variable (OV) of overall organi zational commitment (OOC), overall job satisfaction (OJS), or emotional exhaustion (EE). Correlation Analysis

PAGE 114

101 HO 14a: Physical consequences (PC) are not significantly correlated with overall organizational commitment (OOC). Correlation Analysis HO 14b : Physical consequences (PC) are not significantly correlated with overall job satisfaction (OJS). Correlation Analysis HO 14c : Physical consequences (PC) are not significantly correlated with emotional exhaustion (EE). Correlation Analysis H 15: Overall organizational commitment (OOC), overall job satisfaction (OJS), and emotional exhaustion (E E) are significantly correlated with each other. Correlation Analysis H 15a : Overall organizational commitment (OOC) is significantly correlated with overall job satisfaction (OJS). Correlation Analysis H 15b : Overall organizational commitment (OOC) is significantly correlated with emotional exhaustion (EE). Correlation Analysis H 15c : Overall job satisfaction (OJS) is significantly co rrelated with emotional exhaustion (EE). Correlation Analysis H 16 : Attitude toward the job (ATJ) has a significant positive influence on FSSPs job performance (JP). Regression Analysis H 17 : Overall job satisfaction (OJS) has a significant positive influence on job performance (JP). Regression Analysis H 18: Overall organizational commitment (OOC) has a significant positive influence on job performance (JP). Regression Analysis H 19 : Emotional exhaustion (EE) has a significant negative influence on job performance (JP). Regression Analysis H 20 : Physical consequences (PC) have a significant negative influence on job performance (JP). Regression Analysis H 21 : Job performance (JP) has a significant negative influence on intention to switch positions within the organization (ITS). Regression Analysis H 22 : Intention to switch positions (ITS) within an organization and intention to leave an organi zation (ITL) are correlated. Correlation Analysis H 23 : Job performance (JP) has a significant negative influence on intention to leave the organization (ITL). Regression Analysis Mediation Testing This study examines the relationship betw een the predictor (X) and the criterion (Y) variables, the relationship between the predictor (X) and the mediator (M) variables, and the relationship between the mediator (M) and the criterion (Y) using regression analysis. To test these relati onships, regression analysis was used to: 1) estimate and test for correlation through the path from the predic tor variable to the criterion by treating the mediator as though it does not exist (X Y); 2) using the medi ator as the criterion variable, correlation is examined by estimati ng and testing the path from the predictor

PAGE 115

102 variable to the mediator (X M); 3) to show that the me diator affects the criterion variable, estimate and test for correlation be tween mediator and the criterion variable (M Y); and 4) controlling for the mediator, the effect of (X Y) should be zero. After controlling for the relationship between the mediator and criterion variables, the relationship between the predic tor and criterion should, in th e case of total mediation, be reduced to zero. In the case of partial medi ation, the path between the predictor variable and the criterion variable is reduced in absolute size, but is still different from zero when the mediator is controlled. Chapter Summary Chapter Three connects the c onceptualization and function of the constructs in the model with the operationalization. The chapte r accomplished this by addressing: 1) the research setting, 2) providing a detailed explanation of procedures for securing study participants, 3) addressing non-response bias 4) examining construct development, 5) discussing the foundations and direction for new scale development measures, 6) laying out the data analysis plan a nd justification, 7) presenting a visualization of the proposed data analysis for hypothesized relationships in table format, and 8) discussing mediation testing necessary for correlati on hypothesized in the model. In Chapter Four, the analysis and results of the studies deve loped in the first three chapters of this research are presented. Chapter Four includes the analysis and reports the results of both the pilot study and the final study.

PAGE 116

103 CHAPTER 4 Analysis and Results This chapter presents the findings from the interdisciplinary research framework used to examine the studys six major objectives: 1) To empirically determine the roles of emotional intelligence and emotional labor within self-management of emotionally-based display behavior in the social se rvice delivery encounter from the frontline social service personnels (FSSP) perspective. 2) To empirically examine the infl uence of two neglected constructs (i.e., perceived customer demands an d emotional labor) as sources of job stress. 3) To re-examine role ambiguity, role conflict, and role overload as sources of job stress. 4) To empirically examine the mediating effect of job autonomy, emotion-focused coping, and emotional intelligence.

PAGE 117

104 5) To empirically investigate the in fluence of five selected outcome constructs on job performance. To empirically examine the relationship of social service encounter performance on both intention to switch positions within the organization and intentions to leave the organization. The current study utilizes a three-phase research framework. The first phase focuses on conducting eight (8) in-depth in terviews with frontline social service personnel (FSSP) and management personnel to assess the relevanc y of the proposed manifested indicators representi ng the constructs in a social service environment. Next, a pilot study using a direct cognitive framework is administered to 361 FSSP to determine to what extent each indicator is related to its respective construct. In the last phase, a survey is administered to a large sa mple of 2,500 FSSP to test the hypothesized relationships in the model. This chapter re ports the analyses and findings used in the development of the constructs, purification of the scale measurements, fitness of the CFA structural models, and hypotheses test results. Construct Development In-Depth Interviews In-depth interviews were conducted with one male and three female frontline social service personnel (FSSP) and one male and three female managers of FSSP for the

PAGE 118

105 purpose of obtaining preliminary insights to the nature and scope of the proposed constructs of the model. The transcripts of the audiotapes were content analyzed for the meaning of words used by the respondents, th e frequency and intensity of the comments, and the relationship of a response to an em otion. Preliminary interpretation of the responses suggests initial suppor t of the importance of the proposed models constructs. One of the most significant findings is that all respondents emphasized that perceived customer demands and emotional labor are re levant sources of j ob stress. Regarding perceived customer demands, one male res pondent states that, T he children are the clients and they are the main source of job stress. One female responding to: What is your greatest source of job stress? indirec tly addresses emotiona l labor, through her response I mean you are dealing with abused kids and they may act out their stuff on you and the automatic reaction is to be mad at them, but you cant do that because you are basically recreating their past abuse. We are expected to be very empathic as opposed to being mad back. So, yeah that is ve ry tough. As expected, all respondents acknowledge the importance of role overload. For instance, one respondent stated, You can work 60 hours a week and you cant get ahead of the work. Overall, these types of verbal expressions and emotions provide preliminary confirmation of the constructs to be further developed in the pilot study phase. Pilot Study Cognitive Response Survey The purpose of the pilot study was to gain clearer insights and understanding into which potential specific indicators were mo st relevant in representing the respective constructs. The following findings were based on frontline social service personnel

PAGE 119

106 (FSSP) expressing the extent to which he or she believes the proposed indicators represent the respective construct. Overall Response Rate Using a drop-off delivery method, three hundred and sixty-one (361) surveys were hand delivered to an organizational representative from each of the five participating organizations for distribution to FSSP at a weekly staff meeting. Organizational representatives sent follow-up reminders through the organizations interoffice email two weeks after the initial di stribution of the surve y, resulting in a final response rate of 24%. Table 4.1 below pr esents response rates by organizations. Participating organizations included a very small (i.e., employing five FSSP) privately owned organization o ffering social services to the local community, two small (i.e., employing 11 FSSP and 65 FSSP, respectively) branches of larg er national social services organizations, and the social service departments of two large (i.e., employing >100 FSSP) government-operated organizations. All of these organizations provide services to individuals, as well as families in crisis. Table 4.1 Response Rates by Organization Organization Surveys Delivered Completed Surveys Returned Response Rate (%) #1 183 30 16 #2 11 6 55 #3 100 30 30 #4 62 18 29 #5 5 2 40 Total 361 86 24

PAGE 120

107 Descriptive Profile of the Sampled Respondents Of the 86 respondents (23.8% male and 76.2% female), the age range most frequently reported was 45-54 years with an average age of 40.4. Overall, these respondents are highly educated and experienced in the soci al service field. Over 90 percent of the respondents (94%) hold at le ast a four-year college degree, with 44% holding masters degrees. The average hours of job training through their current employer was 84.6. The average tenure time wi th their current employer was about three (3) years. The average tenure time in the social service field was 9.02 years with 52.4% ranging between three and nine years. S ee Table 4.2 below for a complete profile breakdown. Table 4.2 Sample Demographics (N=86) Age of Respondents (%) Years in the Social Service Field (%) (%) < 19 0.0 >1 3.6 10-11 2.4 20-29 21.4 1 2.4 12-15 10.7 30-34 10.7 2 8.3 16-20 10.7 35-44 25.0 3-5 27.4 > 20 9.5 45-54 27.4 6-9 25.0 > 55 15.5 Mean Age = 40.4 Mean Years in Field = 9.02 Education Level (%) Years of Employment in Current Position (%) (%) Post Grad 0.0 < 1 29.8 10-11 1.2 Masters 44 1 16.7 12-15 3.6 College Grad 50 2 25.0 16-20 0.0 Some College / Technical School 4.8 3-5 17.7 > 20 2.4 High School Grad 1.2 6-9 3.6 Some High School 0.0 Mean Value = 2.96

PAGE 121

108 Gender (%) Number of Hours of Job-Related Training with Current Employer (%) (%) Male 23.8 < 10 25 61-70 1.2 Female 76.2 11-20 6.0 71-80 6.0 21-30 9.5 81-90 0.0 31-40 11.9 91-100 14.3 41-50 8.3 > 100 15.4 51-60 2.4 Mean Value = 84.6 Pilot Study Results Construct Indicators In addition to the preset list of possible manifested indicators, respondents had the opportunity to make additional comments concerning each of the constructs under investigation. About twelve pe rcent (12%) of respondents offe red at least one comment. Overall, respondents commen ts strongly support the relevancy for including the constructs indicators. For example, rega rding receiving assignments without adequate resources and materials to complete them (indicator #8) of the ro le conflict scale, one respondent wrote, Big, Big Fact or after the statement. The same respondent also put a plus mark after the scale descriptor 7 = De finitely a Factor. Her response of ditto after receiving assignment without being adeq uately trained (indicator #9), can be interpreted as an indication that this item is also a Big, Big factor for her. Most of the written comments serve to reinforce that the proposed indicators stir deep emotions for respondents. One responde nt wrote, In a book called Turning Stone the author says that child welfare is like looking a rabid dog in eye and saying nice puppy until you can find a shot gun.

PAGE 122

109 Indicator Purification Given the pilot studys primary objective of assessing which of the 138 proposed manifested indicators best represent the dom ain of each of the models seventeen (17) proposed constructs, the resulting data structures were s ubjected to an iterative principal component analysis with dire ct oblimin rotation, using SPSS 13.0. A combination of skewness analysis and an item-to-total correlation analysis were utilized to determine the relevancy of each indicators contribution to th e proposed construct. Given the fact that a manifested indicators communality represents the percent of variance that indicator accounts for in the retained factor, it was expected that any indicator heavily loading on a factor will, therefore, display a large co mmunality. As such, examining manifested indicators communalities becomes an important component in the decision process of retaining or removing an indicator. The se t of decision rules guiding the elimination or retention of a proposed indicat or are presented below: 1. Positively skewed indicators Decision: An indicator with strongly positive skewness indicates that th e respondents do not see it as a factor relating to the construct and the indica tor should be removed. 2. If the indicators communality is <. 50 (Hair, Jr. et al., 1998) Decision: Remove that indicator.

PAGE 123

110 3. If an indicators rotated factor loading on a given factor is <. 40 or it cross-loads with any other factor > .40 (Hatcher, 1994) Decision: Examine that indicators item-to-tot al correlation prio r to a removal decision. 4. If the item-to-total correlations with other indicators are < .35 (i.e., low) (Nunnally & Bernstein, 1994) Decision: Remove the indicator. Using these known and acceptable decision criteria, the following constructs required manifested indicator re ductions: attitude towa rd the job (from 8 to 7 indicators), emotional intelligence (19 to 15 indicators), emotional labor display (6 to 5 indicators), intention to switch positions within an orga nization (3 to 2 indicat ors), job performance 16 to 7 indicators), job stress (10 to 9 indi cators), overall job satisfaction (6 to 4 indicators), physical consequences (18 to 10 indicators), and role ambiguity (17 to 11 indicators). The i ndicators for all 17 proposed cons tructs did not display strongly positive skewness. The following examples represent some of the manifested indicator retention/removal results from the pilot study: Upon completion of the iterative fa ctoring process and item-to-total analysis of the proposed 19-item emotional intelligence construct,

PAGE 124

111 four (4) indicators were removed. Indicator EIM10 did not load at >.40 and was removed. In the second iteration, EIM8 demonstrated low communality (.272) and was removed. In addition, EIM1 did not factor load at > .40 and was removed. In the third iteration, EIM7 had a low item-to-total correla tion (.311) and low communality (.215) and was removed. The remaining 15 indicators for the proposed emotional intelligence construct met all the decision criteria and were retained. Examinations of the proposed 16-item job performance construct resulted in a decision to remove ni ne (9) indicators. Indicator JP10 had a low communality of .272; therefore, it was removed from further analysis. In the second it eration, communalities were low for JP1 (.490) and JP2 (.413) and both in dicators were eliminated from further analysis. In the third iteration, indicator JP8 had low communality of .440 and was removed. In the fourth iteration, JP9 did not factor load at > .40 and was removed. In continuing the iteration process JP4, JP6, JP7, and JP16 were also removed leaving a final solution that is absent of significant cr oss loadings. All the remaining indicators were acceptable for inclusion.

PAGE 125

112 Eight (8) of the proposed 18-item physical consequence construct indicators were removed. Indicators PC1 and PC11 had low communality of .413 and .424, PC2 and PC4 had indicator loadings < .40, and PC10 cross loads (.431 and -.521) and was removed. In continuing the iteration process, PC5, PC9, and PC17 were removed leaving a final solution that was abse nt of significant cross loadings. All of the remaining indicator s were acceptable for inclusion. Examination of the proposed role am biguity construct using the same data reduction/purification procedures, results in six (6) of the initial seventeen (17) indicators (i.e ., RA6, RA 10, RA 14, RA 15, RA 16, and RA17) being removed from any further analysis. Table 4.3 below presents the pilot studys key statistics on th e final solution of retained indicators used in developing the various scale measurements for each proposed construct. See Appendices C (Inter-Item Co rrelations), D (Item-to-Total Correlation), and E (Pilot Study Descriptive), for comp lete detailed results of the data reduction/construct purification procedures used in the pilo t study. Also, see Appendix F for a summary of the actual retained manifest ed indicators and thei r individual indicator loadings. Items removed are presented as bold and italicized in Appendix F.

PAGE 126

113 Table 4.3 Pilot Study Key Statistical Results (N=86) Construct Composite Mean Std. Dev. Std. Mean Std. Dev. Construct Alpha Initial Items Retained Items Perceived Customer Demands 35.43 6.71 5.91 1.12 .939 6 6 Role Ambiguity 49.59 15.44 4.51 1.40 .937 17 11 Role Conflict 42.85 14.30 4.76 1.589 .932 9 9 Role Overload 28.45 6.17 5.69 1.234 .860 5 5 Emotional Labor Display 16.09 5.15 5.50 1.72 .885 6 5 Emotional Labor Effort 17.05 8.06 3.41 1.60 .881 5 5 Job Autonomy 16.50 5.15 5.50 1.72 .907 3 3 Emotional Intelligence 92.43 12.36 5.78 0.78 .902 19 15 Job Stress 41.33 1471 4.59 1.63 .917 10 9 Emotion-Focused Coping 21.73 5.47 5.43 1.37 .815 4 4 Physical Consequences 38.15 16.38 3.82 1.638 .931 18 10 Attitude Toward the Job 40.65 6.72 5.81 .961 .858 8 7 Overall Job Satisfaction 22.57 4.61 5.64 1.15 .862 6 4 Job Performance 54.38 9.107 6.08 1.04 .920 16 7 Intention to Switch Positions 8.88 4.17 4.44 2.09 .919 3 2 Intention to Leave 13.14 5.75 4.38 1.92 .829 3 3 Data Analysis The purpose of the final study was to colle ct the necessary data to empirically assess the underlying research hypotheses. Therefore, after identifying manifested indicators for each construct and prior to admi nistering the final survey, appropriate sets

PAGE 127

114 of scale point descriptors were applied to each indicator and scale setups and instructions for filling out and returning the instrument were applied. For assurance of clarity and understanding of the survey instrument, a group of 25 frontline social service personnel (FSSP) exam ined the survey instruments physical appearance, (i.e., professionalism), read ability, and ease of understanding of the indicators, instructions, and setups. No problems were id entified as to the physical appearance of the instrument. Some changes believed to improve cl arity and readability of the indicators resulted in modifications such as: expectations for service are changed to customers expectations that the services offered will meet his or her needs are. Overall Response Rate A mailing of the final survey instrument was sent on February 16, 2005 to 2,500 frontline social service personnel (FSSP). The sample consisted of 2,391 FSSP randomly selected from the National Association of Social Workers Florida Chapters 6,583 members and 109 taken from a census of a social services department of a local branch of a national organization. Two weeks after the initial mailings, a post-card reminder to non-respondents followed. The overall respons e rate was 24.4% (i .e., 611 returned surveys) of which 30 came back marked unde liverable by the U.S. postal service and 48 respondents indicated that they did not meet the criteria of currently active in the social service field, resulting in a usable response rate of 21.3%.

PAGE 128

Table 4.4 Response Rates by Organization Organization Surveys Delivered Surveys Returned Overall Response Rate Usable Surveys Response Rate NASW-FL 2,391 571 23.8 493 20.6 #2 109 40 36.7 40 36.7 Total 2,500 611 24.4 533 21.3 Report of Non-Respondents Fifty (50) non-respondents of the previous 1,899 non-respondents received a short survey instrument containing two key constructs and key demographics approximately two weeks after the return date for the original survey. Responses from previous non-respondents were compared to responses of known respondents using one-way analysis of variance (ANOVA) to determine if differences in responses for critical variables between the two groups (respondents / non-respondents) were statistically significant (Lambert & Harrington, 1990). The one-way ANOVAs demonstrated that the mean values of comparison variables (p< 0.10) were not statistically different, implying a low likelihood of non-response bias (Table 4.5). Table 4.5 Non-Respondent Data Constructs T-value Mean Difference Sig. (2-Tailed) Emotional Labor Display -1.368 -2333 .230 Job Autonomy .452 .667 .670 Selected Demographics Age -.237 -1.667 .822 Education 2.000 .667 .102 Gender -2.236 -.500 .076 115

PAGE 129

116 Descriptive Profile of the Sampled Respondents Females made up eighty-six percent (86.3 %) of the respondents. The age range of respondents most frequently reported wa s 55+ years with an average age of 45.3. Respondents reported current employment as 52% Social Worker 27% Case Manager, and 21% other social service positions About ninety-five percent (95.3%) of the respondents held at least a four-year colle ge degree, with 59.1% holding masters, and 27.2% holding postgraduate degrees. Over 60% of the respondents experienced continuous employment with hi s or her current employer for a minimum of 3-5 years. Only 3.2% of the respondents spent less than one year with his or her current employer. The average hours of job training through current employers was 61.3. The average tenure time for respondents in the social service field wa s 13.3 years with 33.6% with 20+ years tenure. See Table 4.6 be low for a complete profile breakdown. Table 4.6 Sample Demographics (N=533) Age of Respondents (%) Years in the Social Service Field (%) (%) < 19 0.0 >1 3.2 10-11 8.4 20-29 12.2 1 3.0 12-15 13.7 30-34 12.6 2 3.8 16-20 8.6 35-44 19.6 3-5 13.3 > 20 33.6 45-54 25.0 6-9 12.2 > 55 30.0 Mean Age = 45.3 Mean Years in Field = 13.3 Education Level (%) Years of Employment in Current Position (%) (%) Post Grad 27.2 < 1 18.0 10-11 5.1 Masters 59.1 1 10.1 12-15 6.2 College Grad 9.0 2 11.8 16-20 3.4 Some College / Technical School 4.5 3-5 26.8 > 20 6.0 High School Grad 0.2 6-9 12.6 Some High School 0.0 Mean Value = 5.7

PAGE 130

117 Gender (%) Number of Hours of Job-Related Training with Current Employer (%) (%) Male 13.7 < 10 13.1 61-70 2.8 Female 86.3 11-20 11.3 71-80 3.2 21-30 7.7 81-90 2.4 31-40 8.3 91-100 3.0 41-50 7.3 > 100 37.1 51-60 3.8 Mean Value = 61.3 Note: Data not provided by 2 respondents for demographics Scale Measurement Purification Exploratory Factor Analyses Results Data from a random sample of 260 of the 533 respondents was subjected to exploratory factor analysis (EFA) to asse ss the measurement properties of the latent variables used in this study. Estimation of factor loadings, items to retain, and underlying structure of items were determined using SPSS 13.1 Principal Components Analysis with an Oblim rotation (Hair, Jr. et al., 1998) Data for this study met the methodological assumptions for performing an EFA (Hatch er, 1994). However, indicators for the Intention to Switch Positions within th e Organization construct violated the interpretability criteria for EFA of a minimum of three indicators. Therefore, Intention to Switch Positions within the Organization was not subjected to explorat ory factor analysis and only scale reliability was reported in Tabl e 4.7 below. The following set of decision rules functioned as a guide for remova l or retention of a proposed indicator:

PAGE 131

118 1. A substantial amount of anti-ima ging correlations >.30 may indicate that there are no true underlying f actors Decision: Do not subject data to an EFA and report only scale reliability. 2. Interpretation of rotated factor pattern: Based on the sample size of 260, any manifested indicator ha ving a factor loading of > .35 which does not cross-load with any other factor at > .35 (i.e., demonstrating a simple structure of high loadi ngs on one factor and near-zero loadings on all other factors) shou ld be retained after determining that the indicators communality is above acceptable levels. Otherwise, remove indicator (Hatcher, 1994). 3. If the indicators communality is <.50 Decision: Remove indicator (Hair, Jr. et al., 1998). 4. If less than two-indicators load on any one factor, remove the indicator and resubmit the data to EFA for further analysis. Based on these known and acceptable decision criteria, the following constructs required manifested indicator reductions: perc eived customer demands (6 to 4), role

PAGE 132

119 ambiguity (11 to 10), role conflic t (9 to 6), emotional intelligen ce (15 to 7), job stress (9 to 6), physical consequences (10 to 8), attitu de toward the job (7 to 6), organizational consequences (9 to 6), emotional exhausti on frequency (9 to 7), emotional exhaustion intensity (9 to 4), and job performance (7 to 5). The following examples represent some of the manifested indicator retenti on/removal decisions from the EFA: Anti-imaging correlations of the job autonomy construct demonstrated that three out of th ree indicators correlate at >.30. Therefore, job autonomy was not submitted to EFA and only scale reliability was reported in Table 4.7 below. Examination of the nine-item emotional exhaustion intensity constructs resulted in the removal of indicators EEID and EEIH for violating the retention communality criterion. Indicator EEIF cross loaded on two factors (.457/.489) and was removed. During the second factor analysis iteration, EEIA cross loaded at >.35 on two factors and was removed. Examina tion of a third iteration revealed that EEIG (.167) was in violation of the communality criterion and was removed. The four remaining indicators load on one factor with loadings > .756 and were retained. Through a series of iterative factor analyses of the 15 manifested indicators proposed to make up the emotional intelligence construct,

PAGE 133

120 two indicators failed to meet the required communality level (EID .368 and EIE .425) and were removed. Indicator EIO cross loaded >.35 on two factors and was removed. During the second iteration, one indicator was removed due to cross loading (EIH .644/.540) across two factors. In iterati on #3, three items violating the communality criterion (EIJ .441, EIN .481, and EIG .337) were removed. From the fourth and fift h iterations (EIO .365) and (EIL .392) were removed due to violations of the communality criterion. The seven (7) remaining indicators loaded on three separate factors with indicator loadings ranging from -.619 to .885 and met all criteria for retention. Three iterative factor analyses of the proposed nine-item role conflict construct resulted in the elimin ation of RCB, RCC, RCE, and RCF based on violations of communalities (.392, .462, .417, .391, respectively). RCI was remove d based on cross loading at > .35 (.789 and .429). The remaining four (4) items loaded on one factor with indicator loadings from .587 to .738 and, therefore, were retained.

PAGE 134

121 Table 4.7 below presents the EFA key statisti cs on the final solutions of retained indicators for each proposed construct. For the EFA factor loadings, see Appendix G. For other detailed EFA statistics, see Appendix H.

PAGE 135

122 Table 4.7 EFA Key Statis tical Results (N=260) Constructs Composite Mean Std. Dev. Mean Std. Dev. Alpha Pilot Study Items EFA Items Perceived Customer Demands 20.63 5.11 5.16 1.28 .904 6 4 Role Ambiguity 16.84 6.05 1.87 0.67 .862 11 10 Role Conflict 14.82 6.22 3.71 1.55 .828 9 4 Role Overload 2127 6.57 4.25 1.31 .887 5 5 Emotional Labor Display 11.44 5.09 2.29 1.02 .788 5 5 Emotional Labor Effort 17.79 7.48 3.56 1.50 .859 5 5 Job Autonomy 16.24 3.97 5.51 1.32 .906 3 3 Emotional Intelligence 43.52 5.90 5.44 .738 .700 15 7 Job Stress 18.18 7.82 3.03 1.30 .807 9 6 Emotion-Focused Coping Frequency 16.58 4.67 4.15 1.67 .803 4 4 Emotion-Focused Coping Effectiveness 23.98 3.24 6.00 .81 .781 4 4 Physical Consequences 12.92 4.42 1.62 .55 .802 10 8 Attitude Toward the Job 29.15 6.37 4.86 1.06 .737 7 6 Organizational Commitment 31.87 7.31 5.31 1.22 .868 9 6 Overall Job Satisfaction 19.77 4.94 4.94 1.24 .842 4 4 Emotional Exhaustion Frequency 13.01 5.07 2.17 .85 .810 9 7 Emotional Exhaustion Intensity 13.508 5.01 3.40 1.25 .796 9 4 Job Performance 29.68 3.35 5.94 .67 .851 7 5 Intention to Switch 5.38 3.72 2.69 1.86 .881 2 2 Intention to Leave 8.58 5.57 2.86 1.86 .874 3 3

PAGE 136

123 Measurement Model Difficulties may arise in obtaining an acceptable fit of a measurement model when the model includes >30 indicators as greater number of indicators may often result in large chi-square values (Bentler & Choa, 1987). Therefore, for the purpose of examining the proposed measurement model consisting of 102 proposed indicators, confirmatory factor analysis (CFA) was pe rformed on six (6) smaller models chosen based on the constructs relationship to ot her constructs in the model using SAS 9.1. Each model demonstrates how the latent variables are operationalized by their corresponding measured va riables, as well as assesses the validity and reliability of the measures (Hair, Jr. et al., 1998; Diamantopoulos & Siguaw, 2000). Model Fit Indices The criteria used to identify indicators for retention through confirmatory factor analysis (CFA) included: Absolute fit indices: If chi squa re is greater than two times the degrees of freedom, then reject the hypothesis of perfect fit and examine incremental fit indices (Hatcher, 1994). Closeness of fit indices: For larg e samples (>200) and/or real world data, RMSEA is considered one of the most informative fit indices (Diamantopoulos & Siguaw, 2000; Hatcher, 1994) [<.05 good fit,

PAGE 137

124 between .05 and .08 reasonable fit, between .08 and .10 mediocre fit and >.10 poor fit]. Decision: Accept < .10, reject >.10. When incremental fit indices fall outside acceptable ranges (CFI > .90, NNFI > .90) examine standard error values (Hair, Jr. et al., 1998). For t-values with zero or near zero standard errors, remove indicator. Where indicator loadings are non-si gnificant or indicator loadings <.35, remove indicator (Hatcher, 1994). When acceptable standard errors a nd significant indicator loadings are present, but fit indices are not in an acceptable range, examine LaGrange modification index of 2 value reduction for indicator removal selection (Hatcher. 1994).

PAGE 138

125 Examine the model for parsimonious fit. The closer to zero the Akaik Information Criterion (AIK) falls, the more parsimonious the model. Accept the model with the lowest AIK that also meets other fit indices requirements (Hair, Jr. et al., 1998).

PAGE 139

126 See Table 4.8 Below for key CFA statistical results: Table 4.8 CFA Key Statistical Results (N=260) Constructs Composite Mean Std. Dev. Mean Std. Dev. Alpha EFA Items CFA Items Perceived Customer Demands 20.55 4.84 5.15 1.28 0.895 4 4 Role Ambiguity 18.622 6.571 1.87 0.677 0.869 10 10 Role Conflict 14.578 6.127 3.65 1.33 0.828 4 4 Role Overload 21.280 6.480 4.25 1.31 0.887 5 5 Emotional Labor 28.857 9.335 2.92 1.02 0.781 10 10 Job Autonomy 16.471 3.873 5.42 1.32 0.906 3 3 Emotional Intelligence 38.460 5.717 5.49 0.768 0.766 7 7 Job Stress 17.848 7.622 3.03 1.304 0.802 6 6 Emotion-Focused Coping Effectiveness 17.931 2.645 6.02 0.851 0.766 4 3 Emotion-Focused Coping Frequency 12.377 3.640 4.17 1.220 0.802 4 3 Physical Consequences 13.212 4.887 1.62 0.533 0.745 8 8 Attitude Toward the Job 29.188 6.415 4.85 1.060 0.775 6 6 Organizational Commitment 26.728 5.791 5.37 1.191 0.821 6 5 Overall Job Satisfaction 19.775 5.057 4.93 1.234 0.861 4 4 Emotional Exhaustion 25.200 8.147 2.78 .889 .838 11 9 Job Performance 29.694 3.262 5.93 0.667 0.841 5 5 Intention to Switch 5.450 3.690 2.85 1.855 0.856 2 2 Intention to Leave 8.574 5.600 2.68 1.855 0.890 3 3

PAGE 140

127 Model Analysis Confirmatory factor analysis (CFA) was performed on data from a separate sample of 260 respondents randomly selected from a sample of 533 FSSP in the final study. Findings of the seven (7) smaller more manageable models that make up the overall model are as follows: Model 1 Proposed Sources of Job Stress consists of six constructs proposed as sources of job stress: role am biguity (3-dimensional), role co nflict (unidimensional), role overload (unidimensional), per ceived customer demands (uni dimensional), and emotional labor (display 2-dimensional/effort uni dimensional). The hypot hesis of perfect fit was rejected ( 2 4975, p<.0001, df 528). RMSEA 0.065 (CI 90% 0.06 to 0.07) was a reasonable fit; therefore, th e hypothesis of close fit coul d not be rejected. Although incremental fit indices (CFI .89 and NNFI .87) were mediocre, with consideration given to close fit, the model was acceptable.

PAGE 141

128 Measurement Model #1 Legend EL Emotional Labor ELD Emotional Labor Display ELE Emotional Laobr Effort PCD Perceived Customer Demands RA Role Ambiguity RC Role Conflict RO Role Overload ROEE19 E18 E17 E16 E15L? L? L? L? F2 R0 VAR1 ROA ROB ROC RODE12 E6 E5 E4 E3 E2 E1 RAF RAE RAD RAC RAB RAA L? L? L? L? L? L? RAG RAHE8 E7L? L? RAK RAIE10 E9L? L? RCH RCG RCD RCAE14 E13 E11 L? L? L? L? F4 RC VAR1 PCDD PCDBE21 E20L? L? PCDE PCDFE23 E22L? L? F3 PCD VAR1 ELDEE26 E25 E24L? L? L? ELDC ELDD ELDA ELDBE27 E28 L? L? ELEAE29 E30 E31 L? L? L? ELEC ELEB ELED ELEEE32 E33L? L? L? C? C? C? C? C? C? C? F4 EL VAR1 F1 RA VAR1 C? C? In order to examine job autonomy, the second model, Model 2a Proposed Job Stress Mediating Relationship Variables combines three proposed constructs: perceived customer demands, job autonomy, and job st ress. The hypothesis of perfect fit was rejected (2 168, p<.0001, df 59). RMSEA 0.085 (CI90% 0.06 to 0.09) was mediocre; however, the hypothesis of close fit was not rejected. Acceptance was also based on other indices within the acceptable range (GFI .91, CFI .94, and NNFI .92) supporting the

PAGE 142

existence of a sound measurement model (MacCullum et al., 1996; Diamantopoulos & Siguaw, 2000). Measurement Model #2a LegendPCD = Perceived Customer DemandsJA = Job AutonomyJS = Job Stress PCDE PCDFE3E4L? PCDD PCDBE1E2L?L? F1PCDVAR=1L? L? JAA JABE5E6 L?L? F2JAVAR=1 JACE7 JSC JSDE10E11L?L? JSB JSAE8E9L?L? F3JSAVAR=1 L? JSH JSIE12E13 L? C?C?C? Model 2b Proposed Job Stress Mediating Relationship Variables examines three proposed multidimensional constructs: emotional labor (three-dimensional) previously examined, emotional intelligence (three-dimensional), and job stress (two-dimensional). Model #2b ( 2 377, p<.0001, df 202) was considered an adequate model and the hypothesis that it is a true model was not rejected. As well, RMSEA 0.057 (CI 90% 0.05 to 0.07) was a reasonable fit and indicates that the hypothesis of close fit could not be rejected. Incremental CFI (.91) falls into the acceptance range supporting the 129

PAGE 143

existence of a good measurement model (MacCullum et al., 1996; Diamantopoulos & Siguaw, 2000). Measurement Model #2b LegendEI = Emotional IntelligenceEL = Emotional LaborELD = Emotional Laobr DisplayELE = Emotional Lbor EffortJS = Job StressE16 EIB EICE13E14 L?L? F2EIVAR=1 EIA EIKE15 L?L? EIME17 L? EIF EIIE11E12 L?L? JSC JSDE20E21 L?L? JSB JSAE18E19L?L? F3JSVAR=1 JSH JSIE22E23 L?L? C?C?C? ELDE ELDC ELDD ELDA ELDB ELEA ELEC ELEB ELED ELEE F1ELVAR=1 E1E10 130 E9 E8E7E6E5E4E3E2 Model 3 Proposed Outcome Variables examines physical consequences, attitude toward the job, organizational commitment, overall job satisfaction, emotional exhaustion frequency, and emotional exhaustion intensity. Fit indices for this 32-indicator model fell outside acceptable ranges for all indicators and demonstrated a high AIK (135.05). Therefore, indicator loadings, standard errors, and t-values were examined. However, no

PAGE 144

131 significant problems were found with indicator loadings, standard er rors, and t-values. Thus, the LaGrange modification index was examined to determine if the chi-square would improve if an indictor influencing more than one factor was reassigned or completely dropped. Based on this index, Indicator A of emotional exhaustion frequency was removed. Subsequently, the chi square was reduced by 101.56 and other indicators improved, as well. After removi ng this indicator, a second model was run and all indices were re-examined. Indices were still not in the acceptable range. Therefore, the LaGrange modification index was examined and the removal of Indicator B for the organizational commitment construct was show n to significantly improve chi square and was, therefore, removed. Although the model improved, it was still outside the acceptable parameters. Upon examination of the third iterative CFA, based on an examination of the modification index, Indi cator E from the emotional exhaustion frequency construct was removed. Fit indices for the respecified model of absolute fit demonstrated that the hypothesis of perfect fit was rejected ( 2 891, p<.0001, df 428). RMSEA 0.064 (CI 90% 0.06 to 0.07) was a reasonable f it and the hypothesis of close fit was not rejected. The incremental fit inde x was in the mediocre range (CFI = .87); however, when considered as part of a global fit, the model was adequate. An AIK = 34.57 demonstrated parsimony of the respecified model.

PAGE 145

132 C?Measurement Model #3 LegendATJ = Attitude Toward the JobEE = Emotional ExhaustionEEF = Emotional Exhaustion FrequencyEEI = Emotional Exhaustion IntensityOC = Organizational CommitmentOJS = Overall Job SatisfactionPC = Physical ConsequencesE14 ATJF ATJGE13 L?L? PCC PCDE1E2 L?L? PCIE3 L? PCE PCGE4E5 L?L? F1PCVAR=1 PCB PCFE6E7 L?L? PCHE8 L? ATJD ATJEE11E12 L?L? ATJB ATJAE9E10L?L? F2ATJVAR=1 OJSI OJSJE20E21 L?L?L? F4OJSVAR=1 OJSHE22 OJSKL?E23 EEFFE26L?L? EEFC EEFBE24E25L? EEFGE27L? EEIEE31 L?L? EEIC EEIBE29E30L?L? F5EEVAR=1 EEFIE28 C?C?C?C?C? OOCE OOCFE15E16 F3OOCVAR=1 OOCGE17 OOCH L?L?L?E18L? L? OOCIE19 EEIIE32 L? C?C?C?C?

PAGE 146

133 Fully accessing the mediating relationships of emotion-focused coping on the five outcome variables required running two sepa rate CFA models. Trying to access these relationships with one CFA model violated variable parameter limitations for confirmatory factor analysis. Hence, Mode l 4a represents mediating relationships of emotion-focused coping and the two proposed new constructs, physical consequences and attitude toward the job. Model 4b de picts the proposed mediating relationship of emotion-focused coping on organizational commitment, overall job satisfaction, and emotional exhaustion. Model 4a Proposed Outcome Mediating Relationships examines emotionfocused coping along with the previously ex amined constructs, physical consequences and attitude toward the job. Due to a lack of model fit in the first two iterative factor analyses, a decision to remove Indicator C from the emotion-focused coping frequency dimension and Indicator A from emoti on-focused coping intensity dimension was based on the LaGrange modifica tion index. The respecified m odel demonstrated that the hypothesis of perfect fit was rejected ( 2 312, p<.0001, df 149). However, RMSEA 0.055 (CI 90% 0.06 to 0.08) indicated a reasonable fit and the hypothesis of close fit was not rejected. A good incremental fit (CFI = .90) provided additional assurance of a good model. An AIK of 14.76 demonstrated grea ter parsimony than two prior models (AIK1 = 262, AIK2 = 137).

PAGE 147

134 Measurement Model #4aL? LegendATJ = Attitude Toward the JobEFC = Emotion-Focused CopingEFCE = Emotion-Focused Coping Extent of UseEFCF = Emotion-Foucsed Coping Frequency of UsePC = Physical ConsequencesL?L?E3 EFCFA EFCFBE1E2 EFCFD L? EFCED EFCECE6E5 L?L? F1EFCVAR=1 EFCEBE4 L? PCC PCDE7E8 L?L? PCIE9 L? PCE PCGE10E11 L?L? F2PCVAR=1 PCB PCFE12E13 L?L? PCHE14 L? ATJD ATJEE17E18 L?L? ATJB ATJAE15E16L?L? ATJF ATJGE19 F5ATJVAR=1E20 L?C?C?C?

PAGE 148

135 Model 4b Proposed Outcome Mediating Relationships examines emotionfocused coping and outcome constructs of organizational commitment, overall job satisfaction, and emotional exhaustion. The model demonstrates that the hypothesis of perfect fit is rejected ( 2 473, p<.0001, df 168). However, RMSEA 0.08 (CI90% 0.06 to 0.08) indicated a reasonable f it and the hypothesis of close fit was not rejected. A mediocre incremental fit (CFI = .84) provi ded additional assurance of an acceptable model. Measurement Model #4b Legend EFC Emotion Focused Coping EFCE Emotion-Focused Copintg Effectivenss of Use EFCF Emotion-Foucsed Coping Frequency of Use OOC Overall Organizational Commttment OJS Overall Job Satisfaction EE Emotional Exhaustion EEF Emotional Exhaustion Frequency EEI Emotional Exhaustion Intensity L? L?E3 EFCFA EFCFBE1 E2 EFCFD L? EFCED EFCECE6 E5 L? L? F1 EFC VAR1 EFCEBE4 L? C? C? OJSI OJSJE12 E13 L? L? L? F3 OJS VAR1 OJS HE14 OJS K L?E15 EEF FE18L? L? EEF C EEF BE16 E17L? EEF GE19L? EEIEE23 L? L? EEIC EEIBE21 E22L? L? F4 EE VAR1 EEFIE20 C? C? OO CE OOC FE7 E8 F2 OOC VAR1 OOC GE9 OOC H L? L?E10L? OOC IE11 EEIIE24 L? L? L?C? C?

PAGE 149

136 In the final model, Model #5 Proposed Performance/Intentions Relationships three constructs (i.e., job performance, intention to switch positions within the organization, and intention to leave the orga nization) were examin ed. The hypothesis of perfect fit was rejected ( 2 1435, p<.0001, df 45). RMSEA 0.05 (CI90% 0.05 to 0.09) was a good fit and the hypothesis of absolute fit wa s not rejected. Acceptance was also based on other indices within the acceptable ra nge (GFI .95, CFI .96, and NNFI .95) supporting the existence of a sound measurement model (MacCullum et al., 1996; Diamantopoulos & Siguaw, 2000). The AIK for this mode l was very low (13.62) indicating a parsimonious model (Hatcher, 1994). Legend ITL = Intenion to Leave ITS = Intention to Switch Postions JP = Job PerformanceMeasurement Model #5 JPC JPDE1 E2 L? L? F1 JP VAR=1 JPEE3 JPF JPG L? L? L?E4 E5 ITLC ITLDE8 E9 L? L? F3 ITL VAR=1 ITLEE10 L? ITSA ITSBE6 E7 L? L? F2 ITS VAR=1 C? C? C?

PAGE 150

137 Reliability The scales used to capture the data for all the major constructs were statistically reliable. Reliabilities of the constructs are considered critical (Gerbing & Anderson, 1988). For each of the key constructs, Cronbachs alphas were greater than .76 demonstrating that there is overall internal consistency within the respective scale indicators measuring a particular construct. Furthermore, the findings indicate 13 out of 18 constructs had alphas greater than .80. In addition, composite reliability values associated with each of the models key c onstructs suggest strong internal consistency among the indicators measuring each composite value, with all constructs having a composite reliability value of > .755. Further support was found by examining the average variance extracted within each construc t. Fifteen (15) out of the 18 constructs have average variance extracted >.51 dem onstrating that the indicators are truly representative of the latent construct (Hair, Jr. et al., 1998). The average variance extracted was slightly lower for three constructs [i.e., attitude toward the job (.455), emotional intelligence (.451), and physical consequences (.493)]. See Table 4.9 below for extracted variance and reliability scores. Overall, the reliability results for each of the constructs were encouraging.

PAGE 151

138 Table 4.9 Scale and Construct Reliabilit ies and Variance Assessments: Measurement Model Construct Cronbachs Composite Reliability Average Percent of Variance Extracted Perceived Customer Demand (PDC) 0.895 0.906 0.597 Role Ambiguity (RA) 0.869 0.927 0.561 Role Conflict (RC) 0.828 0.828 0.553 Role Overload (RO) 0.887 0.870 0.621 Emotional Labor (EL) 0.781 0.932 0.960 Emotional Intelligence (EI) 0.766 0.843 0.451 Job Autonomy (JA) 0.906 0.881 0.769 Emotion-Focused Coping (Effectiveness Dimension) (EFCE) 0.766 0.755 0.514 (Frequency Dimension) (EFCF) 0.774 0.778 0.553 Job Stress (JS) 0.802 0.891 0.579 Physical Consequences (PC) 0.802 0.871 0.493 Attitude Toward the Job (ATJ) 0.775 0.806 0.455 Overall Job Satisfaction (OJS) 0.861 0.852 0.594 Overall Organizational Commitment (OOC) 0.821 0.834 0.506 Emotional Exhaustion (EE) 0.838 0.922 0.531 Job Performance (JP) 0.841 0.851 0.795 Intention to Switch Positions (ITS) 0.856 0.885 0.795 Intention to Leave (ITL) 0.890 0.887 0.730 Note: Significant Level = .001 Convergent Validity Convergent validity indicates that a constr ucts specified indicators are, in fact, correlating with their respect ive constructs. Table 4.10 su mmarizes the standardized factor loadings of each indicator theoretica lly associated with e ach construct. For example, for the role ambiguity construct, the standardized f actor loadings of the six (6) indicators for the dimension company expectations illustrate converg ent validity in the

PAGE 152

139 M C S F L t P V E sense that those six (6) loadings represent strong measures of the company expectations dimension and not measures of either of the other two dimensions (i.e., workload or customer interaction). The results suggest evidence of convergent validity for each construct with indicator loadings ( ) ranging from .48 to .97 significant at (<0.001) (tvalues > 3.291) (e.g., Diamantopoulos & Siguaw, 2000). Table 4.10 Final Measurement Models Factor Loadings odel #1 Proposed Sources of Job Stress onstruct and Scale Items tandardized actor oadings ( ) -Statistics ercent of ariance xplained R ole Ambiguity (Composite Reliability = .927) D H o 1 .49 imension: Company Expectations ow I am expected to handle my customers bjections is 70 2.29 H a 1 .59 W t 1 .64 W h 1 .45 T a 1 .52 H i 1 .57 ow I am expected to handle unusual problems nd situations is 77 4.04 hich specific company strengths I should present o customers is 80 5.05 hich specific service benefits I am expected to ighlight for customers is 67 1.71 he actions required in meeting customers needs re 72 2.88 ow much service I should provide my customers s 77 4.13 R ole Ambiguity D imension: Work Load T 1 he amount of work I am expected to do is 92 3.56 85 H 1 ow much work I am expected to do is 71 0.78 50 R ole Ambiguity D imension: Customer Interaction H c 1 .59 ow I am expected to interact with my ustomers 77 0.60 H t .62 9.02 .38 ow I should behave (with customers) while on he job is R ole Conflict (Composite Reliability = .828) I d 1 .38 have to do things that I believe should be done ifferently. 62 0.47 I e 1 .40 I c 1 .69 I 1 usually receive incompatible requests from ither customers or my supervisors. 63 0.65 receive assignments without the manpower to omplete them. 83 5.45 receive assignments without adequate 86 6.01 74

PAGE 153

140 r esources/material to execute them. R ole Overload (Composite Reliability = .870) M 1 y job requires me to work very fast. 77 4.12 59 M 1 M d 1 .67 T 1 I 1 y job requires me to work very hard. 84 6.19 71 y job leaves me with little time to get things one. 82 5.44 here is a great deal of work to be done. 81 5.32 66 have to do more work than I can do well. 69 2.29 48 P erceived Customer Demands (Composite Reliability = .906) I 1 n terms of quality, the customers I serve are 76 3.97 58 C 1 C o 1 .77 C s 1 .74 ustomers expectations for service are 86 6.71 74 ustomers expectations that the services ffered will meet his or her needs are 88 7.52 ustomers expectations for delivery level of ervice quality are 86 6.91 E motional Labor (Composite Reliability = .932) D imension: Display Expectation of Behavior f s 1 .53 I am expected to make an effort to actually eel the emotions that management believes I hould feel. 73 2.74 t 1 .76 e 1 .59 I am expected to try to actually experience he emotions that I must show on the job. 87 5.89 I am expected to really try to feel the motions I have to show as part of my job. 77 3.57 E motional Labor D imension: Display Required Behavior t 1 .71 I am required to pretend to have emotions hat I dont really have. 84 2.14 s 9 .40 I am required to hide my true feelings about a ituation. 63 .54 E motional Labor D imension: Effort F 1 ake the emotions I show customers. 69 2.00 48 T w 1 .64 S c 1 .46 C m 1 .67 A p 1 .58 alk myself out of feeling what I really feel hen helping customers. 80 4.69 ummon up the feelings I need to show to ustomers. 68 1.74 hange my actual feelings to match those that I ust express to customers. 82 5.00 ttempt to create certain emotions in myself that resent the image my organization desires. 76 3.59

PAGE 154

141 Model #2a Proposed Job Stress Med iating Relationship Variables Construct and Scale Items Standardized Factor Loadings ( ) t-Statistics Percent of Variance Explained Perceived Customer Demands (Com posite Reliability = .906) In terms of quality, the customers I serve are .76 13.92 .58 Customers expectations for service are .86 16.80 .74 Customers expectations that the services offered will meet his or her needs are .88 17.51 .77 Customers expectations for delivery level of service quality are .86 16.77 .74 Job Autonomy (Composite Reliability = .881) ... decide how to go about doing my work. .88 17.56 .77 use my personal initiative or judgment in carrying out the work. .91 18.54 .83 have the opportunity for independence and freedom in how I do my work. .84 16.34 .71 Job Stress (Composite Reliability = .891) Dimension: Anxiety I often feel fidgety or nervousness as a result of my job. .84 12.73 .71 My job irritates me more than it should. .73 15.27 .53 On the job, there are lots of times when my job drives me right up a wall. .83 13.53 .69 Sometimes, when I think about my job I get a tight feeling in my chest. .76 11.27 .58 Job Stress Dimension: Time Pressure I sometimes dread the telephone ringing at home because the call might be job-related. .66 8.64 .44 I frequently get the feeling I am married to the company. .73 9.23 .53 Model #2b Proposed Job Stress Mediating Relationship Variables Construct and Scale Items Standardized Factor Loadings ( ) t-Statistics Percent of Variance Explained Emotional Labor (Composite Reliability = .932) Dimension: Display Expectation of Behavior I am expected to make an effort to actually feel the emotions that mana gement believes I should feel. .73 12.65 .53 I am expected to try to actually experience the emotions that I must show on the job. .88 16.19 .77 I am expected to really try to feel the emotions I have to show as part of my job. .76 13.32 .58

PAGE 155

142 Emotional Labor Dimension: Display of Required Behavior I am required to pretend to have emotions that I dont really have. .79 11.92 .62 I am required to hide my true feelings about a situation. .67 10.28 .45 Emotional Labor Dimension: Effort Fake the emotions I show customers. .69 12.02 .48 Talk myself out of feeling what I really feel when helping customers. .80 14.66 .64 Summon up the feelings I need to show to customers. .68 11.83 .46 Change my actual feelings to match those that I must express to customers. .81 14.97 .66 Attempt to create certain emotions in myself that present the image my organization desires. .76 13.55 .58 Emotional Intelligence (Com posite Reliability = .843) Dimension: Management Self Generally when I feel angry, I am .63 8.06 .40 My ability to control my emotions is .76 9.06 .58 Emotional Intelligence Dimension: Perception Self My ability to figure out the reasons behind my different emotions is .62 7.43 .38 Differentiating between emotions I experience is .85 9.22 .72 Emotional Intelligence Dimension: Management/Perception Others Recognizing emotions that I experience in a particular situation is .48 6.22 .23 When it comes to other peoples feelings at work, acknowledging their feelings is .58 7.29 .34 Examining the feelings, thoughts, and actions of others is .48 6.19 .23 Job Stress (Composite Reliability = .891) Dimension: Anxiety I often feel fidgety or nervousness as a result of my job. .73 12.91 .53 My job irritates me more than it should. .85 15.67 .72 On the job, there are lots of times when my job drives me right up a wall. .75 13.36 .56 Sometimes, when I think about my job, I get a tight feeling in my chest. .65 10.97 .42 Job Stress Dimension: Time Pressure I sometimes dread the telephone ringing at home because the call might be job-related. .73 10.06 .53 I frequently get the feeling I am married to the .65 9.17 .42

PAGE 156

143 company. Model #3 Proposed Outcome Variables Construct and Scale Items Standardized Factor Loadings ( ) tStatistics Percent of Variance Explained Physical Consequences (Com posite Reliability = .871) Dimension: Digestive Diarrhea .65 10.33 .42 Constipation .75 12.21 .56 Stomach cramps (not menstrual) .68 10.82 .46 Physical Consequences Dimension: Emotional Headache .68 9.90 .46 Loss of appetite .71 10.17 .50 Physical Consequences Dimension: Cardiovascular Chest pain .70 11.50 .49 Heart pounding when not exercising .70 11.38 .49 Shortness of breath .74 12.18 .55 Attitude Toward the Job (Composite Reliability = .806) Dimension: Overall Being enthusiastic about my job is .75 13.65 .56 My job tends to be .75 13.86 .56 Overall working conditions are .64 11.24 .41 Overall I see my job as .66 11.55 .44 Attitude Toward the Job Dimension: Boss Feedback from my superiors is .66 8.83 .44 Overall, communications with my boss are .57 7.97 .32 Organizational Commitment (Composite Reliability = .834) I am embarrassed to tell others that I am part of the organization. .66 11.41 .44 This organization really inspires the very best in me in the way of job performance. .69 11.98 .48 I am extremely glad that I chose this organization to work for over others I was considering at the time I joined. .86 16.43 .74 I do not care about the fate of this organization. .61 10.39 .37 For me, this is the best of all possible organizations for which to work. .71 12.62 .50 Overall Job Satisfaction (Composite Reliability = .852) Compared to your ideal job, your current job is .62 10.66 .38 To what extent does your current job match your expectations when you took it .76 13.90 .58 What is your overall satisfaction/dissatisfaction .81 15.35 .66

PAGE 157

144 with your current job Your overall feeling about your job would be .87 17.04 .76 Emotional Exhaustion (Composite Reliability = .922) Dimension: Frequency I feel used up at the end of the workday. .81 9.38 .66 When I get up in the morning to face another day on the job I feel tired. .66 12.87 .44 I feel frustrated with my job. .72 11.62 .52 I feel I am working too hard on my job. .73 11.40 .53 I feel like Im at the end of my rope with my job. .93 5.71 .87 Emotional Exhaustion Dimension: Intensity How do you feel at the end of the workday? .79 10.52 .62 How do you feel in the morning knowing you have to face another day on the job? .67 13.37 .45 How burned out do you feel from your work? .80 10.22 .64 How does your job make you feel? .64 13.90 .41 Model #4 Proposed Outcome Mediating Relationships Construct and Scale Items Standardized Factor Loadings ( ) t-Statistics Percent of Variance Explained Emotion-Focused Coping Frequenc y (Composite Reliability = .778) Turn to hobbies /pastimes. .48 7.50 .23 Talk to understanding friends. .83 14.14 .69 Seek social support. .86 14.68 .74 Emotion-Focused Coping Effectiveness (Composite Reliability = .755) Talk to understanding friends. .83 13.65 .69 Expand interests/activities outside of work. .54 8.37 .29 Seek social support. .75 12.29 .56 Physical Consequences (Com posite Reliability = .871) Dimension: Digestive Diarrhea .64 10.13 .58 Constipation .76 12.30 .58 Stomach cramps (not menstrual) .68 10.89 .46 Physical Consequences Dimension: Emotional Headache .66 9.28 .44 Loss of appetite .73 9.92 .53

PAGE 158

145 Physical Consequences Dimension: Cardiovascular Chest pain .70 11.34 .49 Heart pounding when not exercising .70 11.44 .49 Shortness of breath .74 12.21 .55 Attitude Toward the Job (Composite Reliability = .806) Dimension: Overall Being enthusiastic about my job is .79 13.86 .62 My job tends to be .77 13.22 .59 Overall working conditions are .59 9.44 .35 Overall I see my job as .64 10.46 .41 Attitude Toward the Job Dimension: Boss Feedback from my superiors is .64 7.24 .41 Overall, communications with my boss are .59 6.97 .35 Model #5 Proposed Performanc e/Intentions Relationships Construct and Scale Items Standardized Factor Loadings ( ) t-Statistics Percent of Variance Explained Job Performance (Composite Reliability = .851) Telling the customer the straight facts rather telling them what they want to hear. .69 11.94 .48 Performing your job dependably/ accurately. .72 12.61 .52 Servicing the account with the customers best interest in mind. .78 13.93 .61 Listening attentively to identify as well as understand the concerns of customers. .78 14.15 .61 Working out solutions to each customers questions or concerns. .68 11.72 .46 Intention to Switch (Composite Reliability = .885) I will actively look for another position within this organization .94 11.79 .88 I plan to switch positions within the organization. .84 11.05 .71 Intention to Leave (Composite Reliability = .887) I will actively look for a job with another company. .93 18.93 .87 I plan to switch companies. .97 20.26 .94 I plan to work for my current employer. .62 10.89 .38 Note : All loadings are significant at = 0.001.

PAGE 159

146 Discriminant Validity Assessing discriminant validity is important in efforts to ensure distinctiveness between the models key constructs. The dete rmination of discriminant validity between the constructs aids in ensuring that the scal e indicators used in measuring one particular construct are distinctively different than indica tors used in measuring another construct. The correlation coefficients displayed in Table 4.11 repres ent the degree of interrelationship between the standardized mean value measures of each of the key constructs. The results suggest the existen ce of acceptable distinctiv eness between all of the constructs. For example, the magnitude of the interrelationship among the proposed sources of job stress, the perceived custom er demands construct, and the construct of role ambiguity is -.262 which suggests th at the scale indicato rs used to assess perceived customer demands are distinctivel y different from thos e indicators used to measure role ambiguity.

PAGE 160

147 Constructs Mean SD EigenValue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 P erceived Customer Demands 515 128 3112 .906 2 ole Ambiguity 187 677 4990 1236 1028 -262 R 927 3 ole Conflict 365 133 2637 029 206 R .828 4 ole Overload 425 131 3455 287 -017 311 R .870 5 motional Labor 292 102 2769 1103 3212 160 053 268 273 E .932 6 Job Autonomy 542 132 2534 -052 -137 -231 -145 -314 .881 7 motional Intelligence 549 768 3160 1494 1119 024 -196 -043 090 -095 225 E .843 8 ob Stress 303 1304 3394 1048 104 174 450 421 350 -354 -193 J .891 9 Emotion-Focused Coping ffectiveness 602 851 1900 -031 -141 -122 042 066 171 141 -089 E .755 10 Emotion-Focused Coping equency 417 1220 2439 012 -060 -084 154 106 003 119 -059 413 Fr .778 11 hysical Consequences 162 553 3371 1122 1001 166 057 256 343 218 -244 -033 524 -064 -100 P .871 12 ttitude Toward the Job 485 106 2726 1148 -029 -280 -381 -316 -234 484 255 -596 112 061 -307 A .806 13 verall Organization mmitment 537 1191 3639 007 -300 -302 -114 -158 412 171 -412 162 037 -242 677 Co O .834 14 verall Job Satisfaction 493 1234 2755 -054 -292 -295 -246 -238 433 178 -567 086 034 -292 724 611 O .852 15 motional Exhaustion 278 889 4741 3687 097 178 460 380 296 -310 -088 676 -070 -014 493 -666 -460 -594 E .922 16 ob Performance 593 667 3158 127 -376 -145 052 -066 244 335 -128 187 156 -056 294 290 263 -078 J .851 17 tention to Switch Positions ithin the Organization 285 1855 1789 086 246 237 110 145 -331 -122 335 -008 021 244 -508 -630 -588 412 -223 In w .885 18 ntention to Leave the rganization 268 1855 2402 017 098 128 087 214 -220 -100 185 -055 -030 143 -198 -100 -268 165 -131 246 O I .887 Table 4.11 Degree of Interrela tionship Between Constructs Note. Scale composite reliabilities are on the diagonal. N =260 *Multiple eigenvalues are reported for construc ts that were shown to be multidimensional.

PAGE 161

Construct Validation Construct validity seeks agreement between the theoretical concept and the operational measurement of the construct. In assessing construct validity, discriminant and convergent validity enhances interpretation of construct validity. As presented above, results from examining convergent and discriminant validity demonstrate construct validation. Hypotheses Testing This section provides highlights of the analysis and results of testing the hypothesized relationships summarized in Table 4.12. Table 4.12 Summary of Findings of Proposed Hypotheses Testable Hypotheses Relationships Decision H 1 : Role ambiguity (RA) has a significant positive impact in creating FSSP job stress (JS). RA JS Accept H 2 : Role conflict (RC) has a significant positive impact in creating FSSP job stress (JS). RC JS Accept H 3 : Role overload has a significant positive impact in creating FSSP job stress (JS). RO JS Accept H 4 : Job autonomy (JA) mediates the relationship between perceived customer demands (PCD) and job stress (JS) such that job autonomy (JA) reduces the impact of perceived customer demands (PCD) on job stress (JS). PCD JA JS Reject H 5 : Emotional intelligence (EI) significantly mediates the relationship between emotional labor (EL) and job stress (JS) such that it reduces the impact of emotional labor (EL) on job stress (JS). EL EI JS Accept H6: The stronger the presence of job autonomy (JA), the less likely FSSP will engage in emotion-focused coping (EFC). JA EFC Accept H7a: There is a significant relationship between emotional intelligence (EI) and emotion-focused coping effectiveness (EFCE). EI EFCE Accept 148

PAGE 162

149 H7b: There is a significant relationship between emotional intelligence (EI) and emotion-focused coping frequency (EFCF). EI EFCF Reject H8a: Emotion-focused coping effectiveness (EFCE) significantly mediates the positive relationship between job stress (JS) and emotional exhaustion (EE) such that it reduces the impact of job stress (JS). JS EFCE EE Reject H8b: Emotion-focused coping frequency (EFCF) significantly mediates the positive relationship between job stress (JS) and emotional exhaustion (EE) such that it reduces the impact of job stress (JS). JS EFFC EE Reject H9a: Emotion-focused coping effectiveness (EFCE) significantly mediates the negative relationship between job stress (JS) and attitude toward the job (ATJ) such that it reduces the impact of job stress (JS). JS EFCE ATJ Reject H9b: Emotion-focused coping frequency (EFCF) significantly mediates the negative relationship between job stress (JS) and attitude toward the job (ATJ) such that it reduces the impact of job stress (JS). JS EFCF ATJ Reject H10a: Emotion-focused coping effectiveness (EFCE) significantly mediates the negative relationship between job stress (JS) and overall job satisfaction (OJS) such that it reduces the impact of job stress (JS). JS EFCE OJS Reject H10b: Emotion-focused coping frequency (EFCF) significantly mediates the negative relationship between job stress (JS) and overall job satisfaction (OJS) such that it reduces the impact of job stress (JS). JS EFCF OJS Reject H11a: Emotion-focused coping effectiveness (EFCE) significantly mediates the negative relationship between job stress (JS) and overall organizational commitment (OOC) such that it reduces the impact of job stress (JS). JS EFCE OOC Accept H11b: Emotion-focused coping frequency (EFCF) significantly mediates the negative relationship between job stress (JS) and overall organizational commitment (OOC) such that it reduces the impact of job stress (JS). JS EFCF OOC Reject H12a: Emotion-focused coping effectiveness (EFCE) significantly mediates the positive relationship between job stress (JS) and physical consequences (PC) such that it reduces the impact of job stress (JS). JS EFCE PC Accept H12b: Emotion-focused coping frequency (EFCF) significantly mediates the positive relationship between job stress (JS) and physical consequences (PC) such that it reduces the impact of job stress (JS). JS EFCF PC Reject

PAGE 163

150 HO13: H18: Overall organizational commitment (OOC) has a significant positive influence on job performance (JP). OOC JP Accept H19: Emotional exhaustion (EE) has a significant negative influence on job performance (JP). EEF JP Accept H20: Physical consequences (PC) have a significant negative influence on job performance (JP). PC JP Accept H21: Job performance (JP) has a significant negative influence on intention to switch positions within the organization (ITS). JP ISP Accept H22: Intention to switch positions (ITS) within an organization and intention to leave an organization (ITL) are correlated. ISP ITL Accept H23: JP ITL Job performance (JP) has a significant negative influence on intention to leave the organization (ITL). Accept H15: Overall organizational commitment (OOC), overall job satisfaction (OJS), and emotional exhaustion (EE) are significantly correlated with each other. H15a: Overall organizational commitment (OOC) is significantly correlated with overall job satisfaction (OJS). OOC OJS Accept H15b: Overall organizational commitment (OOC) is significantly correlated with emotional exhaustion (EE). OOC EE Accept H15c: Overall job satisfaction (OJS) is significantly correlated with emotional exhaustion (EE). OJS EE Accept H16: Attitude toward the job (ATJ) has a significant positive influence on FSSPs job performance (JP). ATJ JP Accept H17: Overall job satisfaction (OJS) has a significant influence on job performance (JP). OJS JP Accept HO14: Physical consequences are not significantly correlated with outcome variable (OV) of overall organizational commitment (OOC), overall job satisfaction (OJS), or emotional exhaustion (EE). PC OV Reject HO14a: Physical consequences (PC) are not significantly correlated with overall organizational commitment (OOC). PC OOC Reject HO14b: Physical consequences (PC) are not significantly correlated with overall job satisfaction (OJS). PC OJS Reject HO14c: Physical consequences (PC) are not significantly correlated with emotional exhaustion (EE). PC EE Reject HO13c: Attitude toward the job (ATJ) is not significantly correlated with physical consequences (PC). ATJ PC Reject HO13d: Attitude toward the job (ATJ) is not significantly correlated with emotional exhaustion (EE). ATJ EE Reject Attitude toward the job (ATJ) is not significantly correlated with other outcome variables in the model. HO13a: Attitude toward the job (ATJ) is not significantly correlated with overall organizational commitment (OOC). ATJ OOC Reject HO13b: Attitude toward the job (ATJ) is not significantly correlated with overall job satisfaction (OJS). ATJ OJS Reject

PAGE 164

151 Sources of Job Stress H1: Role ambiguity (RA) has a significant positive impact in creating FSSP job stress (JS). H2: Role conflict (RC) has a significant positive impact in creating FSSP job stress (JS). H3: Role overload (RO) has a significant positive impact in creating FSSP job stress (JS). The model conceptualizes that job stress has five antecedents, two of which were mediated by other relationships. To test H1 through H3 above, a regression analysis was conducted of the three proposed sources of job stress (i.e., role ambi guity, role conflict, role overload) regressed simultaneously as inde pendent variables to estimate their impact on FSSP job stress. The regression results i ndicated that these three sources were statistically significant and explain 30.9% percent of the variance of job stress (Unadjusted R2 = 0.309; F =78.877, p<0.0001). In other words, when any one of these sources of job stress is reported by FSSP as hi gh, then job stress is also expected to be high. Therefore, H1 through H3 were supported. Table 4.13 below shows the significance values of the regression coefficients.

PAGE 165

152 Mediating Relationships When testing any mediating effect for this study, the four-step process advocated by Baron and Kenny (1986) is employed. With this method, four regression analyses were conducted to examine complete mediation: 1) to find if the total effect of the independent variable on the dependent variables is significant, 2) to determine if the path from the independent variable to the mediator is significant, 3) to determine if the mediator has a significant unique effect on the dependent variable, and 4) to determine if the independent variable is associated with the dependent variable after the mediator variable is controlled. However, if there is a significant relationship between the mediator and the dependent variable while controlling for the direct effect of the independent variable, and the relationship from the independent variable to the dependent variable is still significant, then the model is consistent with partial mediation. If needed, a further test for partial mediation is conducted using the formula (z = 1 x 2 / square root of (12 x S22 + 22 x S12). Tests involve examining the 12, the regression coefficient, associated with the independent variable when a regression analysis examines the association between the independent variable and the mediator variable. Also, S12 represents the standard error associated with the independent variable for this regression Standard Table 4.13 Sources of Job Stress Regression Model Parameter Regression Coefficients Error T-Value Role Ambiguity .280 .073 3.821 Role Conflict .261 .037 7.099 Role Overload .214 .034 6.381 N = 260, Significance < .01

PAGE 166

153 analysis. 22, the regression coefficient associated with the mediator when the association of the mediator and the depende nt variable are regressed. Lastly, the S22 represents the standard error associated with the mediator for this regression analysis. A z-value > 1.96 is evid ence of partial mediation at p <.05. That is, the relationship between the independent variab le and the dependent variab le is partially mediated. Job Stress Mediating Relationships It was hypothesized that job autonomy w ill mediate the influence of perceived customer demands on job stress (H4) and emotional intelligence will mediate the influence of emotional labor on job stress (H5). H4: Job autonomy (JA) mediates the relationship between perceived customer demands (PCD) and job st ress (JS) such that job autonomy (JA) reduces the impact of perceived customer demands (PCD) on job stress (JS). Proposed Mediating Influence of Job Autonomy for Perceived Customer Demands and Job Stress A non-significance finding (p=.143) when regressing perceived customer demands and job autonomy demonstrated that job autonomy does not mediate the influence of perceived customer demands on job stress. Hypothesis 4 was not supported.

PAGE 167

Table 4.14 Mediation Analysis of Job Autonomy Relationships Hypothesis 4 Regression Coefficients t p-Value JS = a + 1(PCD) + e .105 2.318 .021 JA = a + 1(PCD) + e .000 .005 .143 154 A post hoc analysis of a moderating effect of job autonomy on the relationship between perceived customer demands and job stress was examined and was analyzed through a regression analysis including the interaction effect for job autonomy and perceived customer demands (i.e., JS = a + 1(PCD) + 2(JA) + 3(JA/PCD) + e). Regression findings show a significant (p = .0001) main effect for the influence of job autonomy on job stress, but no significant moderating (p =.996) influence of job autonomy. H5: Emotional intelligence (EI) significantly mediates the positive relationship between emotional labor (EL) and job stress (JS) such that it reduces the impact of emotional labor (EL) on job stress (JS). To test Hypothesis 5, the mediating relationship of emotional intelligence between emotional labor and job stress was examined. Emotional labor continues to have an effect on job stress when emotional intelligence has been controlled (p=.0001). Proposed Mediating Influence of Emotional Intelligence for Emotional Labor and Job Stress

PAGE 168

155 Hence, emotional intelligence does not completely mediate the influence of emotional labor on job stress (Baron & Kenny, 1986). Further analysis reveals that emotional intelligence partially mediates the relationship between emotional labor and job stress (z = 3.839, significant at .05). Hypothesis H5 was supported. Emotion-Focused Coping Hypotheses H6 and H7a-b. H6: The stronger the presence of job autonomy (JA), the less likely FSSP will engage in emotion-focused coping frequency (EFCF). p-Value Table 4.15 Mediation Analysis of Emotional Intelligence Relationships Hypothesis 5 Regression Coefficients t JS = a + 1(EL) + e .0001 .443 7.928 EI = a + 1(EL) + e .022 -.087 -2.299 JS = a + 1(EI) + e -.375 -5.718 .0001 JS = a + 1(EI) + 2(EL) + e -5.232 -.211 .304 7.563 .0001 .0001 The model conceptualizes a direct influence from job autonomy (JA) to emotion-focused coping frequency (EFCF). In order to examine this relationship, quartiles were examined through a one-way ANOVA for upper and lower quartiles of job autonomy. The analysis showed that the stronger the presence of job autonomy (i.e., highest quartile mean 3.714, std. dev. = 1.339 vs. lowest quartile mean 2.470, std. dev. = .969 significant at .0001) for FSSP, the smaller the value for emotion-focused coping. The findings

PAGE 169

156 H7a: There is a significant relations hip between emotional intelligence (EI) and emotion-focused c oping effectiveness (EFCE). To test H7a, a correlation analysis between emotional intelligence and emotionfocused coping effectiveness demonstrates a weak positive (r = .141, significant at <.05) relationship. Hypothesis 7a was supported. H7b: There is a significant relations hip between emotional intelligence (EI) and emotion-focused coping frequency (EFCF). demonstrated that FSSP with higher levels of job autonomy (mean = 4.344) engage in emotion-focused coping less frequently than FSSP with lesser (means <2.334) levels of job autonomy. Hypothesis 6 was supported. A correlation analysis between emotional intelligence and emotion-focused coping frequency does not demonstrate any signi ficant relationship at p<.05. In other words, FSSP reporting high emotional intellig ence do not report frequently engaging in emotion-focused coping. Hypothesis 7b was not supported.

PAGE 170

157 Test of Emotion-Focused Coping Mediation The model conceptualizes that emotionfocused coping (EFC) mediates between the outcome relationships (i.e., emotional e xhaustion, attitude toward the job, overall job satisfaction, overall organizat ional commitment, and physic al consequences) and job stress. Difficulties in analyzing the data were created by measurements for the two dimensions of emotion-focused coping (i .e., two types of scales: frequency and symmetrical). In order to an alyze the data, Hypotheses H8 through H12 have been split into parts a and b. Thus, a series of two regression analyses for each outcome variable were employed. One set of regres sions examined the influence of emotionfocused coping effectiveness while the other se t of regressions examined the influence of emotion-focused coping frequency for each of the five outcome variables following the Baron and Kenny (1986) four-step process of mediation testing. Table 4.16 presents detailed statistical analysis for emotion-focused coping effectiveness mediating relationships and Table 4.17 for emoti on-focused coping frequency mediating relationships. Both tables are located at the end of this section of the paper. H8a: Emotion-focused coping effectiven ess (EFCE) significantly mediates the positive relationship between job stress (JS) and emotional exhaustion (EE) such that it reduces the impact of job stress (JS). Emotion-Focused Coping Mediation from Job Stress to Em otional Exhaustion Frequency

PAGE 171

158 A finding of non-significance of the rela tionship between the mediator and the independent variable (p=.398) provides evidence that emotion-focused coping effectiveness does not mediat e the influence of job stre ss on emotional exhaustion. Although FSSP may perceive that emotion-focuse d coping is effective, the influence of job stress on emotional exhaustion was not reduced. Therefore, H8a was not supported. H8b: Emotion-focused coping frequenc y (EFCF) significantly mediates the positive relationship between job stress (JS) and emotional exhaustion (EE) such that it reduces the impact of job stress (JS). A non-significant relationship between emotion-focused coping frequency and emotional exhaustion (p=.991) showed that emotion-focused coping frequency did not mediate the influence of job stress on emo tional exhaustion. The frequency of engaging in emotion-focused coping did not reduce the FSSPs emotional exhaustion coming from job stress. Hypothesis H8b was not supported. H9a: Emotion-focused coping effectiven ess (EFCE) significantly mediates the negative relationship between job stress (JS) and attitude toward the job (ATJ) such that it reduces the impact of job stress (JS). Emotion-Focused Coping Mediation from Job Stress to Attitude Toward the Job

PAGE 172

159 Regression analyses show that job stress continues to have an effect on attitude toward the job when emotion-focused coping effectiveness has been controlled. Thus, emotion-focused coping effectiveness does not completely mediate the relationship between job stress and attitude toward the job (p=.0001). Upon further examination of the data for partial mediation, it was found that emotion-focused coping effectiveness was not a partial mediator (z = 1.842241, not sign ificant at .05) for the influence of job stress on attitude toward the job. H9a was not supported. H9b: Emotion-focused coping frequenc y (EFCF) significantly mediates the negative relationship between job stress (JS) and attitude toward the job (ATJ) such that it reduces the impact of job stress (JS). Based on the series of regression analyses examining the role of emotion-focused coping frequency between job stress and attit ude toward the job, it is shown that the relationship between attitude toward the j ob and emotion-focused coping frequency was not significant (p=.254). Therefore, the hypothesized mediating relationship does not exist. In other words, the frequency with which an FSSP engages in emotion-focused coping does not reduce the influence of job stress on attitude toward the job. H9b was not supported.

PAGE 173

160 Emotion-Focused Coping Mediation from Job Stress to Overall Job Satisfaction H10a: Emotion-focused coping effec tiveness (EFCE) significantly mediates the negative relationship be tween job stress (JS) and overall job satisfaction (OJS) such that it reduces the impact of job stress (JS). For the relationship proposed in H10a above, job stress continues to have an effect on the overall job satisfaction when emoti on-focused coping effectiveness has been controlled; therefore, emotion-focused coping effectiveness does not completely mediate the influence of job stress on overall job satisfaction (p=.0001). Upon further examination, partial mediation was also reject ed (z = 1.917 is not signi ficant at .05). H10a was not supported. The findings indicate that although FSSP perceive emotion-focused coping to be effective, the negative influen ce of job stress on overall job satisfaction is not reduced. H10b: Emotion-focused coping frequenc y (EFCF) significantly mediates the negative relationship between j ob stress (JS) and overall job satisfaction (OJS) such that it redu ces the impact of job stress (JS). Consistent with findings from previous regression analyses in this study, emotionfocused coping frequency does not mediate the negative influence of job stress on overall job satisfaction. The relationship between the overall job satisfaction and emotion-

PAGE 174

161 H11a: Emotion-focused coping effec tiveness (EFCE) significantly mediates the negative relationship be tween job stress (JS) and overall organizational commitment (OOC) such that it reduces the impact of job stress (JS). H11b: Emotion-focused coping frequenc y (EFCF) significantly mediates the negative relationship between job stress (JS) and overall organizational commitment (OOC) such that it reduces the impact of job stress (JS). focused coping frequency was not significan t (p=.545). Support for Hypothesis 10b was not found. Emotion-focused coping frequency do es not play a mediating role in reducing the influence of job stress on FSSPs overall job satisfaction. Emotion-Focused Coping Mediation from Job Stress to Overall Organizational Commitment All regression analyses in this media tion evaluation proce ss were significant, demonstrating that emotion-focused coping effectiveness does not completely mediate the relationship between job stress and overall organizati onal commitment. However, partial mediation was found based on the computed z-score of 2.502, p<.05. Hypothesis 11a was supported.

PAGE 175

162 H12a: Emotion-focused coping effec tiveness (EFCE) significantly mediates the positive relationship between job stress (JS) and physical consequences (PC) such th at it reduces the impact of job stress (JS). Findings from the regression analyses we re significant relationships for all regressed relationships. Ther efore, complete mediation was rejected and a partial mediating relationship was examined. Partial mediation was supported at a computed zscore of 2.019733, p<.05, thus indicating that the influence of job stress on physical consequence was reduced when FSSP perceive emotion-focused coping to be effective. Therefore, H12a was supported. The relationship between emotion-fo cused coping frequency and overall organizational commitment was not significant (p = .286). Therefore, there is no evidence of the proposed mediating relations hip of emotion-focused coping frequency between job stress and overall organizational commitment. In other words, emotionfocused coping does not reduce the influence of job stress on overall organizational commitment. Hypothesis 11b was not supported. Emotion-Focused Coping Mediation from Job Stress to Physical Consequences:

PAGE 176

163 H12b: Emotion-focused coping frequenc y (EFCF) significantly mediates the positive relationship between job stress (JS) and physical consequences (PC) such that it reduc es the impact of job stress (JS). The relationship between emotion-focu sed coping frequency and the physical consequences was not significant (p=.107). As evidenced in previous findings for hypotheses of the mediating influence of em otion-focused coping frequency, mediation was also not found for Hypothesis 12b. Em otion-focused coping frequency does not mediate the influence of job st ress on physical consequences.

PAGE 177

164 Table 4.16 Mediation Analysis for Emotion-Focused Coping Effectiveness Hypothesis 8a Regression Coefficients t p-Value EE = a + 1(JS) + e 0.470 20.193 .0001 EFCE = a + 1(JS) + e -.088 -02.934 .0030 EFCE = a + 1(EE) + e .0110 -.108 -02.566 EE = a + 1(EFCE) + 2(JS) + e -.029 0.467 -00.847 19.919 .3980 .0000 Hypothesis 9a ATJ = a + 1(JS) + e -17.459 -.604 .0001 EFCE = a + 1(JS) + e -.088 -02.934 .0030 EFCE = a + 1(ATJ) + e 0.124 03.491 .0010 ATJ = a + 1(EFCE)+ 2(JS) + e 0.091 -.500 02.150 -17.106 .0320 .0001 Hypothesis 10a OJS= a + 1(JS) + e -.571 -16.142 .0001 EFCE = a + 1(JS) + e -.088 -02.934 .0030 EFCE = a + 1(OJS) + e 0.107 03.590 .0001 OJS = a + 1(EFCE) + 2(JS) + e 0.083 -.563 02.321 -15.786 .0210 .0001 Hypothesis 11a OOC= a + 1(JS) + e -.429 -10.951 .0001 EFCE = a + 1(JS) + e -.088 -02.934 .0030 EFCE = a + 1(OOC) + e 0.135 04.154 .0001 OOC = a + 1(EFCE) + 2(JS) + e 0.125 -.413 03.196 -10.553 .0010 .0001 Hypothesis 12a PC= a + 1(JS) + e 0.239 13.222 .0001 EFCE = a + 1(JS) + e -.088 -02.934 .0030 EFCE = a + 1(PC) + e -.221 -0.3571 .0001 PC = a + 1(EFCE) + 2(JS) + e -.092 0.486 -02.429 12.870 .0150 .0001

PAGE 178

165 Table 4.17 Mediation Analysis for Emotion-Focused Coping Frequency Hypothesis 8b Regression Coefficients t p-Value EE = a + 1(JS) + e 0.470 20.193 .0001 EFCF = a + 1(JS) + e -.067 -1.614 .1070 EE = a + 1(EFCF) + e 0.004 0.112 .9110 Hypothesis 9b ATJ = a + 1(JS) + e -.604 -17.459 .0001 EFCF = a + 1(JS) + e -.067 -1.614 .1070 ATJ = a + 1 (EFCF) + e 0.056 01.143 .2540 Hypothesis 10b OJS= a + 1(JS) + e -.571 -16.142 .0001 EFCF = a + 1(JS) + e -.067 -01.614 .1070 OJS = a + 1(EFCF) + e 0.027 0.606 .5450 Hypothesis 11b OOC= a + 1(JS) + e -.429 -10.951 .0001 EFCF = a + 1(JS) + e -.067 -01.614 .1070 OOC = a + 1(EFCF) + e .044 01.067 .2860 Hypothesis 12b PC= a + 1(JS) + e 0.239 13.222 .0001 EFCF = a + 1(JS) + e -.067 -1.614 .1070 PC = a + 1(EFCF) + e -.054 -2.478 .0140 Interrelationships of Outcome Variables HO13: Attitude toward the job (ATJ) is not significantly correlated with other outcome variables in the model. HO13a: Attitude toward the job (ATJ) is not significantly correlated with overall organizational commitment (OOC). HO13b: Attitude toward the job (ATJ) is not significantly correlated with overall job satisfaction (OJS).

PAGE 179

166 HO13c: Attitude toward the job (ATJ) is not significantly correlated with physical consequences (PC). HO13d: Attitude toward the job (ATJ) is not significantly correlated with emotional exhaustion (EE). HO14: Physical consequences are not significantly co rrelated with outcome variables of overall or ganizational commitment (OOC), overall job satisfaction (OJS), or emotional exhaustion (EE). HO14a: Physical consequences (PC) are not significantly correlated with overall organizational commitment (OOC). HO14b: Physical consequences (PC) are not significantly correlated with overall job satisfaction (OJS). HO14c: Physical consequences (PC) are not significantly correlated with emotional exhaustion (EE). H15: Overall organizational comm itment (OOC), overall job satisfaction (OJS), and emo tional exhaustion (EE) are significantly correlated with each other.

PAGE 180

167 H15a: Overall organizational commit ment (OOC) is significantly correlated with overall job satisfaction (OJS). H15b: Overall organizational commit ment (OOC) is significantly correlated with emoti onal exhaustion (EE). H15c: Overall job satisfaction (OJS) is significantly correlated with emotional exhaustion (EE). The hypothesized null interrel ationships proposed in HO13(a-d) between attitude toward the job and other outcome variab les (i.e., physical consequences, overall organizational commitment, overall job satis faction, or emotional exhaustion) were rejected. Correlations shown in Table 4.18 clearly indicate ther e are relationships between attitude toward the job and the proposed outcomes. As such, HO13(a-d) were rejected. That is, all of th e null hypotheses were rejected. For example, high positive correlations exist between attitude toward the job and overall organizational commitment (r = .68) and overall job satisfaction (r = .72) Attitude toward the job was correlated with overall organizational commitment and ove rall job satisfaction such that those who reported having a high positive attitude toward the job also reported high positive overall organizational commitment and high overall job satisfaction. Those who reported high positive attitudes toward the job were found to report low emotional exhaustion (r =

PAGE 181

168 .666). FSSP reporting high positive attitude toward the job also reported fewer occurrences of negative physical consequences (r = -.307). The null interrelationships proposed in HO14(a-c) between physical consequences and other outcome variables (i.e., overal l organizational commitment, overall job satisfaction or emotional exhaustion) were not supported and the null hypotheses were rejected. It was found that st atistically significant weak negative correlations between physical consequences and at titude toward the job (r = -.307), overall organization commitment (r =-.242), and overall job satis faction (r = -.292) exist. FSSP reporting high scores on attitude toward the job also report high scores on overall organizational commitment and overall job satisfaction. T hose reporting low scores on attitude toward the job also report low scor es on overall organizational commitment and overall job satisfaction. The proposed relationship betw een physical consequences and emotional exhaustion was moderately correlated at r = .493. In other words, as emotional exhaustion increases or decreases, negative phys ical consequences are also expected to increase or decease accordingly. Hypotheses 15(a-b) propose that overall organizational commitment is correlated with overall job satisfaction and em otional exhaustion. Hypothesis 15c proposes that overall job satisfaction is correlated with emotional exhaustion. Findings show that moderate to high statistically significant positive correla tions exist between overall organization commitment and attitude toward the job (r = .677) and overall organization commitment and overall job satisfaction (r = .611). FSSP reporting high scores on overall organization commitment also report hi gh scores on attitude toward the job and overall job satisfaction. A statistically significant mode rately correlated inverse

PAGE 182

169 relationship was found between overall organizational commitment and emotional exhaustion (r = -.460). A statistically significant inverse relationship was also demonstrated between emotional exhaustion and overall job satisfaction (r = -.594). FSSP reporting high scores for emotional exhaustion also report low scores for overall job satisfaction. A statistically significant weak negative (r = -.242) correlation exists between overall organizational commitment and physical consequences. In other words, FSSP scoring high on physical consequences score low on overall organizational commitment. FSSP scoring on high overall organizational commitment, score low on physical consequences. Therefore, findings shown in the correlations analysis (see Table 4.18) demonstrate that the hypotheses 15a,b,c were supported. H16: Attitude toward the job (ATJ) has a significant positive influence on FSSPs job performance (JP). Table 4.18 Outcome Correlations PC ATJ OOC OJS EE PC 1.000 ATJ -0.307 1.000 OOC -0.242 0.677 1.000 OJS -0.292 0.724 0.611 1.000 EE 0.493 -0.666 -0.460 -0.594 1.000 Significant at <.05 level (2-tailed) Impact of Job-Related Variables on Job Performance

PAGE 183

170 H17: Overall job satisfaction has a si gnificant positive influence on job performance. H18: Overall organizational commit ment has a significant positive influence on job performance. H19: Emotional exhaustion has a signi ficant negative influence on job performance. H20: Physical consequences have a significant negativ e influence on job performance. A necessary assumption for multiple regres sion and structural equation analysis is that high multicollinearity (>.30) is not presen t. Therefore, with the known presence of multicollinearity, examination of the impact of the five hypothesized outcome variables (i.e., attitude toward the job, overall job sa tisfaction, overall organizational commitment, emotional exhaustion, and physical conseque nces) on job performance occurs through five regression analyses instead of exam ining them in one regression model. Significance R2 Table 4. 19 Job Performance Regression Models Parameter Regression Coefficients Standard Error T-Value Physical Consequences -.162 .046 -3.540 .0001 .023 Attitude Toward the Job .176 .025 6.947 .0001 .082

PAGE 184

171 Overall Organizational Commitment .159 .023 6.765 .0001 .079 Overall Job Satisfaction .152 .021 7.119 .0001 .087 Emotional Exhaustion -.113 .031 -3.646 .0001 .024 Regression findings indicate that the five hypothesized outcome variables influence job performance in the hypothesized directions. All five regressions were statistically significant (<.0001) in the hypothesized direction. As evidenced in Table 4.19 above, Hypotheses 16 through 20 were accepted. As hypothesized, physical consequences and emotional exhaustion inve rsely influence job performance. When FSSP report low job performance, he or she also reports high scores for emotional exhaustion. FSSP reporting low scores on job performance report high scores on physical consequences. A significant positive relationship was found for job performance to attitude toward the job ( = .294), overall organi zational commitment ( = .290), and overall job satisfaction ( = .263). When FSSP report high scores on attitude toward the job, overall organizati onal commitment, and overall job satisfaction, then job performance was also reported as being high. Predicting Behavioral Intentions H21: Job performance (JP) has a significant negative influence on intention to switch positions within the organization (ITSP).

PAGE 185

172 H22: Intention to switch positions (IT SP) within an organization and intention to leave an orga nization (ITL) are correlated. H23: Job performance (JP) has a significant negative influence on intention to leave the organization (ITL). Table 4.20 Intention to Leave Regression Model Parameter Regression Coefficients Standard Error T-Value Job Performance -.632 .121 -5.222 Intention to Switch Positions Regression Model Parameter Regression Coefficients Standard Error T-Value Job Performance .122 -.343 -2.812 Significant < .05 The model conceptualized that intentions (i.e., intention to switch positions within the organization and intention to leave the organization) ha ve one antecedent (i.e., job performance). The findings from two regres sion analyses examining the job performance to intentions relationships support the hypothesized relationships in H21, and H23. Hypothesis 22 proposed a correlation between intenti on to switch positions and intention to leave the organization (r = .273, p < .05) which was supported by the data. The findings from the study show that FSSP may hold an intention to leave the organization simultaneously with an intention to switch pos itions within the organization. The above

PAGE 186

173 findings also demonstrate that it is possi ble for FSSP reporting low job performance to hold high intention to switch positions or hi gh intentions of leav ing the organization. Power analysis enables the researcher to differentiate between a good and bad model (McQuitty, 2004). If the power of the te st is low, the probability of making a Type II error of assuming no relationship exists when, in fact, it does. A pow er analysis is also critical with large sample sizes since large samples can magnify small specification errors leading to rejection of a good model (Type I error). In multiple regression analyses, power is the probability of detecting a statistically significant level of R2 or a regression coefficient at a specific level of significance for a specific sample size (Hair, Jr. et al., 1998). If the power level is low, the probability of failing to reject a bad model leads to accep ting a false theory (McQuitty, 2004). All regression analyses above obtained a significant R2 at a power of .80 for an effect size of .10, .05 [f2 = ( 2 / 12)]; [L = f2(n-k-1)] (Murphy & Myors, 2004). Power Analysis Chapter Summary Chapter 4 examined the analysis and resu lts of the studies de veloped in the first three chapters of this resear ch. The chapter accomplished this through analyses of: 1) indepth interviews, 2) pilot st udy data, 3) exploratory and c onfirmatory analyses, and 4) hypotheses testing using one-way ANOVAs multiple regression, and correlation analyses. Chapter 5 presents a discussion of the findings, implications (i.e., theoretical,

PAGE 187

174 measurement, management), and limitations of the study as well as proposed future research from this study.

PAGE 188

175 Research Purpose and Design CHAPTER 5 Discussion, Conclusions, and Implications Chapter 5 provides a discussion of the key findings of this empirical study, anticipated contributions (i.e., theoretical, measurement, and managerial), limitations of the study, and proposes future research direc tions. The findings presented in Chapter 4 and discussed here, lend support to the notion that the influence of self-management of emotionally-based display behavior and job stress on social service encounter performance are an important and re levant area for services research. This chapter is divided into eight secti ons: 1) research purpose and design, 2) development and assessment of new scale measurements, 3) discussion of empirical findings and conclusions, 4) implications for researchers and practitioners, 5) limitations of the study, 6) contributions of the researc h, 7) directions for future research, and 8) concluding comments. Summary of Research Objectives There are six objectives for the curren t study. One of the primary research objectives of this study is to empirically inve stigate the role of em otional intelligence and emotional labor within self-management of emotionally-based disp lay behavior in the social service delivery encounter from the Fr ontline Social Service Personnels (FSSPs)

PAGE 189

176 perspective. Findings of the study were consistent with the hypothesized model demonstrating that emotional intelligence me diates the relationship between emotional labor on job stress. Another objective is to empirically investigate the in fluence of two neglected constructs (i.e., perceived customer demands and emotional labor) as sources of job stress. This objective was realized by estab lishing that both perceived customer demands (coefficient = .105) and emotional labor (coeffi cient = .443) were, in fact, sources of job stress. The third objective is to re-examine role ambiguity, role conflict, and role overload as sources of job stress. Results of a regression analysis examining role ambiguity, role conflict, and role overload satisfied the third object ive of this study and demonstrated that these three constructs were significant (p<.01) sources of job stress. The fourth objective involves empirically investigating the me diating effects of job autonomy, emotional intelligence, and emotion-focused coping. Hierarchical regression results demonstrated mixed fi ndings of mediating relationships. Job autonomy, for example, did not function as a mediator of the impact of perceived customer demands on job stress. Hierarchical regression results showed that emotional intelligence mediated the imp act of emotional labor on job stress (p<.0001). Emotionfocused coping frequency did not mediate the impact of job stress on any of the five selected outcome constructs (i.e., attitude toward the job, physica l consequences, overall organizational commitment, overall job satisfact ion, or emotional exhaustion). However, the findings do demonstrate the existence of a partial mediating effect of emotion-focused coping effectiveness between job stress and overall organizational commitment and physical consequences (p<.05).

PAGE 190

177 The fifth research objective involves empi rically investigating the relationship of five selected outcome constructs on the job pe rformance construct. All five relationships with job performance were established through a series of regression analysis and found to be signifi cant (p<.0001). The final research objective of empirically investigating the rela tionships of social service encounter performan ce on both a currently unexplored outcome, intention to switch positions within the organization, and on intention to leave the organization was accomplished through regression analysis. Find ings from a regression analysis showed that job performance influences intention to switch positions (coefficient = -.343, t = 2.812, p = .005) and intention to leave (c oefficient = -.632, t = -5.222, p = .0001). Summary of the Research Study A three-phase research framework was used to develop the constructs, purify the measurements, and test hypothesized relati onships in the model. Phase I involved conducting eight in-depth interviews to assess the relevancy of the proposed manifested indicators representing the constructs. In Ph ase II, a pilot study survey was administered to 361 Frontline Social Service Perspective ( FSSP) across five organizations in which a direct cognitive framework was used to ascert ain the extent to which each indicator was related to its respective construct. The th ird and final phase invol ved administering the survey to a large sample (N = 2,500) to collect data to test the hypot hesized relationships in the model.

PAGE 191

178 Development and Assessment of New Scale Measurements The initial model was made up of ten cons tructs that did not have existing or adequate scales for this study. Therefore, new subjective self-report sc ales (i.e., attitude toward the job, emotional in telligence, emotional labo r, emotion-focused coping effectiveness of use and frequency of use, intention to switch positions within the organization, intention to leave the orga nization, overall job sa tisfaction, perceived customer demands, physical consequences, a nd role overload) were developed. Both open-ended and close-ended measures were exam ined through a pilot study using a direct cognitive framework to determine to what extent each indicator was related to its respective construct. Prior to developing the final study, appropriate sets of scale point descriptors were applied to each indicator and scale setups and instructions for filling out and returning the instrument were added to the survey instrument. The survey was also examined for physical appearance, (i.e., pr ofessionalism), readability, and ease of understanding of the indicators, instructions, and setups. Data from the final survey was subjected to exploratory factor analysis to assess the measurement properties of the latent constructs used in this study and to confirma tory factor analysis to demonstrate how the latent constructs were operationalized by their corresponding meas ured constructs, as well as assessing the validity and reliability of the measures. All scales in the study demonstrated construct validity (i.e., disc riminant and convergent). New scales are discussed below.

PAGE 192

179 Perceived Customer Demands This study provides a first attempt at scien tifically developing a scale to measure perceived customer demands that focuses on FSSPs perception of the intensity of the customers demands for overall quality and se rvice expectations in not-for-profit social service organizations. Unlike the Li and Ca lantone (1998) scale that employs endpoints of 1 = truly for none of my clients and 5 = trul y for all of my clients to measure perceived customer demands, the current study uses a ne w direct seven-point continuous scale with endpoints of 1 = not at all demanding to 7 = extremely demanding. Results were encouraging in that overall consistency was high ( =.895) and internal consistency among the eight indicators measuring each co mposite value (reliability = .906) suggested that these eight indicator items represent this latent constr uct and account for, on average, 59.7% of the variance. Emotional Labor Unlike previous attempts to measure em otional labor, which take either a job focus approach (e.g., Wharton & Erickson, 1993) or an emotion-focused approach (e.g., Brotheridge & Lee, 2003; Hochschild, 1983), th e current study operationalizes emotional labor as a multidimensional construct (i.e., effort and display). Analyses of the data for the current study show that emotional labor is multidimensional, but is made up of three dimensions instead of the two proposed dime nsions. The emotional labor scale in the current study consists of ten item s with three indicators measuring display expectation of behavior two measuring display of required behavior and five measuring effort. The emotional labor scale is considered reliable with an =.781 and composite reliability =

PAGE 193

180 .932, suggesting that these ten i ndicator items represent this latent construct and account for, on average, 96% of the variance. Emotional Intelligence Emotional intelligence is operationalized as an ability-based construct. However, because no appropriate scales were found to be available, scales were created based on the works of several researchers including Mayer and Salovey (1993) and Mayer, Salovey, and Caruso (2000). In the current study, emotional intelligence is proposed to have two dimensions (i.e. management of se lf and others and perception of self and others). However, through the factor analyses process, the new semantic differential scale consisting of seven indicator items loaded on three dimensions: management of self, perception of self, and management/perception of others with factors loading >.48. The scale was shown to be reliable with an =.766, composite reliability of .843, and accounts for, on average, 45.1% of the variance. Emotion-Focused Coping Based on the transaction theory of coping (Lazarus, 1966, Lazarus & Folkman 1984) as a dynamic process, a new scale was developed to capture both the frequency with which FSSP engage in emotion-focused coping and his or her perception of its effectiveness of use. Factor analyses revealed that three items loaded on the frequency of use dimension and three items loaded on the effectiveness of use dimension. The threeitem effectiveness of use dimension was measured with a symmetrical scale ( =.766) and frequency of use was measured on a frequency scale ( =. 774). Both scales reported

PAGE 194

181 composite reliabilities >.755 and average ex tracted variances of >.514. Because of difficulties in analyzing the data created by measuring one dimension with a symmetrical scale and one dimension with a frequency scale, the indicator items could not be combined and each mediating relationship was examined separately. Attitude Toward the Job Initially in Chapter 3, attitude toward th e job (ATJ) was operationalized as being an eight-item unidimensional composite cons truct measured by a unique set of bipolar semantic differential scales. Scale purifica tion and dimensionality procedures revealed that ATJ is best captured as a two-dimensional construct (i.e., overall attitude toward the job and attitude toward the boss) consisting of a six-item bipolar scale with a reliability of = .78, a composite reliability of .81, and indicator items accounting for an average 45.5% of the variance. Intention to Switch Positi ons Within the Organization Prior to the current study, scales to measure the intention to switch positions within the organization constr uct did not exist. As proposed by Drolet and Morrison (2001), careful attention was give n to the development of indi cator items to avoid drifting away from the original con ceptual definition of intenti ons and to avoid redundancy. Although originally proposed as a three-item scale, the final purifie d intention to switch scale consisted of only two items. The test results revealed th is new scale holds a reliability of = .86, composite reliability = .89, a nd the indicator items accounted for 79.5% of the variance.

PAGE 195

182 Intention to Leave the Organization The FSSPs likelihood to leave the organi zation was obtained by asking the FSSP about his or her likelihood to leave the organization within a timeframe of 12 months. The three item scale was very reliable (i.e., =.890, composite reliability = .887, with indicator items accounting for, on average, 73 % of the variance) a nd suggested that the intention to leave the organi zation construct was captured th rough the selected three-item scale. Overall Job Satisfaction In order to develop an overall measure of the FSSPs general affective reaction to his or her job without referencing any specific facet of job satisfaction, a scale was developed and assessed based on the Rice, Gentile, and McFarlin (1991), a revised version of the Quinn and Shepard (1974) measur e. It was, however, determined that the Rice, Gentile, and McFarlins (1991) scale type and descript ors were not a ppropriate to measure overall job satisfaction as conceptualized in the current study. Therefore, for the current study, a four-item semantic differential scale was developed, assessed, and found to be reliable (i.e., =.861, composite reliability = .852, with indicator items accounting for, on average, 59.4% of the variance). Physical Consequences A list of 18 physical symptoms from the Spector (1987) causal indicator scale was examined in the current study. After extens ive scale purification, an eight-item scale measuring three dimensions (i.e., digestive, emotional, and cardiovascular) was found.

PAGE 196

183 Role Ambiguity, Role Conflict, and Role Overload as Antecedents to Job Stress Reliability of the eight-item seven-point fr equency of occurrence scale ranging from 1 = never to 7 = always was found to be reliable at =.802, composite reliability = .871, with indicator items accounting for, on average, 49.3% of the variance. Role Overload It has been established that FSSP have more work than can be completed in a normal workday (Gruskin, 2003). For this st udy, a scale to measure role overload for FSSP was based on the revised Quantitative Wo rkload Inventory (QWI) scale (Spector, Dwyer, & Jex, 1988). The revised QWI (1988) eight-item scale lacked uniformity of descriptors and did not lend itself well to mu ltivariate analysis. Thus, a new frequency scale was developed with end descriptors of 1 = never (0%) to 7 = always (100%) in which the respondent answered as to how often in a typica l workday each event occurs. The five-item role overload scale used in th e current study was found to be reliable at = .89, composite reliability = .870, with indica tor items accounting for, on average, 62.1% of the variance. This finding was comparab le to the QWI reliability found in the 1998 job stress study by Spector and Jex (1998) with reliability of = .82. Discussion of Empirical Findings and Conclusions This study re-examined three known sources of job stress (i.e., role ambiguity, role conflict, and role overload) from th e perspective of FSSP a nd found role ambiguity (coefficient = .280, t = 3.821, p = .0001), role conflict (coefficient = .261, t = 7.099, p =

PAGE 197

184 .0001), and role overload (coefficient = .214, t = 6.381, p = .0001) contributed to FSSP job stress which indirectly influenced service encounter performance through its influence on outcomes such as emotional exha ustion, physical consequences, etc. Ereras (1989) study suggests that largeness and complexity of an organization may enhance the existence of these three sour ces of job stress. Although largeness and complexity (i.e., bureaucratically laden) were not directly tested in the curr ent study, it is interesting to note that one common denominator across all of the organizations in th is study is that the organizations are complex (i.e., bureaucratically laden) in that, all of the organizations function under heavy government regulations from several different local, state, and federal governments, as well as regulations from internal Boards of Directors. Role ambiguity is the perception of uncerta inty about what tasks are involved in carrying out a job. We know from previous re search that sources of job stress, such as role ambiguity, are an inherent characterist ic of many jobs for both professional and nonprofessional service providers (Bettencourt & Brown, 2003). In the current study, data were collected from both professional (i.e., degreed >95% of the sample) and nonprofessional (non-degreed <5% of the sample ) social service personnel across various types of jobs (i.e., case workers, case managers family counselors, social workers, etc.). Although Bettencourt and Brown (2003) did not em pirically test the relationship between role ambiguity and job stress, they empirically tested relationships assuming role ambiguity to be a source of job stress. Th e findings of the current study show that role ambiguity directly influences job stre ss (coefficient .280, t = 3.821, p = .0001). This finding supports the notion of the Bettenc ourt and Brown (2003) study that role ambiguity is a source of job stress.

PAGE 198

185 Role conflict is the degr ee of perceived incongruen ce or incompatibility of organizational expectations associated with job performance. Role overload is the imbalance between required job tasks and time allocated to complete those tasks. The Lait and Wallace (2002) study of the work environment of 514 human service workers also found that role conflict ( = .17) and work overload ( = .22) directly influence job stress. The findings of the current study across 260 FSSP in the not-for-profit social services sector of the influence of role conflict (coefficient = .261, t = 7.099, p = .0001) and role overload (coefficient = .214, t = 6.381, p = .0001) on job stress support the findings of Lait and Wallace (2002). Hypothe ses 1, 2, and 3 were supported by the data. As suggested in previous research (Goolsby, 1992), it is apparent from the findings of the current study that role ambi guity, role conflict, and role overload (r2 = .309) are not all inclusive in explaining job stress. Two other sources of job stress hypothesized in this study contributed to job stress: perceived customer demands ( = .104) and emotional labor ( = .350). These two additional sources of job stress are discussed in the following section along w ith hypothesized mediati ng relationship of job autonomy and emotional intelligence. Perceived Customer Demands as an Antecedent to Job Stress Based on the centrality of the role of FSS P and the customer as co-producers of the service offering, perceived customer demands are proposed as a source of job stress for frontline service personnel (Chung & Schne ider, 2002). Perceived customer demands are the perceptions of customers requirem ents and expectations for service quality, reliability, and deliverability. Perceived customer demands have previously had very

PAGE 199

186 Emotional Labor as an Antecedent to Job Stress In an attempt to bring consistency to the service encounter, many organizations attempt to control emotionally -based behavior displayed by frontline service personnel. Emotional labor involves the effort put forth by the frontline service personnel to display emotionally-based behavior deemed appropria te for a given situation (Morris & Feldman, 1996). To date, emotional labor has not been empirically examined as to its influence on the job stress construct. Findings from the current study show that emotional labor significantly influences job stress (coefficient .443, t = 7.928, p = .0001). In addition, this finding is an important first step in th e attempt to examine the mediating relationship of emotional intelligence between emoti onal labor and job stress hypothesized in H5. little empirical examination. Wang and Ne temeyer (2002) assess the salespersons perception of the proportion of his or he r customers who are highly demanding of product/service quality and reliability. Ho wever, to the researchers knowledge, perceived customer demands have not previous ly been empirically examined as to their influence on job stress. Through the investigation of Hypothesis 4 involving the media ting influence of job autonomy between perceived customer de mands and job stress, it was demonstrated that perceived customer demands is a weak source of FSSP job stress (coefficient = .105, t =2.318, p = .021). Further attention is give n to this relationship in the Mediating Effects of Emotional Intellig ence and Job Autonomy on Job Stress section to follow.

PAGE 200

187 Mediating Effects of Job Autonomy and Emotional Intelligence on Job Stress Interestingly, the hypothesized mediati ng influence of job autonomy between perceived customer demands and job stress (H4) was not found. Although perceived customer demands was shown to increase job stress (coefficient = .105, t = 2.318, p = .021), it did not significantly influence j ob autonomy (p = .143). Therefore, job autonomy does not have a mediating effect on reducing the influence of perceived customer demands on job stress. Nevertheless, the finding that job autonomy (coefficient = -.378, t=-9.579, p=.0001) has a di rect inverse influence on job stress for FSSP was significant. In other words, if FSSP have frequent control over job-related ac tivities, then unrelated to perceived customer demands, FSSP experience less job stress. One possible explanation for the weak influence of perceived customer demands on job stress and the lack of a significant finding for Hypothesis 4 may be found in the sample for the current study. The sample was made up of > 95% professional FSSP. FSSP regularly make decisions regarding customer demands fo r which deviation from their training and education may not be an option. That is, professional FSSP are highly educated (i.e., 95.3% have college degrees with 59.1% holdi ng masters and 27.2% of those with college degrees holding postgraduate degrees) and highly trained (i.e ., 61.3 average hours of training with current employer) to deal with demanding people (Chao & Henshaw, 2003). Thus, training and education may prove to moderate the relations hip between perceived customer demands and job stress. In this study, two types of self-managem ent of emotionally-based behaviors were examined across two levels proposed in aff ective event theory as interpersonal (i.e., emotional labor) and between persons (i.e ., emotional intelligence) previously

PAGE 201

188 unexamined in the services literature. The mediating role of emotional intelligence between emotional labor and job stress yielded interesti ng results. The current study offers evidence that substantiates the hypothesized theoretical argument regarding emotional intelligences medi ation role in the job stre ss process (Hypothesis 5). Therefore, the direct effect of emotional labor on job stress (coefficient = .443, t = 7.928, p = .0001) was reduced significantly (coe fficient = .304, t = 7.563, p = .0001) when the mediating effect of emotional intelligence was present. In other words, an emotionally intelligent FSSP is equipped to sense, know, and display appropriate emotionally-based behavior during the service encounter, elim inating the need to act out an imposed emotionally-based display behavior, thus re ducing the impact of emotional labor on job stress. Constructs Direct Influence of Emotion-Focused Coping The coping process arises during the service encounter and changes the relationship between job stress and outcomes (i.e., attitude toward the job, physical consequences, overall organizational commitme nt, overall job satisfaction, and emotional exhaustion). Some research ers suggest that frontline service personnel engage in emotion-focused coping (i.e., passive or reactive strategies such as denial, acceptance, behavioral, and/or mental dise ngagement) in an attempt to reduce the effort to control experienced-stress coming from the work environment (Goolsby, 1992). One unexpected but interesting finding for Hypothesis 7b was that the frequency with which FSSP engaged in emotion-focused coping was not in fluenced by his or her degree of emotional intelligence (coefficient = .075, t = .926, p = .055) However, for Hypothesis 7a, it was

PAGE 202

189 found that when FSSP do engage in emotionfocused coping, the higher the emotional intelligence of FSSP, the greater his or her perception that emotion-focused coping is effective in reducing his or her job stress (coefficient = .141, t = 2.284, p = .023). It is interesting that the findings of th e current study show that highly emotional intelligent FSSP, as well as low emotional inte lligent FSSP, are not significantly different as to the frequency with wh ich they engage in emoti on-focused coping. However, findings do show that highly emotionally in telligent FSSP perceive emotion-focused coping as a more effective means of ma naging emotions than do FSSP with lower emotional intelligence. It may be surmised from these findings that although emotionfocused coping may not function as the pr imary means of managing emotions for emotionally intelligent FSSP, it is perceive d to be an effective way of managing emotions. Emotion-Focused Copings Mediating Ro le on Selected Outcome Constructs The current study also empirically inve stigated the mediating effect of the frequency with which FSSP engaged in emoti on-focus coping between job stress and the constructs including physica l consequences, attitude toward the job, overall organizational commitment, overall job satis faction, and emotional exhaustion. The current study also examined the mediating e ffect of emotion-focused coping effectiveness between job stress and these same five selected outcome constructs.

PAGE 203

190 Impact of Emotion-Focused Coping Effectiveness In cases where FSSP perceived that emo tion-focused coping was effective for them, it was shown that emotion-focus coping effectiveness mediated the impact of job stress on the two outcome constructs: overa ll organizational commitment (z-score = 2.020 p<.05) and physical consequences (z -score = 2.502 p<.05). This should be considered a significant finding in that the cu rrent study is the first empirical assessment of the mediating relationship of emotion-fo cused coping effectiveness between job stress and the five selected outcome constructs. Hypotheses H11a and H12a were supported. Hypotheses H8aH10a were not supported Impact of Emotion-Focused Coping Frequency Overall, emotion-focused coping fre quency results are less than stellar (Hypotheses 8b-12b). Findings of the study show that the frequency with which FSSP engage in emotion-focused coping does not me diate the influence of job stress and any of the proposed outcome constructs. One possibl e explanation for these findings may stem from the fact that job stress develops gradua lly over time masking its intensity. Although emotion-focused coping strategies give FSSP a temporary sense of reprieve, generally, in the long run, these strategies have little effect in reduc ing job stress (Goolsby, 1992). Therefore, FSSP may not sense the need to engage in emotion-focused coping until the intensity of the job stress seems significant e nough to warrant seeking relief. Hence, the hypothesized relationships were not supported.

PAGE 204

191 Interrelationships Among the J ob Related Outcome Constructs The absence of non-significant interrelati onships of the five selected outcome constructs, HO13 HO14, were rejected and H15a,b,c were supported. Interrelationships of attitude toward the job with other outcome constructs in this study have not previously been empirically examined. Interrelationshi ps of physical consequences with other outcome constructs in this study have also not been previously em pirically examined. These findings should be considered an importa nt first step in understanding the behavior of these constructs. Causality was not examin ed in this study, so no conclusions can be made if one or both of thes e constructs (e.g., attitude toward the job or physical consequences) are antecedents for other outcome c onstructs. It is also interesting to note that, as expected, attitude toward the job, as with other attitudinal constructs in the model such as overall job satisfaction and overall organization commitment, have a negative correlation with emotional exha ustion. In other words, attitude toward the job, like overall organizational commitment and overa ll job satisfaction, had a negative impact on emotional exhaustion. Job-Related Outcome Constructs (Antecedents) Impact on Job Performance The finding that strong interrelationships (multicollinearity) exis t between the five selected outcome constructs made it difficu lt to collectively assess their impact on job performance. Therefore, five separa te regression analyses were performed.

PAGE 205

192 Overall Job Satisfaction For the current study, overall job satisfaction was found to have a statistically significant direct positive influence (c oefficient = .152, t = 7.119, p = .0001) on job performance. Fisher (2003) claims that the way research is conceptualized and operationalized contributes to findings that are contrary to the common-sense theory that overall job satisfaction leads to job pe rformance (p. 754). The current study supports the common-sense theory which theorizes th at overall job satisfaction influences job performance and not vice versa. In othe r words, the current study suggests that FSSP whose jobs compare favorable to his or her needs, wants, and expectations for that job can be expected to perform better on the j ob than for those whose expectations, needs, and wants are not met. Hypothesis H17 was supported. Attitude Toward the Job Attitude toward the job did perform as proposed in Chapter 2 and had a significant positive influence (coefficient .176, t = 6.947, p = .0001) on job performance. FSSPs job performance at the service encounter may be the only surrogate indicator for service quality (Shostack, 1977). Therefore, a ny finding of a source of variance provides useful information. Hy pothesis 16 was supported. Overall Organizational Commitment Because controversial empirical findi ngs as to the relationship between organizational commitment and job perfor mance exist, this study re-examined the relationship (e.g., Brown & Peterson, 1993). An interesting finding from the current

PAGE 206

193 study was that, as found in the Bettencourt and Brown (2003) study in the commercial sector, frontline service personnel in the not-f or-profit sectors job performance was also influenced by FSSPs overall organizational commitment. The finding of the current study (coefficient = .159, t = 6.765, p = .0001) de monstrated support of Hypothesis 18 and suggests that in an attempt to help th e organization achieve its objectives, a FSSP highly committed to the organization will be motivated to perform well on the job. Emotional Exhaustion Wright and Cropanzano (1998) find that emotional exhaustion is negatively correlated with job performance (r = -.27, p< .05). The findings of this study show causality (coefficient -.113, t = -3.646, p = .0001) not shown in the Wright and Cropanzano (1998) study. Hypothesis 19 was s upported. These findings suggest that as emotional exhaustion increases for FSSP, that his or her job performance will decline. Physical Consequences DeJonge et al. (2001) clai m that poor health of any kind may cause bad working conditions and Madsen (2003) proposes that improving the physical wellness of an individual influences his or her job performance. H ypothesis 20 was supported by the data. The claims made by DeJonge et al. (2001) and Madsen (2003) were supported with the findings of the current empirical study demonstrating that as the occurrence of physical consequences goes up, the level of job performance goes down (coefficient = .162, t = -3.540, p<.0001). As stated earlier, al though this finding is weak, any finding of a source of variance provides useful information.

PAGE 207

194 Job Performance as a Predicto r of Behavioral Intentions The current study shows that as expect ed for Hypothesis 23, job performance influences intention to leave the organiza tion. It is interesting to note that job performance also significantly influences a FSSPs intention to switch positions within the organization (Hypothesis 21). But, probabl y a more interesting finding is support for Hypothesis 22 where it is found that FSSP holding intentions to switch positions within the organization may also hold intentions to leave the organization and vice versa ( r = .273, p <.05). This phenomenon could mean that a FSSP switching positions may also be waiting for something better to become ava ilable outside the orga nization. Switching behavior may add additional cost to the organization in the fo rm of training costs for the FSSPs new position within the organization, for someone to fill the position he or she left behind, and then to fill another positi on vacated by the same FSSP when leaving the organization after temporarily switching positio ns within the organization. Maybe even more important is the cost of switching or l eaving intentions to the customer. Customers seeking social services, voluntar ily or involuntarily, are exp ected to reveal personally sensitive information with FSSP. Changing FSSP requires the customer to reveal often highly personal information to a new FSSP. If the customer is uncomfortable in confiding personal information to the new FSSP, he or she may be reluctant to work with the new FSSP and/or could leave the organiza tion from which they are seeking help, possibly foregoing or postponing much n eeded assistance (Andreasen, 1995).

PAGE 208

195 Implications for Researchers and Practitioners Although in any particular job several s ources of job stress may be present, the cumulative influence of job stress cannot be ignored. Typically, however, the ability to reduce the influence of any of these sources is not under the direct c ontrol of most FSSP. If FSSPs job stress from these sources is to be reduced, management must examine each source individually, as well as collectivel y. In many cases, where several different government regulatory bodies dictate organizational behavi or, role ambiguity, role conflict, and role overload, it may be diffi cult for even management to reduce or eliminate job stress. Thus, it may be necessary for management to pr ovide assistance in coping with these sources of job stress. With an anticipated increase in serviceoriented employment opportunities in the U.S. of 5.1 million between 2000 and 2010, and with over half of those jobs being within the health, business, and social services indus tries (Hecker, 2001), focusing on the critical issues underlying customers perceptions of se rvice providers, such as service encounter performance, has gained renew interest among researchers (Chenet, Tynan, & Money, 1999). Based on the findings in this study that one form of self-management of emotionally-based display behavior (i.e., emotional labor) increases job stress while another (i.e., emotional intelligence) redu ces the influence of emotional labor on job stress, self-management of emotionally-based display behavior should be of particular interest to FSSP and management of social service organizations, especially in cases where FSSP are entrusted to make life and deat h decisions at the se rvice encounter such as those dealing with abused and ne glected children. A lthough implementing emotionally-based behavior display rules has been proposed to bring consistency to the

PAGE 209

196 service encounter (Morris & Feldman, 1996), the current findings bri ng in to question the value of such a decision. For example, this study indicates that in an attempt to bring consistency to the service encounter, usi ng a strategy involving implementation of emotionally-based display rules for FSSP incr eases job stress. Job stress is shown to reduce the effectiveness of FSSP by increasing physical consequences and emotional exhaustion for which the influence of engagi ng in emotion-focused coping is minimally effective. Additionally, job stress decr eases the effectivenes s of FSSP by negatively influencing overall job satisfa ction, attitude toward the job, and overall organizational commitment for which emotion-focused coping has no effectiveness. Therefore, implementing strategies such as alignment of all subsystems (i.e., human resource management, operations, marketing, etc.) in an attempt to foster a service focus consistent with customer expectations is e xpected to reduce job st ress, as well as the influence of job stress on emotional exha ustion, physical consequences, overall job satisfaction, attitude toward the job, and ove rall organizational commitment. Findings of the current study indicate that emotional e xhaustion, physical cons equences, overall job satisfaction, attitude toward the job, and ove rall organizational commitment influence job performance. Thus, by reducing the influen ce from various sources of job stress, the influence of job stress on the above outcome variables also improves job performance (i.e., service quality, reliability, and respons iveness). These findings advance knowledge of marketing for the service sector by examin ing issues of value cr eation in the service delivery encounter. Because both employees and customers are involved in creating value in a service encounter, organizations cannot afford to mistreat either. The institution of strategies that increase job stress, thereby ne gatively influencing the FSSP

PAGE 210

197 The current study takes a fr esh look at job stress and emotionally-based display behavior and provides in sights for researchers and practitione rs as to the role of emotions (i.e., emotional intelligence, emotiona l exhaustion, emotion-focused coping, and emotional labor) in relationship to job stress. The study also offers new findings that are exploratory in nature and are anticipated to evoke future research. For example, the finding of no mediating influence of job au tonomy between perceived customer demands and job stress. Findings of the study also reveal the possibility for examining relationships in this study across other orga nizational and job types in both the not-forprofit service sector and the commercial sector. and his or her service delivery, negatively impacts both the value and the reputation of the firm (McAlexander et al., 1994). Howe ver, through the reduction of job stress, service firms may augment their care services adding value to the firm. Limitations of the Study When considering the interpretation of this study, certain limitations should be considered. The study relies on subjective self-r eport measures that may reflect response bias. However, steps taken during data collection to reduce response bias were anticipated to enhance generalizability. For example, item reversals were dispersed throughout the study, respondents were guarant eed anonymity, and responses are mailed back directly to the researcher in an attempt to reduce social responding. Another concern should be kept in mind. There may be subtle artifacts due to sampling bias. A large portion of the sample came from the National Associat ion of Social Workers-FL which consists predominately of professional FSSP, and therefore, may not adequately

PAGE 211

198 represent the non-professional FSSP populati on thereby limiting generalizability. Readers should also note that several significant relationshi ps reported in this study are relatively weak. This suggests that some stat istically significant re lationships may have minimal practical significance and warrant further empirical study such as the relationship between the selected outcome cons tructs and job performance. For example, the relationship between emotional exhaustion (coefficient -.113, t-3.646, p =.024) and job performance warrants further investigation. Contributions of the Research For more than 20 years, job stress in the U.S. has continued to increase significantly (Princeton Survey Research Association, 1997). Sources of job stress and reactions to job stress have empirically been shown to vary across types of employment and job levels (Narayanan et al., 1999). Th is study makes a signifi cant contribution to the services literature by examining the antece dents and consequences of job stress in the not-for-profit social services sector focu sing on self-management of emotionally-based display behavior. To date, the relationship of emotionally-based disp lay behavior to job stress has received little empiri cal investigation in the not-for-profit sector. Therefore, this study makes a contribution by concep tualizing, empirically examining, and evaluating the influence of three types of self-management of emotionally-based behaviors (i.e., emotional labor, emotional intelligence, and emotion-focused coping) on job stress. In addition, findings of the curr ent study provide insights for other similar service encounters where imposed emotiona lly-based behavior display rules are employed in an attempt to bring consiste ncy to service encounter performance.

PAGE 212

199 Contributions come from conceptualizing a nd empirically investig ating the previously unexamined direct influence of emotional in telligence on emotion-focused coping and the mediating influence of emoti onal intelligence between emoti onal labor and job stress. Emotion-focused coping is also conceptuali zed and empirically examined as to its mediating influence between job stre ss and selected outcome constructs. Scale Measurement and Methodology Contributions A significant contribution of the current study comes fr om scale development. Specifically, this study makes an important contribution by developing and assessing new scales for ten constructs (i.e ., attitude toward the job, emo tional intelligence, emotional labor, emotion-focused coping, intention to switch positions with in the organization, intention to leave the organi zation, overall job satisfaction, perceived customer demands, physical consequences, and role overload) as well as adapting and psychometrically testing existing scales for seven constructs (i.e., role ambiguity role conflict, job autonomy, job stress, emotional exhaustion, jo b performance, and overall organizational commitment). All scales in the study demons trate reliability and c onstruct validity (i.e., convergent and discriminant). Therefore, th e development of reliable (.806 to .932) and valid (i.e., construct) scales, where such scales did not exist, is considered a significant contribution to services research. See th e Development and Assessment of New Scale Measurements section above beginning on page 176 for a detailed discussion of ten new scales developed and tested as part of th is study. Although these scales are developed and tested in the social service sector, they present high reliabilities and construct validity and, therefore, could be app lied in a non-social service environment service sector.

PAGE 213

200 Theoretical Contributions Contributions to theory are also an important part of this study. Findings of this study offer several important theoretical contri butions for the not-for-p rofit service area. Affective event theory (AET) purports that events in the workplace generate emotional reactions (Weiss & Cropanzano, 1996). Th e current study contributes by examining constructs at various levels of AET. The study presents fi ndings of relationships for constructs at three of the five levels of AET: between person (i.e., emotional intelligence), interpersonal interactions (i.e., emotional labor and emotion-focused coping), and within-person (i.e., emotional exha ustion). The current study also presents findings across levels of AET for a previ ously untested mediating relationship of emotional intelligence between emotional labor and job st ress of FSSP. The current study extends AET by finding statistically sign ificant relationships across the between person (i.e., emotional intellig ence) and interpersonal intera ctions (i.e., emotional labor and emotion-focused coping) levels of A ET, as well as rela tionships across the interpersonal interactions and within-person levels of AET. This study contributes to role theory by offering fi ndings from an empirical examination of the previously untested infl uence of perceived customer demands as a source of job stress and the proposed media ting relationship of j ob autonomy between perceived customer demands and job stress. Making the connection between perceived customer demands and job stress, as well as emotional labor and job stress, is important to role theory. The finding that perceived customer demands and emotional labor are additional sources of job stress for FSSP provides another link between the FSSP and

PAGE 214

201 Managerial Contributions organizational level of research and scientific inquiry by providing clarity to the summation of the mechanism through which systems confront their members as individuals, as well as building blocks of the social system (Kahn et al., 1964). The current study offers management a be tter understanding of a complex issue, job stress, by examining two new sources of job stress (i.e., percei ved customer demands and emotional labor) through the eyes of FSSP. In addition, this is the first study in the not-for-profit services area that investigates the mediating influence of job autonomy between perceived customer demands and j ob stress and the mediating influence of emotional intelligence between emotional labor and job stress. It is also the first such study to examine the mediating effects of emotion-focused coping on the relationship between job stress and five selected outco me constructs. Even though findings of the study did not demonstrate that job autonom y mediates between perceived customer demands and job stress, findings of the st udy show that job autonomy by itself reduces job stress for FSSP (coefficient = -.378, t= -9.579, p=.0001). The current study also demonstrates that emotional intelligence re duces job stress coming from emotional labor and that engaging in emotion-focused coping frequently has no effect in reducing job stress. The study also shows th at the mediating influence of engaging in emotion-focused coping is limited to having an influence between job stress and two outcomes (i.e., overall organizational commitme nt and physical consequence). Many organizations have come to realize th at consistency at th e service encounter is important (Hochschild, 1983). Intuitively, unregulated job stress during a social

PAGE 215

202 service delivery encounter can be counterprodu ctive for all the participants. Therefore, the service encounter is no place for FSSP to demonstrate inappropriate emotionallybased behavior. FSSP are in close and consta nt proximity to customers who often are unwilling or reluctant to seek professional help in making needed behavioral changes. Customers of social service providers often are angry or frustrated and tend to make unreasonable demands on FSSP (Kotler et al., 2002). The current study provides insights into the consequences of bringing consistency to the service encounter by controlling the FSSPs emotionally-based display behavior th rough instituting adherence to emotionallybased display behavior rules. FSSPs job pe rformance at the service encounter may be the only surrogate indicator for service quality in an organization where credence attributes are high (Shostack, 1977). Therefor e, since job stress (i.e., role ambiguity .31 and role conflict -.14) is shown to have a dir ect inverse relationship on job performance (Singh, 2000), poor job performa nce may indicate poor service quality. Because social service customers are expected to trust FSSP with something as important as a behavior change, job perfor mance is an important indicato r of service quality. In the current study, role ambiguity, role conflict, role overload, perceived customer demands, and emotional labor are all sources of job st ress (p <.01). Therefore, by adding to the FSSPs job stress, emotional labor can be expected to negatively influence job performance and may reduce the customers pe rception of service qua lity (Singh et al., 1994). Unlike other sources of job stress such as role ambiguity, role conflict, and role overload for which it may be difficult for ma nagement to reduce their influence on job stress in a bureaucratically laden organizati on such as many social service organizations, job stress from adherence to or avoidance of organizationally sanctioned emotionally-

PAGE 216

203 based behavior display rules (i.e., emotional labor) may be reduced by not instituting or sanctioning emotionally-based behavior display rules. Some previous research suggests that ro le ambiguity, role conflict, and role overload are not significant s ources of job stress (Lord, 1996; Narayanan et al., 1999). However, the researchers contention that th ese are sources of job stress relevant for FSSP is supported in the findings of the curr ent study. Therefore, attention to these sources of job stress should be addressed by the management of FSSP. For example, it may be possible for management to reduce job stress from role ambiguity by defining more explicitly what is expected of FSSP on the job. However, it is unlikely that job stress will ever be totally eliminated for FSSP. It is well known that much of the influence of job stress is under the direct co ntrol of management and not of the FSSP. Where job stress cannot be eliminated or reduc ed, the research suggests that management may find that implementing intervention program s (i.e., supportive behavior that provides FSSP with information to help him or her cope with job stress) to be effective (Goolsby, 1992). However, the current study shows that me rely attempting to escape the influences of job stress through emotion-focused coping is not very productive ov erall. With this finding in mind, development of intervention providing social support programs (i.e., emotional support) for FSSP may not be the solu tion. It is, however, apparent from the study that emotional intelligence is effective in circumventi ng the influence of job stress created by emotional labor. In light of th is finding, it is suggested that management approach a decision to institute emotionally-based behavior display rules carefully, weighing the benefits of consistency at the service encounter to the influence of job stress caused by emotional labor on outcomes such as physical consequences, emotional

PAGE 217

204 exhaustion, attitude toward the job, overall organizational commitment, and overall job satisfaction. Sales force, customer service representatives, retail store managers, and many others must demonstrate self-control, conscientiousness, empathy, and service orientation (Goleman, 2000). In these and ot her jobs where a strong customer orientation is crucial, emotional intelligence is an e ssential ingredient for a successful service encounter and the cost to the employee (i.e., emotional exhaustion and physical consequences) and to the organization (i.e., ne gative influence of job stress on attitude toward the job, overall organizational commit ment, overall job satisfaction) from sources of stress such as emotional labor may outwei gh the benefit of consis tency at the service encounter (Goleman, 2000). Emotional intel ligence is an FSSPs lear ned ability to selfmonitor his/her own emotions and the feelings and emotions of customers resulting in useful information for guiding his or her own thinking and actions (Salovey & Mayer, 1990). Therefore, alternatives to imposing em otionally-based behavi or display rules are to test the emotional intelligence of poten tial new hires, hire FSSP that meet the organizations standards of emotional intellig ence, and/or offer training programs aimed at increasing the FSSPs emotional intelligen ce. Certainly selec tion of adaptable, emotionally aware, optimistic, and socially skilled individuals has potential benefits (Zeidner, Matthews, & Roberts, 2004). Thus, testing potential new hires for emotional intelligence ability and providing training se ssions to increase employees emotional competencies must focus on the qualities most important for the job and on how these qualities will influence the organization in the short and long run. It has also been suggested that creating an encouraging and supportive e nvironment for intervention,

PAGE 218

205 modeling desired skills, and providing ongoi ng emotional support to be effective in increasing emotional intelligence levels of employees (Cherniss & Goleman, 2001). It is also apparent from the findings of this study that many FSSP are highly educated and highly trained. A natu ral assumption here is that as human capital good FSSP are a valuable asset to the organization. It is in the interest of the organization to keep good FSSP. We know from previous research that intentions play a central role in turnover models (Johnston & Futrell, 1989; Lee & Mowday, 1987; Sager et al., 1989; Sager & Menon, 1994) and should not be igno red by management. Findings from the current study offer valuable managerial insights that previously had not been demonstrated empirically such as the relatio nship between job perf ormance and intention to switch positions and the positive interr elationship between intentions to switch positions within the organization and in tentions to leave the organization. Directions for Future Research This study provides many avenues for future research in both the not-for-profit service area and the commercial sector. Ba sed on statistically significant, although sometimes weak relationships between constructs in the study, fu ture research of relationships is warranted. For example, the relationship between perceived customer demands and job stress is exploratory in nature and should be considered a first step in investigating the relationship. Due to the nature of the FSSPs j ob (e.g., social service providers) in the current study, FSSPs sensitiv ity to what constitutes customer demands may be another contributor to the weak relationship between perceived customer demands and job stress. Furt her research of the relations hip of perceived customer

PAGE 219

206 demands and job stress across other job types is suggested such as frontline service providers in health services and law enforcement. A central focus of this study is to gain a better understanding of the role of selfmanagement of emotionally-based display behavior (i.e., emotional intelligence, emotional labor, emotion-focused coping, and emotional exhaustion). The relationships found in the current study i nvolving self-management of emotionally-based display behavior show promise in clarifying the ro le of self-management of emotionally-based display behavior in a social service organization. An exam ple is the finding that the direct effect of emotional labor on job stress (coefficient = .443, t = 7.928, p = .0001) is reduced significantly (coefficient = .304, t = 7.563, p = .0001) when the mediating effect of emotional intelligence is present. Resu lts of this study involving the influence of emotionally-based display behaviors for FSSP should be examined for generalizability across other service industries. It is also apparent from the findings of the study that valuable pieces of the job stress puzzle are still missing. Many FSSP deal with clients in vary ing locations and are away from the home office a great deal of the time opening up the possibility that workplace isolation may be a potential source of job stress worthy of investigation in the not-for-profit service area. According to FSSP interviewed in the current study, they work many long hours (e.g., Gruskin, 2003) and are often on call weekends and nights. Therefore, work-family conflict may be a potential source of job stress for FSSP worthy of investigation. Future studies should also be conducted focusing explicitly on the relationship between emotional labor and job stress. If future findings support that emotional labor

PAGE 220

207 increases job stress, which is shown to have negative consequences for both the emotional (i.e., increased emotional exhaus tion) and physical (i.e ., increased physical consequences) well-being of FSSP, then whethe r social service organizations as stewards of the publics well-being and trust should make such business decisions as imposing emotionally-based behavior display rule s becomes an important consideration. Because of the large portion of pr ofessional FSSP in the current study, a comparison study of non-professional FSSP s hould be conducted to determine if relationships in the study in which emoti onal intelligence is hypot hesized holds across both types of FSSP. A study could be conducte d to study the influence of the emotional intelligence training programs on FSSPs emotional intelligence. The impact of the programs isolated and causal in ference could be made as to the value of such programs in increasing FSSPs emotional intelligence. Additionally, it is recognized that not all frontlin e service personnel in not-forprofit organizations are social service providers. Therefore, it would be interesting to examine the relationships found in this st udy across other not-for-p rofit organizations which have previously been iden tified as stressful in nature (i.e., health services and law enforcement), but for which frontline service personnel are not prim arily social service providers. In law enforcement, the core se rvice is providing prot ection for the public. Police officers may also provide a supplementa l service that adds value to the service encounter by counseling victims of crime. Health service providers provide social support as a supplemental service, yet the primary function of his or her job is providing health care. In both cases, job stress has pr eviously been shown to be high and selfmanagement of emotionally-based display behavi or is vital to the FSSPs safety as well

PAGE 221

208 as to the successful serv ice encounter (Hecker, 2001; Princeton Survey Research Association, 1997; Goleman 2000). Attitude toward the job and physical cons equences have not been examined prior to the current study as to their relationship with the other ou tcome constructs selected for this study. Although the current study demonstrates directionality (i.e., negative versus positive relationships) of interrelationships between outcome constructs, it did not examine causality. In other words, do attit ude toward the job or physical consequences act as antecedents to other outcome constructs or to each other? Such insights as to whether physical consequences decrease attitude toward th e job, overall organizational commitment, etc., or if attitude toward the job increases or decreases physical consequences could aid management in assi gning valuable, and often scarce, financial resources where they are of the greatest benefit when attempting to increase job performance and reduce turnover. Concluding Comments This research endeavor allows for a gr eater understanding of the complexity that job stress places not only on FSSPs job perfor mance, but also on FSSPs desire to switch positions within the organization or to leave the organization. By taking a global look at the antecedents and consequences of job st ress from the FSSPs prospective, it is the researchers belief that the study provides a cl earer picture of the job stress process for FSSP. Findings of the study show that there are at least five construc ts that impact job stress (i.e., role ambiguity, role conflict, role overload, perceived customer demands, and emotional labor), but it is also evident that other sources of jo b stress exist that still need

PAGE 222

209 to be investigated. Finally, it is also known that job autonomy reduces job stress, but not job stress coming from perceived customer de mands. Also, the mediating influence of emotional intelligence demonstrates its valu e in reducing job stress. A global look shows that emotion-focused coping frequency of use and effectiveness of use have limited application in reducing job stre ss coming from the five sour ces of job stress on the five outcome constructs (i.e., attitude toward the job, physical consequences, overall organizational commitment, overall job satisfaction, and emotional exhaustion) examined in this study. It is also a pparent that the influence of these five outcomes constructs impact job performance and that job performa nce impacts intentions to switch and to leave the organization. And, fi nally, it is shown that these intentions are correlated. Without taking a holistic appro ach in empirically examining these relationships, it is unlikely that vital relationships in the progressi on from antecedents of job stress to final outcomes would not have painted as clear a picture as is offere d through this holistic approach. It is hoped that findings of this research endeavor will create interest among other researchers to study this important research su bject. It is believed that understanding coming from this research is relevant for sp ecific marketing-oriented situations such as customer call centers and sales force management.

PAGE 223

210 Bashaw, R. E. & Grant, E. S. (1994, Spring). Exploring the Distinc tive Nature of Work Commitments: The Relationships with Pers onal Characteristics, Job Performance, and Propensity to Leave. Journal of Personal Sel ling and Sales Management, 2, 41-56. REFERENCES Aghazadeh, S. (1999). Human Resource Management: Issues and Challenges in the New Millennium. Management Research News, 22, 19-32. Aker, M. & Porter, G. L. (2003, March). Your EQ Skills: Got What It Takes Journal of Accountancy, 65-69. Altman, I. & Taylor, D. (1973). Social Penetration: The Development of Interpersonal Relationships New York, NY: Holt, Rine hart and Winston Inc. Andreasen, A. R. (1995). Marketing Social Change San Francisco, CA: Jossey-Bass. Ashforth, B. E. & Humphrey, R. H. (1993). Emotional Labor in Service Roles: The Influence of Identity. Academy of Management Review, 18, 88-115. Ashkanasy, N. M. (2003, November). Emotions in Organizations: A Multilevel Perspective. Presentation at USF. Athanassopoulos, A. D. (2000, March). Cust omer Satisfaction Cues to Support Market Segmentation and Explain Switching Behavior. Journal of Business Research, 47, 191-207. Babakus, E., Cravens, D. W., Johnston, M., & Moncrief, W. C. (1999, Winter). The Role of Emotional Exhaustion in Sales Force Attitude and Behavior Relationships. Journal of Academy of Marketing Science, 27, 58-70. Bagozzi, R. P. (1978, November). Salesforce Performance and Satisfaction as a Function of Individual Difference, Interperso nal, and Situational Factors. Journal of Marketing Research 15, 517-531. Baron, R. M. & Kenny, D. A. (1986). The Mode rator-Mediator Variab le Distinction in Social Psychological Research: Concep tual, Strategic, and Statistical Considerations. Journal of Marketing Research, 19, 29-239. Bateman, T. S. & Strasser, S. (1984, Ma rch). A Longitudinal Analysis of the Antecedents of Organizational Commitment. Academy of Management Journal, 27, 95-112.

PAGE 224

211 Bridges, T. & Lauer, N. C. (2003, February). Plan to Re form DCF Coming Under Fire. Tallahassee Democrat. www.tallahassee.com/mld/tallahassee/512433.htm Bedeian, A. G. & Armenakis, A. A. (1981, June). A Path-Analysis Study of the Consequences of Role Conflict and Ambiguity. Academy of Management Journal, 24, 417-424. Behrman D. N. & Perreault, Jr., W. D. (1984, Fall). A Role Stress Model of the Performance and Satisfaction of Industrial Salespersons. Journal of Marketing, 48, 9-21. Bentler, P. M. & Choa, C. (1987). Pract ical Issues in Structural Modeling. Sociological Methods and Research, 16, 78-117. Bettencourt, L. A. & Brown, S. W. (2003). Role Stressors and Customer-Oriented Boundary-Spanning Behaviors in Service Organizations Journal of the Academy of Marketing Science, 31, 394-408. Biddle, B. J. (1979). Role Theory: Expectations, Identities and Behaviours. New York, NY: Academic Press, Inc. Bitner, M. J., Booms, B. H., & Tetreault, M. S. (1990, January). The Service Encounter: Diagnosing Favorable and Un favorable Incidents. Journal of Marketing, 54, 7184. Bluedorn, A. C. (1982). A Unified Mode l of Turnover from Organizations. Human Relations, 35, 135-153. Bowen, D. E. & Schneider, B. (1985). Boundary-Spanning Role Employees and the Service Encounter: Some Guidelines fo r Management and Research. In J. Czepiel, M. Solomon, & C. Supernant (Eds.), The Service En counter: Managing Employee/Customer Interaction in Service Businesses (pp. 127-148). Lexington, MA: Lexington Books. Broderick, A. J. (1999, April). Role Theory and the Management of Service Encounters. The Service Industries Journal, 19, 117-131. Brotheridge, C. M. & Lee, R. T. (2003, September). Development and Validation of the Emotional Labour Scale. Journal of Occupational and Organizational Psychology, 76, 365-379. Brown, S. P. & Peterson, R. A. (1993, February). Antecedents and Consequences of Salesperson Job Satisfaction: Meta-Analysis and Assessment of Causal Effects. Journal of Marketing Research, 30, 63-77.

PAGE 225

212 Burke, R. J. & Deszcea, E. (1986, June). Correlates of Psychological Burnout Phases Among Police Officers. Human Relations, 39, 487-491. Burnham, T. A., Frels, J. R., & Mahajan, V. (2003, Spring). Consumer Switching Costs: A Typology, Antecedents, and Consequences. Journal of Academy of Marketing Science, 31, 109-126. Busch, P. & Bush, R. F. (1978, August). Women Contrasted to Me an in the Industrial Salesforce: Job Satisfaction, Va lues, Role Clarity, Performance, and Propensity to Leave. Journal of Marketing Research, 15, 438-448. Byrne, B. M. (1998). Structural Equation Modeling with LISREL, PRELIS, and SIMPLIS: Basic Concepts, Applications and Programming Mahwah, NJ: Lawrence Erlabum Associates. Cano, C., Carrillat, F., & Jaramillo, F. (2004, June). A Meta-Analysis of the Market Orientation-Business Performance Relations hip: Evidence from Five Continents. International Journal of Research in Marketing, 21 179-200. Carlzon, J. (1987). Moments of Truth Cambridge, MA: Ballinger Publishing Company. Center for Womens Healthcar e. (1998, October). How Stress Affects Your Physical Health. Cornell University Medical College: Womens Health Advisor 2, 1-14. Chao, E. L. & Henshaw, J. L. (2003). Guid elines for Preventing Workplace Violence for Health-Care and Social Service Workers. U. S. Department of Labor, OSHA, 3148. Chebat, J. C. & Kollias, P. (2000, August). The Impact of Empowerment on Customer Contact Employees Role in Service Organization. Journal of Services Research 3, 66-82. Chenet, P., Tynan, C., & Money, A. (1999, October). Service Performance Gap: ReEvaluation and Redevelopment. Journal of Business Research 46, 133-147. Cherniss, C. & Goleman, D. (2001). The Emotionally Intelligent Workplace: How to Select for, Measure, and Improve Emo tional Intelligence in Individuals, Groups, and Organizations San Francisco, CA: Jossey-Bass. Cherniss, C. (1980). Staff Burnout, Job Stress in the Hu man Services: Sage Studies in Community Mental Health 2 London, UK: Sage Publications. Chung, B. G. & Schneider, B. (2002). Serving Multiple Masters: Role Conflict Experienced by Service Employees. Journal of Services Marketing, 16, 70-87.

PAGE 226

213 Churchill, Jr., G. A., Ford, N. M., & Walker, Jr., O. C. (1976, November). Organizational Climate and Job Sa tisfaction in the Salesforce. Journal of Marketing Research, 13, 323-332. Churchill Jr., G. A., Ford, N. M., Hartley, S. W., & Walker, Jr., O. C. (1985, May). The Determinants of Salesperson Pe rformance: A Meta-Analysis. Journal of Marketing Research, 22, 103-118. Cohen, A. (1993). Organizational Commit ment and Turnover: A Meta-Analysis. Academy of Manageme nt Journal, 36, 1140-1157. Cooper, C. L., Mallinger, M. & Kahn, R. (1978, September). Identifying Sources of Occupational Stress Among Dentists. Journal of Occupational Psychology, 51, 227-234. Cordes, C. & Dougherty, T. W. (1993, Octobe r). A Review and An Integration of Research on Job Burnout. Academy of Management Review, 18, 621-656. Cordes, C. L., Dougherty, T. W. & Blum, M. (1997, November). Patterns of Burnout Among Managers and Professionals: A Comparison of Models. Journal of Organizational Behavior, 18, 685-701. Cropanzano, R., Rupp, D. E., & Byrne, Z. S. (2003). The Relationship of Emotional Exhaustion to Work Attitudes, Job Perfor mance, and Organizational Citizenship Behaviors. Journal of Applied Psychology, 88, 160-169. Czepiel, J. A. (1990, January). Servi ce Encounters and Service Relationships: Implications for Research. Journal of Business Research, 20, 13-21. DeJonge, J., Dormann, C., Janssen, P. P. M ., Dollard, M. F., Landeweerd, J. A., & Nijhuis, F. J. N. (2001). Testing Reciprocal Relationships Between Job Characteristics and Psychological Well-Being: A Cross-Lagged Structural Equation Model. Journal of Occupational and Organizational Psychology, 74, 29-46. Dewe, P. & ODriscoll, M. (2002). Stress Management Interventions: What Do Managers Actually Do? Personnel Review, 31, 143-165. Diamantopoulos, A. & Siguaw, J. A. (2000). Introducing LISREL: A Guide for the Uninitiated London, UK: Sage Publications. Diefendorff, J. M. & Gosserand, R. H. ( 2003, September). Understanding the Emotional Labor Process: A Control Theory Perspective. Journal of Organizational Behavior, 24, 945-959.

PAGE 227

214 Donnelly, Jr., J. H. & Ivancevich, J. M. (1975, January). Role Clarity and the Salesman. Journal of Marketing, 39, 71-74. Drolet, A. L. & Morrison, D. G. (2001, February). Do We Rea lly Need Multiple-Item Measures in Service Research? Journal of Service Research 3, 196-204. Erera, I. P. (1989). Role Ambiguity in Public Welfare Organizations. Administration in Social Work, 13, 67-82. Fisher, C. D. (2003). Why Do Lay People Believe That Satisfaction and Performance are Correlated? Possible Sources of a Commonsense Theory. Journal of Organizational Behavior, 24, 753-777. Fisher, C. D. & Ashkanasy, N. M. (2000). The Emerging Role of Emotion in Work Life: An Introduction. Journal of Organizational Behavior, 21, 123-129. Folkes, V. S. & Patrick, V. M. (2003, June). The Positive Effect in Perception of Service: Seen One, Seen Them All? Journal of Consumer Research 30, 125137. Folkman, S. & Lazarus, R. (1988, March). Coping as a Mediator of Emotion. Journal of Personality and Social Psychology, 54, 466-475. Gardner, H. (1983). Frames of Mind: The Theory of Multiple Intelligence New York, NY: Basic Books. Inc. Gerbing, D. W. & Anderson, J. C. (1988, May). An Updated Paradigm for Scale Development Incorporating Unidimensionality and Its Assessment. Journal of Marketing Research, 25, 186-192. Goffman, E. (1959). The Presentation of Se lf in Everyday Life Garden City, NJ: Doubleday Anchor Books. Goffman, E. (1967). Interaction Ritual: Essays on Face-to-Face Behaviors New York, NY: Doubleday and Company, Inc. Goffman, E. (1969). Strategic Interaction Philadelphia, PA: University of Pennsylvania Press. Goleman, D. (1995). Emotional Intelligence: Why It Can Matter More Than IQ New York, NY: Bantam Books. Goleman, D. (1998, March/April). Th e Emotionally Competent Leader. The Healthcare Forum Journal, 41, 36, 38, 76.

PAGE 228

215 Hair, Jr., J. F., Anderson, R., Tatham R. L., & Black, W. C. (1998). Multivariate Data Analysis (4th ed.), New Jersey: Prentice Hall. Hatcher, L. (1994). Step-By-Step Approach to Using the SAS System for Factor Analysis and Structural Equation Modeling Cary, NC: SAS Institute Inc. Goleman, D. (2000). Working with Emotional Intelligence New York, NY: Baton Books. Goolsby, J. R. (1992, Spring). A Theory of Role Stress in Boundary Spanning Positions of Marketing Organizations. Journal of Academy of Marketing Science, 2, 155164. Gruskin, S. (2003). Swamped Social Wo rkers in South Florida Collect Massive Overtime Paychecks. Knight Ridder Tribune Business News 1+. Hackman, J. R. & Oldham. G. R. (1975). Development of the Job Diagnostic Survey. Journal of Applied Psychology, 60, 159-170. Hair, Jr., J., Bush, R. P., & Ortinau, D. J. (2003). Marketing Research: Within A Changing Information Environment. Boston, MA: McGraw-Hill Irwin. Hall, E. M. (1989). Gender, Work Contro l, and Stress: A Theo retical Discussion and Empirical Test. International Journal of Health Services, 19, 725-745. Hampton, R., Dubinsky, A. J., & Skinner, S. J. (1986, Fall). A Model of Sales Supervisor Leadership Behavior and Reta il Salespeoples Job-Related Outcomes. Journal of Academy of Marketing Science, 14, 33-43. Hecker, D. E. (2001, November). Occupational Employment Projections to 2010. Monthly Labor Review 57-84. Hochschild, A. R. (1983). The Managed Heart: Commercialization of Human Feeling Berkley, CA: University of California Press Ltd. Homans, G. C. (1961). Social Behaviour: Its Elementary Forms New York, NY: Harcourt Brace and World. Houkes, I., Janssen, P. P. M., DeJonge, J., & Bakker, A. B. (2003, December). Specific Determinants of Intrinsic Work Moti vation, Emotional Exhaustion, and Turnover Intention: A Multisample Longitudinal Study. Journal of Occupational and Organizational Psychology, 76, 427-450.

PAGE 229

216 Idaszak, J. R. & Drasgow, F. (1987, Febr uary). A Revision of the Job Diagnostic Survey: Elimination of Measurement Artifact. Journal of Applied Psychology, 72, 69-74. Jackson, S. E., Schwab, R. L., & Schuler, R. S. (1986, November). Toward an Understanding of the Burnout Phenomenon. Journal of Applied Psychology, 71, 630-640. Jacobson, E., Charters, Jr., W.W., & Lieberman, S. (1951). The Use of the Role Concept in the Study of Complex Organizations. Journal of Social Issues 7, 18-27. Jamal, M. & Baba, V. V. (1992, September). Shiftwork and Department-Type Related to Job Stress, Work Attitudes, and Behavioral Intentions: A Study of Nurses. Journal of Organizational Behavior, 13, 449-464. Jex, S. M. & Elacqua, T. C. (1999, March). Self-Esteem as a Moderator: A Comparison of Global Organization-Based Measures. Journal of Occupational and Organizational Psychology, 72, 71-81. Johlke, M. C. & Dunham, D. F. (2000, November). Supervisor Communication Practices and Service Employee Job Outcomes. Journal of Service Research 3, 154-165. Johlke, M. C. & Dunham, D. F. (2001, Sp ring). Supervisor Communication Practices and Boundary Spanner Role Ambiguity. Journal of Managerial Issues 21, 87101. Johnston, M. W. & Futrell, C. M.. (1989, Ma rch). Functional Salesforce Turnover: An Empirical Investigation Into the Positive Effects of Turnover. Journal of Business Research, 18, 141-157. Judge, T. A., Thoresen, C. J., Bono, J. E., & Patton, G. K. (2001). The Job SatisfactionJob Performance Relationship: A Qualitative and Quantitative Review. Psychological Bulletin 127, 376-407. Kahn, R. L., Wolf, D. M., Quinn, R. P., Snock, J. D., & Rosenthal, R. A. (1964). Organizational Stress: Studies in Role Conflict and Ambiguity New York, NY: Wiley. Karasek, Jr., R. A. (1979, June). Job Demands, Job Decision Latitude, and Mental Strain: Implications for Job Redesign. Administrative Science Quarterly 24, 285308. Kelley, H. & Thibaut, J. (1978). Interpersonal Relations: A Theory of Interdependence New York, NY: Wiley.

PAGE 230

217 Kidwell, Jr., R. E. & Bennett, N. (2001). Perceived Work Context and Job Neglect. American Business Review, 19, 64-74. Kimes, J. D. (1977, September). Handling Stress in the Accounting Profession. Management Accounting, 59, 19-25. King, L. A. & King, D. W. (1990). Role Conflict and Role Ambiguity: A Critical Assessment of Construct Validity. Psychological Bulletin, 107, 48-64. Kotler, P., Roberto, N., & Lee, N. (2002). Social Marketing: Improving the Quality of Life Thousand Oaks, CA: Sage Publications. Kruml S. M. & Geddes, D. (2000, August). Exploring the Dimensions of Emotional Labor. Management Communications Quarterly, 14, 8-49. Lait, J. & Wallace, J. E. (2002, Summer). Stress at Work: A Study of OrganizationalProfessional Conflict and Unmet Expectations. Relations Industrielles 57, 463488. Lam, T., Baum, T., & Pine, R. (2001). St udy of Managerial Job Satisfaction in Hong Kongs Chinese Restaurants. International Journal of Contemporary Hospitality Management, 31, 35-42. Lambert, D. M. & Harrington, T.C. (1990). Measuring Nonresponse Bias in Customer Service Mail Surveys. Journal of Business Logistics, 11, 1-25. Lazarus, R. S. (1966). Psychological Stress and the Coping Process NewYork, NY: McGraw-Hill. Lazarus, R. S. & Folkman, S. (1984). Stress, Appraisal and Coping New York, NY: Springer Publishing Company. Lee, K., Carswell, J. J. & Allen, N. J. ( 2000). A Meta-Analytic Re view of Occupational Commitment: Relations Within Pers on and Work-Related Variables. Journal of Applied Psychology, 85, 799-811. Lee, R. T. & Ashforth, B. E. (1993, January ). A Further Examination of Managerial Burnout: Toward an Integrated Model. Journal of Organizational Behavior ,14, 320. Lee, T. W. & Mowday, R. T. (1987, Decembe r). Voluntarily Leaving an Organization: An Empirical Investigation of Steers and Mowdays Model of Turnover. Academy of Manageme nt Journal, 30, 721-743.

PAGE 231

218 Levi-Strauss, C. (1949). The Elementary Structures of Kinship. Leiter, M. P. (1991, March). Coping Pattern s as Predictors of Burnout: The Function of Control and Escapist Coping Patterns. Journal of Organizational Behavior, 12 123-144. Lerner, J. S. & Keltner, D. ( 2001). Fear, Anger, and Risk. Journal of Personality and Social Psychology, 81, 146-159. Oxford, U.K.: Alden Press. Li, T. & Calantone, R. J. (1998, Octobe r). The Impact of Market Knowledge Competence on New Product Advantage: Conceptualizati on and Empirical Examination. Journal of Marketing, 62, 13-29. Madsen, S. R. (2003, Spring). Wellness in the Workplace: Preparing Employees for Change. Organization Development Journal 21, 46-55. Lord, V. B. (1996). An Impact of Commun ity Policing: Reported Stressors, Social Support, and Strain Among Police Officer s in a Changing Police Department. Journal of Criminal Justice, 24, 503-522. MacCallum, R. C., Browne, M. W., & Sugawa ra, H. M. (1996). Power Analysis and Determination of Sample Size fo r Covariance Structure Modeling. Psychological Methods, 1 (2), 130-149. Mangione, T. W. (1995). Mail Surveys: Improving the Quality Thousand Oaks, CA: Sage Publications. Maslach, C. & Jackson, S. E. (1979, May). Burned-Out Cops and Their Families. Psychology Today, 12, 58-62. Maslach, C. & Jackson, S. E. (1981, Novemb er). The Measurement of Experienced Burnout. Journal of Occupational Behaviour, 2, 99-113. Mayer, J. D. & Caruso, D. R. (1993). The Intelligence of Emotional Intelligence Intelligence, 17, 433-442. Mayer, J. D., Caruso, D. R., & Salovey, P. (2000). Emotional Intelligence Meets Traditional Standards for an Intelligence. Intelligence, 27, 267-298. McAlexander, J. H., Kaldenberg, D. O., & Ko enig, H. F. (1994, Fall). Service Quality Measurement. Journal of Health Care Marketing, 14, 34-40.

PAGE 232

219 McGrath, Joseph E. (1976). St ress and Behavior in Organizations. In M. D. Dunnette (Ed.), Handbook of Industrial and Organizational Psychology (pp. 1351-1395), Chicago, IL: Rand McNally College Publishing Company. McQuitty, S. (2004, February). Statistical Power and Structural Equation Models in Business Research. Journal of Business Research 57, 175-183. Merton, R. K. (1957, June). The Role Se t: Problems in Sociological Theory. The British Journal of Sociology, 8, 106-120. Miles, R. H. & Perreault, Jr., W. D. (1976). Organizationa l Role Conflict: Its Antecedents and Consequences. Organizational Behavior and Human Performance 17, 19-44. Moncrief, W., Babakus, E., Cravens, D. W., & Johnston, M. (1997, January). Examining the Antecedents and Consequences of Salesperson Job Stress. European Journal of Marketing, 31, 786-798. Morris, J. A. & Feldman, D. C. (1996). The Dimensions, Antecedents, and Consequences of Emotional Labor. Academy of Management Review, 21, 9961010. Mowday, R. T., Steers, R. M., & Porter, L. W. (1979). The Measurement of Organizational Commitment. Journal of Vocational Behavior 14, 224-247. Munz, D. C., Huelsman, T. J., Konold, T. R., & McKinney, J. J. (1996). Are There Methodological and Substantive Roles for Affectivity in Job Diagnostic Survey Relationships? Journal of Applied Psychology, 81, 795-805. Murphy, K. R. & Myors, B. (2004). Statistical Power Analysis: A Simple and General Model for Traditional and Modern Hypothesis Tests. Mahwah, NJ: Lawrence Erlbaum. Narayanan, L., Menon, S., & Spector, P. E. (1999). Stress in the Workplace: A Comparison of Gender and Occupations. Journal of Organizational Behavior, 20, 63-73. Nelson, K. & Bowen, J. (2000, April). The Effect of Employee Uniforms on Employee Satisfaction. Cornell Hotel and Restaurant Administration Quarterly, 41, 86-93. Netemeyer, R. G., Johnston, M. W., & Burton, S. (1990). Analysis of Role Conflict and Role Ambiguity in a Structural Equations Framework. Journal of Applied Psychology, 75, 148-157.

PAGE 233

220 Netemeyer, R. G., Burton, S., & Johnston, M. W. (1995, January). A Nested Comparison of Four Models of the Conse quences of Role Perception Variables. Organizational Behavior and Huma n Decision Processes, 61, 77-93. Normann, R. (1983). Service Management: Strategy and Leadership in Service Businesses. New York, NY: J ohn Wiley and Sons. Nunnally, J. C. & Bernstein, I. H. (1994). Psychometric Theory New York, NY: McGraw-Hill, Inc. Ortinau, D. J. & Brensinger, R. T. (1992). An Empirical Inves tigation of Perceived Qualitys Intangible Dimensionality Thr ough Direct Cognitive Structural (DCS) Analysis. In R. L. King (Ed.), Marketing: Perspective for the 1990s. Southern Marketing Association, Ne w Orleans, LA., 2/14-2/19. Parker, D. F. & Decotiis, T. A. (1983). Organizational Determinants of Job Stress. Organizational Behavior and Human Performance 32, 160-177. Parker, S. K., Wall, T. D., & Jackson, P. R. (1997, August). Thats Not My Job: Developing Flexible Employee Work Orientations. Academy of Management Journal 40, 899-929. Petrick, J. F. (2002). Development of A Multi-Dimensional Scale for Measuring the Perceived Value of A Service. Journal of Leisure Research, 34, 119-134. Princeton Survey Research Association. (1997). Labor Day survey: State of Workers. Princeton, NJ. Quinn, R .P. & Shepard, L. G. (1974). The 1972-1973 Quality of Employment Survey Ann Arbor, MI: University of Michig an Institute for Social Research. Rafaeli, A. & Sutton, R. I. (1987). Expressi on of Emotion as Part of the Work Role. The Academy of Management Review, 12, 23-37. Rice, R. W., Gentile, D. A., & McFarlin, D. B. (1991). Facet Importance and Job Satisfaction. Journal of Applied Psychology 76, 31-39. Rizzo, J., House, R. J., & Lirtzman, S. I. (1970, June). Role Conflict and Role Ambiguity in Complex Organizations. Administrative Science Quarterly, 15, 150163. Robertson, I. T. & Cooper, C. (1990). The Validity of the Occupational Stress Indicator. Work and Stress, 4, 29-39.

PAGE 234

221 Sauter, S., Murphy, L., Colligan, M., Swanson, N ., Hurrell, Jr., J., Scharf, Jr., F., Sinclair, R., Grubb, P., Goldenhar, L., Alterman, T ., Johnston, J., Hamilton, A., & Tisdale, J. (1999). Stress At Work. http://www.cdc/gov/niosh/stesswk.html Date Accessed: August 5, 2003. Rogers, J. D., Clow, K. E., and Kash, T. J. (1994). Increasing Job Satisfaction of Service Personnel. The Journal of Services Marketing, 8, 14-27. Sager, J. K., Futrell, C. M., & Varadaraja n, R. (1989, June). Exploring Salesperson Turnover: A Causal Model. Journal of Business Research, 18, 303-326. Sager, J. K. & Menon, A. (1994, March). The Ro le of Behavioral Intentions in Turnover of Salespeople. Journal of Business Research, 29, 179-188. Saks, A. M. & Cronshaw, S. F. (1990, May). A Process Investigation of Realistic Job Previews: Mediating Variables a nd Channels of Communication. Journal of Organizational Behavior, 11, 221-236. Sales, S. M. (1969, September). Organizati onal Role as a Risk Factor in Coronary Disease. Administrative Science Quarterly 14, 325-336. Salovey, P. & Mayer, J. D. ( 1990). Emotional Intelligence. Imagination, Cognition and Personality, 9, 185-211. Sarbin, T. R. (1954). Role Theory. In G. Lindzey (Ed.), Handbook of Social Psychology (pp. 223-258). Cambridge, MA: Addison-Wesley. Savicki, V. & Cooley, E. J. (1994, Decembe r). Burnout in Child Protective Service Workers: A Longitudinal Study. Journal of Organizational Behavior, 15, 655666. Saxton, M. J., Phillips, J. S., & Blakeney, R. N. (1991). Antecedents and Consequences of Emotional Exhaustion in the Airl ine Reservations Service Sector. Human Relations, 44, 583-595. Schaubroeck, J. & Jones, J. R. (2000). Antecedents of Workplace Emotional Labor Dimensions and Moderators of Thei r Effects on Physical Symptoms. Journal of Organizational Behavior, 21, 163-183. Schaubroeck, J., Ganster, D. C., Sime, W. E., & Ditman, D. (1993, Spring). A Field Experiment Testing Supervis ory Role Clarification. Personnel Psychology, 46, 125.

PAGE 235

222 Schneider, B., Ashworth, S. D., Higgs, A. C., & Carr, L. (1996). Design, Validity, and Use of Strategically Focused Employee Attitude Surveys. Personnel Psychology, 49, 695-705. Schuler, R. S. (1980). Defi nition and Conceptualization of Stress in Organizations. Organizational Behavior and Human Performance 25, 184-215. Shenkar, O. & Zeira, Y. (1991, August). Role Conflict and Role Ambiguity of Chief Executive Officers in International Joint Ventures. Journal of International Business Studies 55-75. Shostack, G. L. (1977, April). Breaking Free From Product Marketing. Journal of Marketing 41, 73-80. Simmel, G. (1908/1950). Types of Social Relationships by Degrees of Reciprocal Knowledge of the Participants. In K. Wolff (Ed.), The Sociology of Georg Simmel (pp. 317-329). New York, NY: Free Press. Singh, J. & Rhoads, G. K. (1991, August). Boundary Role Ambiguity in MarketingOriented Positions: A Multidimensional, Multifaceted Operationalization. Journal of Marketing Research, 28, 328-338. Singh, J., Goolsby, J. R., & Rhoads, G. K. (1994, November). Behavioral and Psychological Consequences of Bounda ry Spanning Burnout for Customer Service Representatives. Journal of Marketing Research 31, 558-569. Singh, J., Verbeke, W., & Rhoads, G. K. (1996, July). Do Orga nizational Practices Matter in Role Stress Process? A Study of Direct and Moderating Effects for Marketing-Oriented Boundary Spanners. Journal of Marketing, 60, 69-86. Singh, J. (1993, April). Boundary Role Ambigu ity: Facets, Determinants, and Impacts. Journal of Marketing 57, 11-31. Singh, J. (1998, July). Striking a Balan ce in Boundary-Spanning Positions: An Investigation of Some Un conventional Influences of Role Stressors and Job Characteristics on Job Outc omes of Salespeople. Journal of Marketing 62, 69-86. Singh, J. (2000, April). Performance Productivi ty and Quality of Frontline Employees in Service Organizations. Journal of Marketing 17, 15-34. Smith, C. S., Tisak, J., & Schmieder, R. A. (1993, June). The Measurement Properties of the Role Conflict and Role Ambiguity Scales: A Review and Extension of the Empirical Research. Journal of Organizational Behavior 14, 37-48.

PAGE 236

223 Sojka, J. Z. & Deeter-Schmelz, D. R. (2002, Spring). Enhancing the Emotional Intelligence of Salespeople. Mid-American Journal of Business 17, 43-50. Solomon, M. R., Surprenant, C., Czepiel, J. A., & Gutman, E. G. (1985, Winter). A Role Theory Perspective on Dyadic In teractions: The Service Encounter. Journal of Marketing 59, 99-111. Spector, P. E. (1987). Interactive Effect s of Perceived Control and Job Stressors on Affective Reactions and health Ou tcomes for Clerical Workers. Work and Stress, 1, 155-162. Spector, P. E. and Jex, S. M. (1998). Develo pment of Four Self-Report Measures of Job Stressors and Strain: Interpersonal Conflict at Work Scale, Organizational Constraints Scale, Quantitative Worklo ad Inventory, and Physical Symptoms Inventory. Journal of Occupati onal Health Psychology 3, 356-367. Spector, P. E., Dwyer, D. J., & Jex, S. M. ( 1988). Relation of Job Stressors to Affective, Health, and Performance Outcomes: A Comparison of Multiple Data Sources. Journal of Applied Psychology 73, 11-19. Spector, P., Jex, S., & Chen, P. (1995). Relations of Incumbent Affect-Related Personality Trait with Incumbent and Obj ective Measures of Characteristics of Jobs. Journal of Organizational Behavior, 16, 59-65. Steel, R. P. and Rentsch, J. R. (1997). The Dispositional Model of Job Attitudes Revisited: Findings of a 10-Year Study. Journal of Applied Psychology 82, 873879. Steinberg, R. J. (1999, January). Emotional Labor in Job Evaluation: Redesigning Compensation Practice of Political Social Science. The Annals of the American Academy 561, 143-157. Stone, A. A., Greenberg, M. A., Kennedy-Moor e, E., Newman. M. G. (1991). SelfReport, Situation-Specific Coping Questi onnaire: What are They Measuring? Journal of Personality and Social Psychology 61, 648-658. Strutton, D. and Lumpkin, J. R. (1994, Sp ring). The Relationship Between Optimism and Coping Styles of Salespeople. Journal of Personal Selling and Sales Management 2, 71-82. Sweeney, J. C., Soutar, G. N., & Johnson, L. W. (1999, Spring). The Role of Perceived Risk in the Quality-Value Relationship: A Study in a Retail Environment. Journal of Retailing 75, 77-105.

PAGE 237

224 Teas, R. K. (1983, February). Supervis ory Behavior, Role Stress, and The Job Satisfaction of Industrial Salespeople. Journal of Marketing Research 20, 84-91. Thorndike, E. L. (1920). Intelligence and Its Uses. Harpers Magazine 140, 227-235. Tyagi., P. K. (1985, Summer). Relative Importance of Key Job Dimensions and Leadership Behaviors in Motivating Salesperson Work Performance. Journal of Marketing 49, 76-86. Verbeke, W. (1997, September). Individual Differences in Emotional Contagion of Salespersons: Its Effect on Performance and Burnout. Psychology and Marketing 14, 617-636. Vitello-Cicciu, J. M. (2003, October). Emotional Intelligence. Nursing Management 28-33. Walker, O., Churchill, Jr., G. A. & Ford, N. M. (1975, January). Organizational Determinants of the Industrial Salesmens Role Conflict and Ambiguity. Journal of Marketing 39, 32-39. Wang, G. & Netemeyer, R. G. (2002). The Effects of Job Autonomy, Customer Demandingness, and Trait Competitiveness on Salesperson Learning, SelfEfficacy, and Performance. Journal of the Academy of Marketing Science 30, 217-228. Weiss, H. M. & Cropanzano, R. (1996). Affective Events Theory: A Theoretical Discussion of the Structure, Causes and C onsequences of Affective Experiences at Work. In Research in Organizational Behav ior Human Decisions Process (pp. 125) Greenwich, CT: J.A.I. Press. Wharton, A. S. & Erickson, R. J. (1993). Managing Emotions on the Job and At Home: Understanding the Consequences of Multiple Emotional Roles. Academy of Management Review 18 (3), 457-486. Wheelwright, S. C. & Clark, K. B. (1992). Revolutionizing Product Development New York, NY: The Free Press. Wright, T. A. & Cropanzano, R. (1998). Em otional Exhaustion as a Predictor of Job Performance and Voluntary Turnover. Journal of Applied Psychology, 83, 486493. Zeidner, M., Matthews, G., Roberts, R. D. (2004, July). Emotional Intelligence in the Workplace: A Critical Review. Applied Psychology, 53 (3), 371-399.

PAGE 238

225 APPENDICES

PAGE 239

226 Appendix A Pilot Study Survey To what extent do you think each of the following factors influences how you react toward your job ? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Trying to maintain a positive attitude toward customers 1 2 3 4 5 6 7 2. Being enthusiastic about your job 1 2 3 4 5 6 7 3. Perceiving your job as pleasant 1 2 3 4 5 6 7 4. Finding the job interesting 1 2 3 4 5 6 7 5. Getting the feedback from your boss that you expect 1 2 3 4 5 6 7 6. Communication with the boss 1 2 3 4 5 6 7 7. Overall working conditions 1 2 3 4 5 6 7 8. Morale (in other words is the job uplifting or dismal) 1 2 3 4 5 6 7 Is there something else that should be included? To what extent do you think each of the following factors represents your ability to perceive emotions in others as well as yourself ? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Recognizing emotions present in a particular situation 1 2 3 4 5 6 7 2. Figuring out the reasons behind different emotions 1 2 3 4 5 6 7 3. Differentiating between the emotions you experience 1 2 3 4 5 6 7 4. Thinking about the emotions behind your actions 1 2 3 4 5 6 7 5. Awareness of how your feelings affect you 1 2 3 4 5 6 7 6. Acknowledging feelings of others at work 1 2 3 4 5 6 7 7. Examining the feelings, thoughts, and actions of others 1 2 3 4 5 6 7 8. Attempting to understand why other people feel the way they do 1 2 3 4 5 6 7 9. Observing how others react to you in an effort to understand your own behavior 1 2 3 4 5 6 7 Is there something else that should be included?

PAGE 240

227 Appendix A (Continued) To what extent do you think e ach of the following statements reflects your ability to manage your emotions as well as others emotions ? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. When frustrated or angr y you consider options available 1 2 3 4 5 6 7 2. Composure even when feeling angry 1 2 3 4 5 6 7 3. Expecting that you will succeed for most endeavors he or she takes on 1 2 3 4 5 6 7 4. Reacting to challenges 1 2 3 4 5 6 7 5. Failing or succeeding to control your emotions 1 2 3 4 5 6 7 6. Seeking out uplifting activities 1 2 3 4 5 6 7 7. Arranging event that others enjoy 1 2 3 4 5 6 7 8. Helping others feel better when they are down 1 2 3 4 5 6 7 9. Handling conflict 1 2 3 4 5 6 7 10. Involving yourself in other peoples problems 1 2 3 4 5 6 7 Is there something else that should be included? To what extent do you think each of the following factors reflects whether or not you must adhere to specific or implied rules on the job when displaying your emotions ? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. You believe that you must hold back your true feelings 1 2 3 4 5 6 7 2. You must pretend to have emotions that you dont really have 1 2 3 4 5 6 7 3. There is an expectation that you will hide your true feelings about a situation 1 2 3 4 5 6 7 4. There is the expect ation that you must make an effort to actually feel an emotion that management believes you should feel 1 2 3 4 5 6 7 5. There is the expectation that you will try to actually experience an emotion that you are supposed to show 1 2 3 4 5 6 7 6. There is the expectation that you will really try to feel emotions that are deemed part of your job 1 2 3 4 5 6 7 Is there something else that should be included?

PAGE 241

228 Appendix A (Continued) To what extent do you think each of the following factors requires some degree of effort on your part to display emotions deemed appropriate on the job ? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Faking emotions you show the customer 1 2 3 4 5 6 7 7 2. Talking yourself out of feeling what you really feel when helping customers 1 2 3 4 5 6 3. Conjuring up feelings you need to show the customer 1 2 3 4 5 6 7 4. Changing your actual feelings to match the feeling you must express to customers 1 2 3 4 5 6 7 5. Attempting to create emotions in yourself that present an image the organization desires 1 2 3 4 5 6 7 Is there something else that should be included? To what extent do you think each of the following factors reflects ways you cope with job stress ? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Turn to hobbies and pastimes 1 2 3 4 5 6 7 2. Talk to understanding friends 1 2 3 4 5 6 7 3. Expand interests and activities outside of work 1 2 3 4 5 6 7 4. Seek social support 1 2 3 4 5 6 7 Is there something else that should be included? To what extent do you think each of the following factors reflects a frontline social service persons intention of switching positions within an organization in the next 12 months? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Actively looking for another position within the organization 1 2 3 4 5 6 7 2. Planning to switch positions within the organization 1 2 3 4 5 6 7 3. Planning to keep your position within the organization 1 2 3 4 5 6 7 Is there something else that should be included?

PAGE 242

229 Definitely a Factor Appendix A (Continued) To what extent do you think each of the following factors reflects a frontline social service persons intention to leave their organization in the next 12 months? Circle one number ONLY for each of the following. Not At All A Factor 1. Actively looking for a job with another organization 1 2 3 4 5 6 7 7 2. Planning to switch companies 1 2 3 4 5 6 3. Planning to stay with your current employer 1 2 3 4 5 6 7 Is there something else that should be included? To what extent do you think e ach of the following statements represents whether frontline social service personnel are satisfied or dissatisfied with their job ? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. If given a chance to decide all over again, knowing what he or she knows now, he or she would take the job if it were offered 1 2 3 4 5 6 7 2. If a friend applies for a job like his or hers with the same employer, he or she would recommend the job to him or her 1 2 3 4 5 6 7 3. Judging his or her job against his or her ideal job 1 2 3 4 5 6 7 4. The measurement of his or her expectations when he or she took the job against the actual job 1 2 3 4 5 6 7 5. All things considered, the level of satisfaction with his or her current job 1 2 3 4 5 6 7 6. His or her general feeling about his or her job 1 2 3 4 5 6 7 Is there something else that should be included? Thinking about customers demands to what extent do you think each of the following factors reflects an element of customer demands ? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Level of overall service expected 1 2 3 4 5 6 7 2. Level of service quality expected 1 2 3 4 5 6 7 3. Level of service reliability expected 1 2 3 4 5 6 7 4. Overall expectations for service 1 2 3 4 5 6 7 5. Customers expectation that the services offered meets his or her needs 1 2 3 4 5 6 7 6. Expectations of a specific level of service delivery quality 1 2 3 4 5 6 7 Is there something else that should be included?

PAGE 243

230 Appendix A (Continued) To what extent do you think the following physical problems are consequences of job stress for frontline social service personnel? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Backache 1 2 3 4 5 6 7 2. Skin rash 1 2 3 4 5 6 7 3. Acid indigestion or heart burn 1 2 3 4 5 6 7 4. An infection 1 2 3 4 5 6 7 5. Upset stomach or nausea 1 2 3 4 5 6 7 6. Chest pain 1 2 3 4 5 6 7 7. Constipation 1 2 3 4 5 6 7 8. Diarrhea 1 2 3 4 5 6 7 9. Dizziness 1 2 3 4 5 6 7 10. Eye strain 1 2 3 4 5 6 7 11. Fever 1 2 3 4 5 6 7 12. Headache 1 2 3 4 5 6 7 13. Heart pounding when not exerting 1 2 3 4 5 6 7 14. Loss of appetite 1 2 3 4 5 6 7 15. Shortness of breath 1 2 3 4 5 6 7 16. Stomach cramps (not menstrual) 1 2 3 4 5 6 7 17. Tiredness or fatigue 1 2 3 4 5 6 7 18. Trouble sleeping 1 2 3 4 5 6 7 Is there something else that should be included? To what extent do you think e ach of the following statements reflect workload for frontline social service personnel? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. The job requires him or her to work very fast 1 2 3 4 5 6 7 2. The job requires him or her to work very hard 1 2 3 4 5 6 7 3. The job leaves little time to get things done 1 2 3 4 5 6 7 4. The job requires a great deal of work to be done 1 2 3 4 5 6 7 5. The job requires more work than can be done well 1 2 3 4 5 6 7 Is there something else that should be included?

PAGE 244

231 Appendix A (Continued) To what extent do you think e ach of the following statements creates uncertainty about what frontline social service personnel are required to do on the job? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Managements expectation for interaction with customers 1 2 3 4 5 6 7 2. The amount of service to be provided 1 2 3 4 5 6 7 3. The behavior expected by management for frontline social service personnel w ith customers on the job 1 2 3 4 5 6 7 4. How to handle customer objections 1 2 3 4 5 6 7 5. How to handle unusual problems and situations 1 2 3 4 5 6 7 6. How to deal with customer criticism 1 2 3 4 5 6 7 7. Which specific strengths to point out to customers 1 2 3 4 5 6 7 8. Which specific benefits to highlight for customers 1 2 3 4 5 6 7 9. What actions to take in meeting customers needs 1 2 3 4 5 6 7 10. How to handle non-routine activities on the job 1 2 3 4 5 6 7 11. The amount of work frontline social service personnel are expected to do 1 2 3 4 5 6 7 12. Which cases to give priority 1 2 3 4 5 6 7 13. How much work you are expected to do 1 2 3 4 5 6 7 14. How free time should be handled on the job 1 2 3 4 5 6 7 15. What to do to get a promotion 1 2 3 4 5 6 7 16. Vulnerability to being terminated 1 2 3 4 5 6 7 17. Not knowing what are the critical factors in getting promoted 1 2 3 4 5 6 7 Is there something else that should be included?

PAGE 245

232 Appendix A (Continued) To what extent do you think e ach of the following statements represents conflict on the job for frontline social service personnel? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Having to do things that he or she believes should be done differently 1 2 3 4 5 6 7 2. The need to break company policy rules to carry out an assignment 1 2 3 4 5 6 7 3. Having to work on unnecessary things 1 2 3 4 5 6 7 4. Perceiving requests as incompatible from one or more people 1 2 3 4 5 6 7 5. Doing what will be accepted by one person but not by another 1 2 3 4 5 6 7 6. Working with two or more groups who operate very differently 1 2 3 4 5 6 7 7. Receiving assignments w ithout the manpower to complete them 1 2 3 4 5 6 7 8. Receiving assignments without adequate resources and materials to complete them 1 2 3 4 5 6 7 9. Receiving assignments without being adequately trained 1 2 3 4 5 6 7 Is there something else that should be included? To what extent do you think each of the following factors impacts the way you do your job ? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Ability to decide how to go about doing his or her work 1 2 3 4 5 6 7 2. Ability use his or her personal initiative or judgment in carrying out his or her job 1 2 3 4 5 6 7 3. Opportunity for independence and freedom on how to do the job 1 2 3 4 5 6 7 Is there something else that should be included?

PAGE 246

233 Appendix A (Continued) To what extent do you think each of the following factors make up job stress anxiety for frontline social service personnel? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Feeling fidgety or nervous as a result of the job 1 2 3 4 5 6 7 2 7 2. The job gets to him or her more than it should 1 3 4 5 6 3. On the job there are a lot of things that drive him or her right up a wall 1 2 3 4 5 6 7 5 6 4. Thinking about the job creates a tight feeling in his or her chest 1 2 3 4 7 5. He or she feels guilty when taking time off from work 1 2 3 4 5 6 7 Is there something else that should be included? To what extent do you think each of the following factors represents time pressure job stress for frontline social service personnel? Circle one number ONLY for each of the following. Not At All A Factor Definitely a Factor 1. Too much work to do and too little time to do it 1 2 3 4 5 6 7 2. Burnout caused by excessive job demands 1 2 3 4 5 6 7 3. Dreading the telephone ringing at home because the call might be job-related 1 2 3 4 5 6 7 4. The feeling that he or she never has a day off 1 2 3 4 5 6 7 5. Feeling that he or she is married to the job 1 2 3 4 5 6 7 Is there something else that should be included?

PAGE 247

234 Appendix A (Continued) Not At All A Factor To what extent do you think each of the following factors represent an element of job performance for frontline social service personnel? Circle one number ONLY for each of the following. Definitely a Factor 1. Taking the initiative to help your customers even when it is not part of your responsibility 1 2 3 4 5 6 7 2. Taking the time to help your customers at the expense of not meeting daily productivity goals 1 2 3 4 5 6 7 3. Overall, developing customer trust and confidence in your service 1 2 3 4 5 6 7 4. Responding in a timely manner to customer requests despite your busy schedule 1 2 3 4 6 5 7 5. Consistency in following up on promises made to your customers 1 2 3 4 5 6 7 6. Consistency in providing prompt service to all of your customers 1 2 3 4 5 6 7 7. Consistency in resolving customer concerns the first time 1 2 3 4 5 6 7 8. Consistency in demonstrating emotionally-based behavior deemed approp riate by the company 1 2 3 4 5 6 7 9. Providing accurate or correct information to the customer 1 2 3 4 5 6 7 10. Even thought it is not your responsibility, making sure other departments follow through with your customers requests 1 2 3 4 5 6 7 11. Telling the customer the st raight facts instead of telling them what they want to hear 1 2 3 4 5 6 7 12. Overall performing your job dependably and accurately 1 2 3 4 5 6 7 13. Servicing the account with the customers best interest in mind 1 2 3 4 5 6 7 14. Listening attentively to identify and understand the concerns of customers 1 2 3 4 5 6 7 15. Working out solutions to each customers questions or concerns 1 2 3 4 5 6 7 16. Overall, providing individualized attention to each customers concerns 1 2 3 4 5 6 7 Is there something else that should be included?

PAGE 248

235 Appendix B Personal Interview Sample Questions For Key Constructs Qualify Questions How many year have you worked in the social service field? What is your job title? What are the job responsibilities for someone in your position? Do you experience stress from you job? Construct of Interest General Question Job Stress What do you feel is the most significant cause of job stress for you? What changes are needed to reduce your job stress? Emotionally-Based Behavior Are you required to display emotions on the job that you do not feel? If so, how does faking emotions on the job make you feel? Perceived Customer Demands In what ways do you believe that customers wants and needs are similar or different from what management believes the customer wants or needs? If there is one thing you could change about handling customer demands, what would it be? Job Autonomy When it comes to making decisions on the job, how much freedom do you have? What would make your decision making process easier?

PAGE 249

Appendix C: Pilot Study Inter-Item Correlations Perceived Customer Demands Indicator Item PCD1 PCD2 PCD3 PCD4 PCD5 PCD6 1. Level of overall service expected 1.000 2. Level of service quality expected .776 1.000 3. Level of service reliability expected .784 .887 1.000 4. Overall expectations for service .811 .861 .834 1.000 5. Customers expectation that the services offered meets his or her needs .634 .608 .630 .605 1.000 6. Expectations of specific level of service delivery quality .716 .678 .723 .684 .642 1.000 Significant at <0.01 level (2-tailed) Role Ambiguity Indicator Item RA1 RA2 RA3 RA4 RA5 RA7 RA8 RA9 RA11 RA12 RA13 1. Managements expectation for interaction with customers 1.000 2. The amount of service to be provided .722 1.000 3. The behavior expected by management for frontline social service personnel with customers on the job .672 .546 1.000 4. How to handle customer objections .550 .457 .673 1.000 5. How to handle unusual problems and situations .422 .409 .613 .696 1.000 7. Which specific strengths to point out to customers .480 .490 .607 .591 .536 1.000 8. Which specific benefit to highlight for customers .519 .471 .655 .538 .529 .880 1.000 236

PAGE 250

Appendix C (Continued) Indicator Item RA1 RA2 RA3 RA4 RA5 RA7 RA8 RA9 RA11 RA12 RA13 9. What actions to take in meeting customer needs .672 .671 .617 .723 .577 .541 .727 1.000 11. The amount of work frontline social service personnel are expected to do .481 .540 .608 .586 .583 .591 .484 .584 1.000 12. Which cases to give priority .375 .450 .734 .559 .594 .550 .613 .683 .558 1.000 13. How much work you are expect to do .488 .644 .569 .529 .485 .446 .492 .616 .728 .594 1.000 Significant at <0.01 level (2-tailed) Role Conflict Indicator Item RC1 RC2 RD3 RD4 RC5 RD6 RD7 RC8 RC9 1. Having to do things that he or she believes should be done differently 1.000 2. The need to break company policy rules to carry out an assignment .612 1.000 3. Having to work on unnecessary things .535 1.000 .629 .854 4. Perceiving requests as incompatible from one or more people .619 .739 1.000 5. Doing what will be accepted by one person but not by another .629 .655 .577 .583 1.000 6. Working with two or more groups who operate very differently .421 .528 .669 .674 .715 1.000 7. Receiving assignments without the manpower to complete them .484 .586 .539 .552 .679 .745 1.000 8. Receiving assignments without adequate resources and materials to complete them .476 .522 .540 .544 .634 .693 .840 1.000 9. Receiving assignments without being adequately trained .597 .582 .540 .628 .539 .583 .588 .677 1.000 Significant at <0.01 level (2-tailed) 237

PAGE 251

Appendix C (Continued) Role Overload Indicator Item RO5 RO1 RO2 RO3 RO4 1. The job requires him or her to work very fast 1.000 2. The job requires him or her to work very hard .702 1.000 3. The job leaves little time to get things done .629 .601 1.000 4. The job requires a great deal of work to be done .464 .591 .645 1.000 5. The job requires more work than can be done well Indicator Item .511 .360 .663 .531 1.000 Significant at <0.01 level (2-tailed) Emotional Labor Display ELD2 ELD3 ELD4 ELD5 ELD6 2. You must pretend to have emotions that you dont really have 1.000 3. There is an expectation that you will hide your true feelings about a situation .729 1.000 4. There is the expectation that you must make an effort to actually feel an emotion that management believes you should feel .536 .619 1.000 5. There is the expectation that you will try to actually experience an emotion that you are supposed to show .461 .541 .903 1.000 6. There is the expectation that you will really try to feel emotions that are deemed part of your job .441 .500 .686 .689 1.000 Significant at <0.01 level (2-tailed) 238

PAGE 252

Appendix C (Continued) Emotional Labor Effort Indicator Item ELE2 ELE3 ELE4 ELE5 ELE6 1. Faking emotions you show the customer 1.000 2. Talking yourself out of feeling what you really feel when helping customers .654 1.000 3. Conjuring up feelings you need to show the customer .742 .638 1.000 4. Changing your actual feelings to match the feeling you must express to customers .544 .496 .755 1.000 5. Attempting to create emotions in yourself that present an image the company desires .366 .488 JA1 .574 .725 1.000 Significant at <0.01 level (2-tailed) Job Autonomy Indicator Item JA2 JA3 1. Ability to decide how to go about doing his or her work 1.000 2. Ability to use his or her personal initiative or judgment in carrying out his or her job .865 1.000 3. Opportunity for independence and freedom on how to do the job Significant at <0.01 .660 level (2-tailed) .765 1.000 239

PAGE 253

Appendix C (Continued) Emotional Intelligence Perception Indicator Item EIP3 EIP1 EIP2 EIP4 EIP5 EIP6 EIP7 EIP8 EIP9 P1. Recognizing emotions present in a particular situation 1.000 P2. Figuring out the reason behind different emotions .596 1.000 P3. Differentiating between the emotions you experience .634 .582 1.000 P4. Thinking about the emotions behind your actions .634 .523 .786 1.000 P5. Awareness of how your feelings affect you .639 .501 .713 .768 1.000 P6. Acknowledging feelings of other at work .593 .422 .559 .636 .757 1.000 P7. Examining the feelings, thoughts, and actions of others .661 .654 .692 .667 .538 .521 1.000 P8. Attempting to understand why other people feel the way they do .534 .585 .530 .541 .558 .583 .682 1.000 P9. Observing how others react to you in an effort to understand your own behavior .559 .554 .554 .577 .624 .706 .682 .715 1.000 Significant at <0.01 level (2-tailed) 240

PAGE 254

Appendix C (Continued) Emotional Intelligence Management Indicator Item EIM2 EIM3 EIM4 EIM5 EIM6 EIM9 M2. Composure even when feeling angry 1.000 M3. Expecting that you will succeed in most endeavors you take on .373 1.000 M4. Reacting to challenges .326 .600 1.000 M5. Failing or succeeding to control your emotions .402 .559 .421 1.000 M6. Seeking out uplifting activities .341 .400 .449 .248 1.000 M9. Handling conflict .344 .360 .349 .382 .329 1.000 Significant at <0.01 level (2-tailed) 241

PAGE 255

Appendix C (Continued) Emotion-Focused Coping Indicator Item EFC1 EFC2 EFC3 EFC4 1. Turn to hobbies and pastimes 1.000 2. Talk to understanding friends .499 1.000 3. Expand interests and activities outside of work .779 .465 1.000 4. Seek social support .363 .645 .439 1.000 Significant at <0.01 level (2-tailed) Job Stress Indicator Item JSA1 JSA2 JSA3 JSA4 JSA5 JSTP1 JSTP2 JSTP3 JSTP5 A1. Feeling fidgety or nervous as a result of the job 1.000 A 2. The job gets to him or her more than it should .782 1.000 A 3. On the job there are a lot of thing that drive him or her right up a wall .614 .785 1.000 A 4. Thinking about the job creates a tight feeling in his or her chest .598 .671 .708 1.000 A 5. He or she feels guilty when taking time off from work .538 .511 .383 .574 1.000 T1. Too much work to do and too little time to do it .492 .583 .526 .530 .504 1.000 T2. Burnout caused by excessive job demands .498 .577 .605 .496 .441 .809 1.000 T3. Dreading the telephone ringing at home because the call might be job-related .433 .521 .496 .580 .387 .398 .502 1.000 T5. Feeling that he or she is married to the job .558 .608 .558 .674 .538 .568 .540 .675 1.000 Significant at <0.01 level (2-tailed) 242

PAGE 256

Appendix C (Continued) Physical Consequences Indicator Item PC12 PC3 PC6 PC7 PC8 PC13 PC14 PC15 PC16 PC18 3. Acid indigestion or heart burn 1.000 6. Chest pain .625 1.000 7. Constipation .433 .515 1.000 8. Diarrhea .441 .648 .667 1.000 12. Headache .525 .660 .429 .423 1.000 13. Heart pounding when not exerting .516 .827 .540 .616 .581 1.000 14. Loss of appetite .423 .631 .588 .621 .432 .634 1.000 15. Shortness of breath .485 .778 .600 .690 .412 .776 .747 1.000 16. Stomach cramps (not menstrual) .542 .671 .482 .648 .459 .600 .609 .745 1.000 18. Trouble sleeping .511 .633 .366 .522 .646 .580 .430 .500 .547 1.000 Significant at < 0.05 level (2-tailed) Attitude Toward the Job t at < 0.05 level (2-tailed) ** Significant at <0.01 level (2-tailed) Significan Indicator Item ATJ2 ATJ3 ATJ4 ATJ5 ATJ6 ATJ7 ATJ8 2. Being enthusiastic about your job 1.000 3. Perceiving your job as pleasant .441** 1.000 4. Finding the job interesting .216* .429** 1.000 5. Getting the feedback from your boss that you expect .330** .456** .494** 1.000 6. Communication with the boss .466** .536** .403** .727** 1.000 7. Overall working conditions .461** .518** .274* .438** .558** 1.000 8. Morale (in other words is the job uplifting or dismal) .430** .639** .409** .403** .490** .702** 1.000 243

PAGE 257

Appendix C (Continued) Overall Job Satisfaction Indicator Item OJS3 OJS4 OJS5 OJS6 3. Judging his or her job against his or her ideal job 1.000 4. The measurement of his or her expectations when he or she took the job against the actual job .584 1.000 5. All things considered, the level of satisfaction with his or her current job .492 .639 1.000 6. His or her general feeling about his or he job .523 .629 .946 1.000 Significant at < 0.05 level (2-tailed) Job Performance Indicator Item JP3 JP5 JP11 JP12 JP13 JP14 JP15 3. Overall, developing customer trust and confidence in your service 1.000 5. Consistency in following up on promises made to your customers .774 1.000 11. Telling the customer the straight facts instead of telling them what they want to hear .416 .444 1.000 12. Overall performing your job dependably and accurately .522 539 .652 1.000 13. Servicing the account with the customers best interest in mind .585 .607 .652 .832 1.000 14. Listening attentively to identify and understand the concerns of the customers .464 .477 .611 .809 .856 1.000 15. Working out solutions to each customers questions or concerns .495 .585 .612 .760 .796 .777 1.000 Significant at < 0.05 level (2-tailed) 244

PAGE 258

Appendix C (Continued) Intention to Switch Positions Indicator Item ITS1 ITS2 1. Actively looking for another position within the organizations 1.000 2. Planning to switch positions within the organization .854 1.000 Significant at < 0.05 level (2-tailed) Intention to Leave Indicator Item ITL1 ITL2 ITL3 1. Actively looking for a job with another organization 1.000 2. Planning to switch companies .866 1.000 3. Planning to stay with your current employer .467 .501 1.000 Significant at < 0.05 level (2-tailed) 245

PAGE 259

Appendix D Pilot Study Item-to-Total Correlation Scale Development Analysis Perceived Customer Demands (PCD) Role Ambiguity (RA) Role Conflict (RC) Role Overload (RO) Scale Item Item-Total Correlation Scale Item Item-Total Correlation Scale Item Item-Total Correlation Scale Item Item-Total Correlation 1 .846 1 .681 1 .664 1 .703 2 .867 2 .684 2 .705 2 .669 3 .883 3 .814 3 .751 3 .792 4 .865 4 .730 4 .758 4 .671 5 .690 5 .682 5 .783 5 .617 6 .773 7 .734 6 .785 8 .756 7 .783 9 .827 8 .768 11 .719 9 .729 12 .725 13 .707 246

PAGE 260

Appendix D (Continued) Emotional Labor Display (ELD) Emotional Labor Effort (ELE) Emotional Intelligence Management (EIM) Emotional Intelligence Perception (EIP) Scale Item Item-Total Correlation Scale Item Item-Total Correlation Scale Item Item-Total Correlation Scale Item Item-Total Correlation 2 .636 1 .683 2 .438 1 .694 3 .713 2 .672 3 .454 2 .608 4 .829 3 .832 4 .460 3 .725 5 .775 4 .765 5 .437 4 .698 6 .675 5 .632 6 .408 5 .712 9 .384 6 .696 7 .757 8 .633 Job Autonomy (JA) Scale Item Item-Total Correlation 2 .814 3 .895 4 .739 9 .703 Job Stress Anxiety (JSA) Job Stress Time Pressure (JSTP) Emotion-Focused Coping (EFC) Attitude Toward the Job (ATJ) Scale Item Item-Total Correlation Scale Item Item-Total Correlation Scale Item Item-Total Correlation Scale Item Item-Total Correlation 1 .719 1 .686 1 .648 2 .508 2 .812 2 .699 2 .661 3 .679 3 .742 3 .632 3 .670 4 .494 4 .783 5 .761 4 .576 5 .654 5 .607 6 .732 7 .661 8 .678 247

PAGE 261

Appendix D (Continued) Overall Job Satisfaction (OJS) Physical Consequences (PC) Job Performance (JP) Intention to Switch Positions (ISP) Scale Item Item-Total Correlation Scale Item Item-Total Correlation Scale Item Item-Total Correlation Scale Item Item-Total Correlation 3 .591 3 .627 3 .632 1 .854 4 .707 6 .862 5 .683 2 .854 5 .794 7 .645 11 .665 6 .804 8 .744 12 .833 12 .637 13 .883 13 .810 14 .802 14 .721 15 .809 15 .819 16 .751 18 .665 Intention to Leave (ITL) Scale Item Item-Total Correlation 1 .780 2 .814 3 .501 248

PAGE 262

Appendix E Pilot Study Summary of Factor Analysis Results Scale Scale Composite Mean SEM Scale Weighted Mean SEM Min. Max. Alpha Sampling Adequacy Variance Original Scale Items Final Scale Items Perceived Customer Demands (PCD) 35.43 .723 5.91 .121 2 7 .939 .906 77.359 6 6 Role Ambiguity 53.95 1.789 4.50 .149 1 7 .937 .868 61.606 17 11 Role Conflict 42.85 1.542 4.76 .171 1 7 .932 .888 64.883 9 9 Role Overload 28.45 .665 5.69 .133 1 7 .860 .770 65.719 5 5 Emotional Labor Display 16.09 .827 3.22 .165 1 7 .885 .777 69.137 6 5 Emotional Labor Effort 17.05 .864 3.41 .173 1 7 .881 .778 68.100 5 5 Job Autonomy 16.50 .556 5.50 .185 1 7 .907 .684 84.328 3 3 Emotional Intelligence 87.337 1.262 5.82 .084 1 7 .902 .873 59.480 19 15 Job Stress 41.33 1.586 4.59 .176 1 7 .917 .846 61.415 10 9 EmotionFocused Coping 21.73 .589 5.43 .147 1 7 .815 .650 64.991 4 4 249

PAGE 263

Appendix E (Continued) Scale Scale Composite Mean SEM Scale Weighted Mean SEM Min. Max. Alpha Sampling Adequacy Variance Original Scale Items Final Scale Items Physical Consequences 38.15 1.766 3.82 .177 1 7 .931 .898 61.969 18 10 Attitude Toward the Job 40.65 54.894 .725 5.81 .104 3 7 .858 .819 8 7 Overall Job Satisfaction 22.57 .497 5.64 .124 2 7 .862 .707 73.117 6 4 Job Performance 42.57 7 .816 6.08 .117 2 7 .920 .886 68.948 16 Intention to Switch Positions 8.88 .450 2 4.44 .225 1 7 .919 .500 92.711 3 Intention to Leave 13.14 .620 4.38 .207 1 7 .829 .624 74.770 3 3 250

PAGE 264

Appendix F Pilot Study Item Indicator Retention Perceived Customer Demands (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 Level of overall service expected 5.94 1.240 -1.479 0.898 2 Level of service quality expectedLevel of service reliability e -1.770 xpected 5.95 5 Skewness 5.92 1.304 0.918 3 1.207 -1.553 0.927 4 Overall expectation for service 5.97 1.202 -1.844 0.915 Customers expectation that the services offered meets his or her needs 5.87 1.353 -1.542 0.770 6 Expectations of specific level of service delivery quality 5.78 1.35 -1.556 0.839 Role Ambiguity (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Std.Dev. Factor Loading 1 Managements expectation for interaction with customers 4.41 1.811 0.047 0.738 2 The amount of service to be provided 4.83 1.836 -0.449 0.739 3 The behavior expected by management for frontline social service personnel with customers on the job 4.37 1.885 -0.158 0.854 4 How to handle customer objections 4.27 1.683 -0.070 0.783 5 How to handle unusual problems and situations 4.92 1.675 -0.592 0.741 6 How to deal with customer criticism 4.36 1.762 -0.120 X 7 Which specific strengths to point out to customers 3.92 1.867 -0.023 0.787 8 Which specific benefit to highlight for customers 3.97 1.831 -0.018 0.805 9 10 What actions to take in meeting customer needs 4.27 1.765 -0.182 0.863X How to handle non-routine activities on the job 4.56 1.760 -0.424 251

PAGE 265

Appendix F (Continued) Role Ambiguity (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items 5.19 1.913 Mean Skewness Std.Dev. Factor Loading 11 The amount of work frontline social service personnel are expected to do 1.641 -0.502 0.770 12 Which cases to give priority 4.56 -0.334 0.781 13 How much work you are expect to do 4.91 1.786 -0.427 0.761 14 How free time should be handled on the job 3.78 2.089 1.926 0.190 X 15 What to do to get a promotion 4.09 -0.084 X 16 Vulnerability to being terminated 3.84 2.185 0.054 X X 17 Not knowing what are the critical factors in getting promoted 3.99 2.026 0.025 Role Conflict (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 Having to do things that he or she believes should be done differently 4.86 1.610 -0.686 0.728 2 The need to break company policy rules to carry out an assignment 4.17 0.821 4.98 2.138 -0.233 0.770 3 Having to work on unnecessary things 4.91 1.938 -0.610-0.289 0.809 4 Perceiving requests as incompatible from one or more people 4.64.38 1.830 0.815 5 Doing what will be accepted by one person but not by another 2.0532.006 -0.309 0.837 6 Working with two or more groups who operate very differently 4.63 -0.528 0.838 7 Receiving assignments without the manpower to complete them 5.08 2.036 -0.849 0.834 8 Receiving assignments without adequate resources and materials to complete them 5.23 2.010 -0.924 9 Receiving assignments without being adequately trained 2.114 -0.741 0.789 252

PAGE 266

Appendix F (Continued) Role Overload (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 The job requires him or her to work very fast 5.49 1.643 -1.298 0.818 2 The job requires him or her to work very hard 5.94 1.392 -1.368 0.806 3 The job leaves little time to get things done 5.646.08 1.564 -1.267 0.878 4 The job requires a great deal of work to be done 1.150 -1.302 0.796 5 The job requires more work than can be done well 5.30 1.867 -1.076 0.749 Emotional Labor Display (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 You believe that you must hold back your true feelings 4.48 1.787 -0.537 X 2 You must pretend to have emotions that you dont really have 3.24 0.878 1.940 0.339 0.747 3 There is an expectation that you will hide your true feelings about a situation 3.56 1.974 0.104 0.808 4 There is the expectation that you must make an effort to actually feel an emotion that management believes you should feel 2.95 1.821 0.609 0.913 5 There is the expectation that you will try to actually experience an emotion that you are supposed to show 2.87 1.720 0.614 6 There is the expectation that you will really try to feel emotions that are deemed part of your job 3.47 1.807 0.058 0.800 253

PAGE 267

Appendix F (Continued) Emotional Labor Effort (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 Faking emotions you show the customer 3.08 1.886 0.515 0.804 2 Talking yourself out of feeling what you really feel when helping customers 3.73 1.943 0.122 0.790 3 Conjuring up feelings you need to show the customer 3.42 1.943 0.195 0.907 4 Changing your actual feelings to match the feeling you must express to customers 3.42 1.949 0.277 0.858 5 Attempting to create emotions in yourself that present an image the company desires 3.4 2.013 0.316 0.758 Job Autonomy (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 Ability to decide how to go about doing his or her work 5.34 1.864 -1.088 0.919 2 Ability to use his or her personal initiative or judgment in carrying out his or her job 5.52 1.926 -1.315 0.958 3 Opportunity for independence and freedom on how to do the job 5.64 1.821 -1.481 0.876 254

PAGE 268

Appendix F (Continued) Emotional Intelligence (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading P1 Recognizing emotions present in a particular situation 6.16 1.039 -1.560 0.800 P2 Figuring out the reason behind different emotions 5.76 1.147 -0.844 0.740 P3 Differentiating between the emotions you experience 5.69 1.258 -1.341 0.838 P4 Thinking about the emotions behind your actions 5.81 1.297 -1.234 0.847 P5 Awareness of how your feelings affect you 6.16 1.094 -1.712 0.839 P6 Acknowledging feelings of other at work 5.95 1.217 -1.113 0.789 P7 Examining the feelings, thoughts, and actions of others 5.87 1.196 -1.184 0.838 P8 Attempting to understand why other people feel the way they do 5.93 1.272 -1.411 0.785 P9 Observing how others react to you in an effort to understand your own behavior 5.77 1.262 -1.058 0.815 M1 When frustrated or angry you consider options available 5.91 0.903 -0.403 X M2 Composure even when feeling angry 5.84 1.061 -0.936 0.635 M3 Expecting that you will succeed in most endeavors you take on 5.67 1.202 -1.254 0.809 M4 Reacting to challenges 5.56 1.184 -1.338 0.766 M5 Failing or succeeding to control your emotions 5.41 1.426 -1.105 0.723 M6 Seeking out uplifting activities 5.73 1.212 -1.375 0.639 M7 Arranging events that others enjoy 5.09 1.476 -0.972 X M8 Helping others feel better when they are down 5.86 1.097 -1.139 X M9 Handling conflict 6.02 1.106 -1.329 0.634 M10 Involving yourself in other peoples problems 4.66 1.649 -0.472 X 255

PAGE 269

Appendix F (Continued) Emotion-Focused Coping (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 Turn to hobbies and pastimes 5.67 1.560 -1.456 0.830 2 Talk to understanding friends 5.51 1.727 -1.157 0.804 3 Expand interests and activities outside of work 5.70 1.602 -1.494 0.842 4 Seek social support 4.85 1.907 -0.571 0.744 Job Stress (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading A1 Feeling fidgety or nervous as a result of the job 4.37 2.098 -0.329 0.787 A2 The job gets to him or her more than it should 4.59 3 T1 4.29 2.043 -0.582 0.865 A3 On the job there are a lot of thing that drive him or her right up a wall 4.58 2.095 -0.534 0.814 A4 A5 Thinking about the job creates a tight feeling in his or her chest He or she feels guilty when taking time off from work 3.91 2.194 0.025 0.8310.68 4.33 2.257 -0.329-1.395 Too much work to do and too little time to do it 5.835.83 1.716 0.767 T2 Burnout caused by excessive job demands 1.576 -1.576 0.775 T3 Dreading the telephone ringing at home because the call might be job-related 3.67 2.427 0.194 0.703 T4 The feeling that he or she never has a day off 3.88 4.79 X T5 Feeling that he or she is married to the job 4.22 2.393 -0.183 0.812 256

PAGE 270

Appendix F (Continued) Physical Consequences (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 Backache 4.36 2.092 -0.312 X 2 5.00 Skin rash 3.014.23 1.937 0.609 X 3 Acid indigestion or heart burn 2.157 -0.316 0.691 4 An infection 3.21 1.9892.072 0.511 X 5 Upset stomach or nausea 4.22 -0.246 X 6 Chest pain 3.95 2.163 -0.060 0.896 7 Constipation 2.93 1.921 0.591 0.711 8 Diarrhea 3.12 2.026 0.525 0.802X 9 Dizziness 2.92 1.936 0.795 10 Eye strain 4.272.65 2.020 -0.261 X 11 Fever 1.858 0.764 X 12 Headache 2.1422.194 -0.794 0.699 13 Heart pounding when not exerting 3.72 0.105 0.858 14 Loss of appetite 3.52 2.085 0.317 0.783 15 Shortness of breath 3.23 2.118 0.493 0.869 16 Stomach cramps (not menstrual) 3.28 2.107 0.418 0.807 17 Tiredness or fatigue 5.59 1.811 -1.486 X 18 Trouble sleeping 5.16 1.939 -0.968 0.724 257

PAGE 271

Appendix F (Continued) Attitude Toward the Job (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 Trying to maintain a positive attitude toward customers 5.91 1.411 -1.58 X 2 Being enthusiastic about your job 6.00 5 plifting or dismal) 1.063 -1.505 0.63 3 Perceiving your job as pleasant Finding the job interesting 5.79 1.128 -0.731 0.784 4 5.99 1.333 -1.81 0.602 5 Getting the feedback from your boss that you expect 5.44 1.508 -1.214 0.746 6 Communication with the boss 5.58 1.545 -1.302 0.818 7 Overall working conditions Morale (in other words is the job u 5.85 1.153 -0.922 0.776 8 6.00 1.337 -1.874 0.797 Overall Job Satisfaction (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 If given a chance to decide all over again, knowing what he or she knows now, he or she would take the job if it were offered 5.77 1.493 -1.72 X 2 If a friend applies for a job like his or hers with the same employer, he or she would recommend the job to him or her 5.55 1.614 -1.445 X 3 Judging his or her job against his or her ideal job 5.23 1.614 -0.698 0.737 4 The measurement of his or her expectations when he or she took the job against the actual job 5.34 1.377 -0.744 0.829 5 All things considered, the level of satisfaction with his or her current job 6.02 1.208 -1.646 0.918 6 His or her general feeling about his or he job 5.98 1.246 -1.562 0.923 258

PAGE 272

Appendix F (Continued) Job Performance (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 Taking the initiative to help your customers even when it is not part of your responsibility 5.63 1.624 -1.518 X 2 Taking the time to help your customers at the expense of not meeting daily productivity goals 5.36 1.202 -2.034 1.608 -0.975 X 3 Overall, developing customer trust and confidence in your service 5.97 1.483 -1.756 0.716 4 Responding in a timely manner to customer requests despite your busy schedule 6.00 1.138 -1.473 X 5 Consistency in following up on promises made to your customers 6.066.05 -1.489 0.748 6 7 Consistency in providing prompt service to all of your customers Consistency in resolving customer concerns the first time 1.105 1.234 -1.379 X 5.77 -1.044 X 8 Consistency in demonstrating emotionally-based behavior deemed appropriate by the company 5.41 1.537 -0.995 X 9 Providing accurate or correct information to the customer 6.33 1.011 -2.236 X 10 Even thought it is not your responsibility, making sure other departments follow through with your customers requests 5.38 1.667 -0.930 X 11 Telling the customer the straight facts instead of telling them what they want to hear 5.92 1.465 -1.695 0.754 12 Overall performing your job dependably and accurately 6.29 1.245 -2.672 0.892 13 Servicing the account with the customers best interest in mind 6.10 1.265 -2.128 0.928 14 Listening attentively to identify and understand the concerns of the customers 6.21 1.294 -2.271 0.874 15 Working out solutions to each customers questions or concerns 6.02 1.236 -1.957 0.875 16 Overall, providing individualized attention to each customers concerns 6.07 1.344 X 259

PAGE 273

Appendix F (Continued) Intention to Switch (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Skewness Std.Dev. Factor Loading 1 Actively looking for another position within the organizations 4.45 2.273 -0.390 0.854 2 Planning to switch positions within the organization 4.43 Skewness 2.061 -0.385 0.854 3 Planning to keep your position within the organization 4.15 1.85 -0.215 X Intention to Leave (Continuous Scale with anchors of: 1 = not at all a factor and 7 = definitely a factor) Items Mean Std.Dev. Factor Loading 1 Actively looking for a job with another organization 4.71 2.386 -0.578 0.922 2 Planning to switch companies 4.48 2.2272.034 -0.427 0.933 3 Planning to stay with your current employer 3.95 0.013 0.723 260

PAGE 274

Appendix G Exploratory Factor Analysis Scale Purification Bartletts Test of Sphericity Sampling Adequacy Construct Smallest Sampling Adequacy Line Item Total Variance Factor Loading Scale Reliability Factors Line Items Original Scale Line Items Retained Perceived Customer Demands 0.000 0.864 .836 77.809 >.838 0.904 1 6 4 Role Ambiguity 0.000 0.843 .713 69.141 > .613 0.862 3 11 10 Role Conflict 0.000 0.749 .707 65.926 >.766 0.828 1 9 4 Role Overload 0.000 0.887 0.801 .757 69.106 >.759 1 5 5 Emotional Labor Display 0.000 >.631 0.730 .673 77.445 0.788 2 5 5 Emotional Labor Effort 0.000 .822 0.842 64.232 >.753 0.859 1 5 5 Job Autonomy N/A N/A N/A N/A N/A 0.906 1 3 3 Emotional Intelligence 0.000 0.666 .583 71.267 >.619 0.701 3 15 7 Job Stress 0.000 0.763 .720 69.215 >.721 0.807 2 9 6 Emotion-Focused Coping Frequency 0.000 4 0.683 .668 62.999 >.757 0.803 1 4 Emotion-Focused Coping Effectiveness of Use 0.000 0.649 .630 N/A N/A 0.781 1 4 4 261

PAGE 275

Appendix G (Continued) Bartletts Test of Sphericity Sampling Adequacy Construct Sampling Adequacy Line Item Total Variance Factor Loading Scale Reliability Factors Line Items Original Scale Line Items Retained Physical Consequences 0.000 0.779 .786 68.680 >.706 .802 3 10 8 Attitude Toward the Job 0.000 0.771 .715 64.582 >.629 0.737 2 7 6 Organizational Commitment 0.000 0.850 .795 60.652 >.710 0.868 1 9 6 Overall Job Satisfaction 0.000 0.842 0.807 .763 68.879 >.745 1 4 4 Emotional Exhaustion Frequency 0.000 0.803 .754 65.514 >.588 0.810 2 9 7 Emotional Exhaustion Intensity 0.000 .782 .730 62.297 >.756 0.796 1 9 4 Job Performance 0.000 0.832 .809 63.160 >.759 0.851 1 7 5 Intention To Switch 0.000 0.500 .500 N/A N/A 0.881 1 2 2 Intention To Leave 0.000 0.655 .600 80.055 >.793 0.874 1 3 3 262

PAGE 276

Appendix H Exploratory Factor Analysis Item Indicator Retention Perceived Customer Demands (Continuous Scale with anchors of: 1 = not at all demanding and 7 = extremely demanding) Items Mean Skewness Std.Dev. Factor Loading Factor A In terms of service, the customers I serve are 4.74 1.415 -0.485 X X B In terms of quality, the customers I serve are 4.72 1.505 -0.498 0.838 1 C In terms of reliability, the customers I serve are 4.67 1.547 -0.280 X X D Customers expectations for service are 5.23 1.418 -0.766 0.897 1 E Customers expectations that the services offered will meet his or her needs are 5.37 1.492 -0.954 0.899 1 F Customers expectations for delivery level of service quality are 5.32 1.378 -0.858 0.893 1 Role Ambiguity (Symmetrical Scale with anchors of: 1 = very uncertain and 7 = very certain) Items Mean Skewness Std.Dev. Factor Loading Factor A How I am expected to interact with my customers 1.43 0.729 2.801 0.773 3 B How much service I should provide my customers is 1.82 0.973 1.696 1.724 0.613 1 C How I should behave (with customers) while on the job is 1.29 0.539 0.862 3 D How I am expected to handle my customers objections is 1.88 0.929 1.406 0.624 1 E How I am expected to handle unusual problems and situations is 2.18 1 1.097 1.252 0.764 1 F Which specific company strengths I should present to customers is 2.11 1.043 1.215 0.876 G Which specific service benefits I am expected to highlight for customers is 2.05 1.031 1.106 0.779 1 263

PAGE 277

Appendix H (Continued) Role Ambiguity (Symmetrical Scale with anchors of: 1 = very uncertain and 7 = very certain) Continued Items Mean Skewness Std.Dev. Factor Loading Factor H The actions required in meeting customers needs are I 1.93 1.028 1.351 0.830 1 I The amount of work I am expected to do is I 1.85 1.105 1.797 0.848 2 J Which tasks I should give priority is I How much work I am expected to do is 2.21 1.170 1.409 X X K I 2.15 1.388 1.675 0.952 2 Role Conflict (Likert Scale with anchors of: 1 = strongly disagree and 7 = strongly agree) Items Mean Skewness Std.Dev. Factor Loading Factor A I have to do things that I believe should be done differently. 3.88 1.838 -.021 .766 1 B I have to break company policy rules in order to carry out an assignment. 1.939 D le requests from either customers or my supervisors. I 2.45 1.538 1.077 X X C I work on unnecessary things. I usually receive incompatib 3.45 .320 X X 3.33 1.765 .440 .774 1 E I do things that are apt to be accepted by some people and not accepted by others. 4.24 1.841 -.256 X X F I work with two or more groups who operate quite differently. 4.54 1.985 -.466 X X G I receive assignments without the manpower to complete them. 3.86 2.011 .125 .844 1 H I receive assignments without adequate resources/material to execute them. 3.75 2.033 .205 .859 1 I receive assignments for which I am not adequately trained. 2.58 1.676 1.051 X X 264

PAGE 278

Appendix H (Continued) Role Overload (Frequency scale ranging from 1=never to 7=always) Items Mean Skewness Std.Dev. Factor Loading Factor A My job requires me to work very fast. 4.16 1.541 0.077 0.813 1 B My job requires me to work very hard. 4.87 1.543 -0.358 0.865 1 C My job leaves me with little time to get things done. 3.84 1.548 Emotional Labor Display (Frequency scale ranging from 1=never to 7=always) During any service encounter 0.467 0.8610.854 1 D There is a great deal of work to be done. 5.09 1.626 -0.480 1 E I have to do more work than I can do well. 3.31 1.662 0.671 0.759 1 Items Mean Skewness Std.Dev. Factor Loading Factor A I am required to pretend to have emotions that I dont really have. 2.09 1.142 1.401 0.764 2 B I am required to hide my true feelings about a situation. 2.90 1.513 1 1.438 0.813 0.941 2 C I am expected to make an effort to actually feel the emotions that management believes I should feel. 2.10 1.342 0.631 1 D I am expected to try to actually experience the emotions that I must show on the job. 2.08 1.440 1.694 0.913 E I am expected to really try to feel the emotions I have to show as part of my job. 2.27 1.529 1.379 0.923 1 265

PAGE 279

Appendix H (Continued) Emotional Labor Effort (Continuous scale ranging from 1 = no effort 7= extreme amount of effort) Items Mean Skewness Std.Dev. Factor Loading Factor A Fake the emotions I show customers. 4.01 2.032 0.037 0.756 1 B Talk myself out of feeling what I really feel when helping customers. 3.73 1.834 0.147 0.840 1 C Summon up the feelings I need to show to customers. 3.00 1.714 0.720 0.753 1 D Change my actual feelings to match those that I must express to customers. 3.53 1.823 0.330 0.847 1 E Attempt to create certain emotions in myself that present the image my organization desires. 3.52 1.940 0.327 0.806 1 Emotional Intelligence (Semantic Differential Scale with bipolar end points of 1 and 7) Items Mean Skewness Std.Dev. Factor Loading Factor A Recognizing emotions that I experience in a particular situation is 5.72 1.261 -1.338 -.619 3 B My ability to figure out the reasons behind my different emotions is 5.27 1.552 -.932 -.926 2 C Differentiating between emotions I experience is 5.25 1.455 .943 -.861 2 D I think about the emotions underlying my actions 4.73 1.515 -.327 X X E When it comes to how my feelings are affecting me, I am 5.46 1.244 -.947 X X F Generally when I feel angry, I am 5.43 .917 -.677 .885 1 G On most things I try, I expect to 6.03 .844 -1.578 X X H When taking on challenges where there is a strong chance that I may fail, I feel 4.46 1.234 .195 X X 266

PAGE 280

Appendix H (Continued) Emotional Intelligence (Semantic Differential Scale with bipolar end points of 1 and 7) Continued Items Mean Skewness Std.Dev. Factor Loading Factor I My ability to control my emotions is 5.66 .935 -.694 .835 1 J I seek out activities that make me 5.85 1.078 -.945 X X K When it comes to other peoples feelings at work, acknowledging their feelings is 5.80 1.248 -1.320 -.678 3 L Ones ability to understand why other people feel the way they do is 6.53 .817 -2.685 X X M Examining the feelings, thoughts, and actions of others is 5.47 1.253 -.876 -858 3 N Observing how other people react to me helps me better understand my own behavior 5.56 1.520 -1.378 X X O My handling of conflict is 5.18 1.520 -.756 X X Job Stress (7 Point Likert Scale 1 = Strongly Disagree and 7 = Strongly Agree) Items Mean Skewness Std.Dev. Factor Loading Factor A I often feel fidgety or nervousness as a result of my job. 2.97 1.712 0.564 0.868 1 B My job irritates to me more than it should. 3.1 1.788 0.516 0.864 1 C On the job, there are lots of times when my job drives me right up a wall. 3.36 1.835 0.283 0.749 1 D Sometimes when I think about my job I get a tight feeling in my chest. 2.68 1.786 0.801 0.721 1 E I do not feel guilty when I take time off from the job. 3.52 2.112 0.202 X X F I have too much work to do and too little time to do it. 4.46 1.846 -0.354 X X G Very few frontline social service personnel in my company get burned out because of job demands. 4.84 1.797 -0.566 X X 267

PAGE 281

Appendix H (Continued) Job Stress (7 Point Likert Scale 1 = Strongly Disagree and 7 = Strongly Agree) Continued Items Mean Skewness Std.Dev. Factor Loading Factor H I sometimes dread the telephone ringing at home because the call might be job-related. 2.83 1.887 0.812 0.868 2 I I frequently get the feeling I am married to the company. 3.23 1.941 0.459 0.837 2 Emotion-Focused Coping Extent of Use (Frequency Scale 1 = Never and 7 = Always) Items Mean Skewness Std.Dev. Factor Loading Factor A Turn to hobbies /pastimes. 3.93 1.494 0.323 0.757 1 B Talk to understanding friends. 4.42 1.418 0.168 0.798 1 C Expand interests/activities outside of work. 4.09 1.399 0.361 0.799 1 D Seek social support. 4.15 1.568 0.216 0.820 1 Emotion-Focused Coping Effectiveness (Symmetrical Scale 1 = Definitely Hinders and 7 = Definitely Helps) Items Mean Skewness Std.Dev. Factor Loading Factor A Turn to hobbies /pastimes. 5.91 1.015 -0.736 0.765 1 B Talk to understanding friends. 6.17 0.913 -1.131 0.807 1 C Expand interests/activities outside of work. 5.97 1.054 -1.274 0.795 1 D Seek social support. 5.93 1.166 -1.405 0.756 1 268

PAGE 282

Appendix H (Continued) Physical Consequences (Frequency 1 = Never and 7 = All of the Time) Items Mean Skewness Std.Dev. Factor Loading Factor A Acid indigestion or heartburn 1.90 1.116 1.501 X X B Chest pain 1.39 .714 1.829 -.788 3 C Diarrhea 1.64 .874 1.325 .854 1 D Constipation 1.58 .912 1.668 .816 1 E Headache 2.45 1.183 .729 .889 2 F Heart pounding when not exercising 1.57 .795 1.164 -.835 3 G Loss of appetite 1.51 .911 2.513 .793 2 H Shortness of breath 1.32 .665 3.030 -.787 3 I Stomach cramps (not menstrual) 1.46 .840 2.482 .706 1 J Trouble sleeping 2.72 1.385 .996 X X Attitude Toward the Job (7-point Semantic differential with bipolar ends) Items Mean Skewness Std.Dev. Factor Loading Factor A Being enthusiastic about my job is I 5.35 1.523 -0.91 0.785 1 B My job tends to be 5.17 1.317 -0.926 0.821 1 C My job is usually I 5.77 5.01 1.289 -1.111 X X D Feedback from my superiors is I 1.722 -0.776 0.801 2 E Overall, communications with my boss are 1.38 1.741 -0.426 0.824 2 F Overall working conditions are I 4.89 1.626 -0.528 0.629 1 G Overall I see my job as 4.35 1.714 -0.327 0.847 1 269

PAGE 283

Appendix H (Continued) Organizational Commitment (Likert Scale 1 = Strongly Disagree and 7 = Strongly Agree) Items Mean Skewness Std.Dev. Factor Loading Factor A I am willing to put in a great deal of effort beyond that normally expected in order to help this organization be successful. 5.62 1.343 -1.379 X X B I talk up this organization to my friends as a great organization to work for. 4.99 -0.776 X 5.16 X 6.11 4.95 1 5.21 1 1 1.810 1.689 0.838 1 C I would accept almost any type of job assignment in order to keep working for this organization. 3.72 1.808 0.222 X D I find that my values are very different than the organizations values. I 1.633 -0.664 X E I am embarrassed to tell others that I am part of the organization. I 1.302 -1.714 0.736 1 F This organization really inspires the very best in me in the way of job performance. 1.591 -0.683 0.747 G I am extremely glad that I chose this organization to work for over others I was considering at the time I joined. 1.596 -0.868 0.857 H I do not care about the fate of this organization. I 6.06 1.371 -1.778 0.710 I For me, this is the best of all possible organizations for which to work. 4.56 -0.472 0.774 1 270

PAGE 284

Appendix H (Continued) Overall Job Satisfaction (7-point Semantic differential with bipolar ends) Items Mean Skewness Std.Dev. Factor Loading Factor H Compared to your ideal job, your current job is 4.71 1.639 -0.622 0.745 1 I To what extent does your current job match your expectations when you took it 4.78 1.432 -0.532 0.826 1 J What is your overall satisfaction/dissatisfaction with your current job 4.92 1.587 -0.596 0.856 1 K Your overall feeling about your job would be 5.36 1.327 -0.718 0.886 1 Emotional Exhaustion Frequency (Frequency 0 = never and 7 = 7 days a week) Items Mean Skewness Std.Dev. Factor Loading Factor A I feel emotionally drained from my work. 2.30 1.292 1.188 .900 1 B I feel used up at the end of the workday. 2.47 1.385 .786 .873 1 C When I get up in the morning to face another day on the job I feel tired. 1 .854 .677 2.12 1.153 1.480 .588 D Working with people all day is really a strain for me. 1.98 2.061 X .686 X E I feel burned out from my work. 2.02 1.101 1.914 2 F I feel frustrated with my job. 2.03 1.165 1.798 2 G I feel I am working too hard on my job. 2.39 1.398 1.386 .693 1 X H Working directly with people really puts a strain on me. 1.97 2.06 .880 2.772 X I I feel like Im at the end of my rope with my job. .920 2.670 .853 2 271

PAGE 285

Appendix H (Continued) Emotional Exhaustion Intensity (Semantic differential scale with 1 to 7 points) Items Mean Skewness Std.Dev. Factor Loading Factor A How emotionally draining is your work? 3.60 1.852 -.028 X X B How do you feel at the end of the workday? 4.35 1.521 -.176 .776 1 C How do you feel in the morning knowing you have to face another day on the job? 3.26 X 1.473 .358 .767 1 D E How do you feel after working with people all day? 3.93 1.597 -.029 X .756 X How burned out do you feel from your work? 2.67 1.710 .701 1 F How frustrated are you on the job? 2.88 1.729 .586 X G How hard do you feel you must work on the job? 5.12 1.596 -1.001 X X X H How does working directly with people make you feel? How does your job make you feel? 4.93 1.484 -.630 X I 3.30 1.647 .488 .854 1 Job Performance (Continuous Scale with 1 = Truly Terrible and 7 = Outstanding) Items Mean Skewness Std.Dev. Factor Loading Factor A Developing customer trust/ confidence in your service provided. 5.80 0.928 -1.207 X X B Following up on promises make to your customers. 5.83 0.811 -0.554 X X C Telling the customer the straight facts rather telling them what they want to hear. 5.79 1 0.818 -0.487 0.771 D Performing your job dependably/ accurately. 6.07 0.751 -0.660 0.793 1 E Servicing the account with the customers best interest in mind. 5.99 0.836 -1.018 0.82 1 F Listening attentively to identify as well as understand the concerns of customers. 6.10 0.839 -1.425 0.828 1 G Working out solutions to each customers questions or concerns. 5.74 0.971 -1.265 0.759 1 272

PAGE 286

273 Appendix H (Continued) Intention to Leave (Symmetrical Scale Ranging from 1 = Very Unlikely to 7 = Very Likely) Items Mean Skewness Std.Dev. Factor Loading Factor C I will actively look for a job with anot her company. 2.9 2.108 0.646 0.937 1 D I plan to switch companies. 2.9 2.108 0.658 0.947 1 E I plan to work for my current employer. 2.78 2.013 0.936 0.793 1 Note: = Reversals for all of the above tables

PAGE 287

ABOUT THE AUTHOR Doreen (Dee) Sams is currently an A ssistant Marketing Professor at Georgia College and State University, Mi lledgeville. During her years in the Doctoral program at the University of South Florida (USF), Depa rtment of Marketing, College of Business Administration, she published in Policing: An International J ournal of Police Strategies & Management and the Marketing Educator with proceedings published in: Southern Management Association Conf erence, CHARM Historical Conference, Society for Marketing Advances Conference, 2nd International AmFiTan Development Ethics Conference, and Academy of Marketing Science She studied at the University of Cost a Rica and earned her MBA at USF in 1999. She earned a B.A. in Marketing at USF in 1998. She is a member of Beta Gamma Sigma, Golden Key, Alpha Sigma Lambda, Phi Kappa Phi honor societies and the Deans List of Scholars. Professional membership include: Academy of Marketing Science, American Marketing Association, Associa tion of Consumer Research, Society for Marketing Advancement.