USF Libraries
USF Digital Collections

Personality and work-family conflict

MISSING IMAGE

Material Information

Title:
Personality and work-family conflict the mediational role of coping styles
Physical Description:
Book
Language:
English
Creator:
Bryant, Rebecca H
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Conscientiousness
Extraversion
Neuroticism
Locus of control
Problem solving
Dissertations, Academic -- Psychology -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Although an extensive body of literature exists on the consequences of work-family conflict (WFC), comparatively little research has examined the construct's antecedents. Research on two sets of antecedent variables, personality and coping style, is particularly scarce. Thus, the present study expands the literature by examining four personality variables (conscientiousness, extraversion, neuroticism, and locus of control) and five coping styles (problem solving, support seeking, positive cognitive restructuring, rumination, and escape) in relation to work-interference-with-family (WIF) and family-interference-with-work (FIW) conflict. Additionally, coping style, which was assessed separately for managing work stressors and for managing family stressors, was examined as a potential mediator between personality and both directions of WFC. Two hundred and four participants, recruited from a snowball approach, completed surveys. Additionally, significant others provided ratings of conscientiousness, extraversion, and neuroticism. Of the four personality variables, only neuroticism related to WIF and FIW. Furthermore, among the hypothesized relationships between coping and WFC, only rumination and escape for work stressors related to WIF, though several cross-domain relationships were observed. Overall, the present study found little support for coping as a mediator between personality and WFC, though there was some evidence that rumination mediated the relationship between neuroticism and WIF. As a supplementary analysis, coping was examined as a moderator between personality and WFC. Theoretical and practical implications, as well as future directions, are discussed.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2009.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Rebecca H. Bryant.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 146 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 - 002029487
oclc - 436924179
usfldc doi - E14-SFE0002908
usfldc handle - e14.2908
System ID:
SFS0027225:00001


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 2200397Ka 4500
controlfield tag 001 002029487
005 20090917122816.0
007 cr mnu|||uuuuu
008 090917s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002908
035
(OCoLC)436924179
040
FHM
c FHM
049
FHMM
090
BF121 (Online)
1 100
Bryant, Rebecca H.
0 245
Personality and work-family conflict :
b the mediational role of coping styles
h [electronic resource] /
by Rebecca H. Bryant.
260
[Tampa, Fla] :
University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains 146 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2009.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
3 520
ABSTRACT: Although an extensive body of literature exists on the consequences of work-family conflict (WFC), comparatively little research has examined the construct's antecedents. Research on two sets of antecedent variables, personality and coping style, is particularly scarce. Thus, the present study expands the literature by examining four personality variables (conscientiousness, extraversion, neuroticism, and locus of control) and five coping styles (problem solving, support seeking, positive cognitive restructuring, rumination, and escape) in relation to work-interference-with-family (WIF) and family-interference-with-work (FIW) conflict. Additionally, coping style, which was assessed separately for managing work stressors and for managing family stressors, was examined as a potential mediator between personality and both directions of WFC. Two hundred and four participants, recruited from a snowball approach, completed surveys. Additionally, significant others provided ratings of conscientiousness, extraversion, and neuroticism. Of the four personality variables, only neuroticism related to WIF and FIW. Furthermore, among the hypothesized relationships between coping and WFC, only rumination and escape for work stressors related to WIF, though several cross-domain relationships were observed. Overall, the present study found little support for coping as a mediator between personality and WFC, though there was some evidence that rumination mediated the relationship between neuroticism and WIF. As a supplementary analysis, coping was examined as a moderator between personality and WFC. Theoretical and practical implications, as well as future directions, are discussed.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: Tammy D. Allen, Ph.D.
653
Conscientiousness
Extraversion
Neuroticism
Locus of control
Problem solving
690
Dissertations, Academic
z USF
x Psychology
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2908



PAGE 1

Personality and Work-Family Conflict: Th e Mediational Role of Coping Styles by Rebecca H. Bryant, M. A. A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Tammy D. Allen, Ph.D. Walter C. Borman, Ph.D. Paul E. Spector, Ph.D. Vicky Phares, Ph.D. Judith B. Bryant, Ph.D. Date of Approval: March 13, 2009 Keywords: conscientiousness, extraversion, neuroticism, locus of control, problem solving, support seeking, cognitive restruct uring, escape, rumination, WIF, FIW, WFC Copyright 2009, Rebecca H. Bryant

PAGE 2

Dedication To my parents, for their uncondition al love and support, and for always encouraging me to reach for the stars. And to my husband, for believing in me more than I believe in myself; I would never have made it this far without his continual encouragement and patience. I am incredibly lu cky to have all three of you in my life.

PAGE 3

Acknowledgements I would like to acknowledge the members of my dissertation committee, for their time, feedback, and guidance. My adviso r, Dr. Tammy Allen, deserves a special acknowledgement, for her constant encouragem ent and support, for being a true mentor, and for being the best advisor that I could ever hope for. Finally, thank you to all the people who participated in my study, as well as those who assisted with recruitment efforts; I really appreciate your time and willingness to help.

PAGE 4

i Table of Contents List of Tables iii List of Figures vi Abstract vii Chapter One – Introduction 1 Work-Family Conflict: An Overview 2 Personality and WFC: A Summar y of Previous Research 5 Conscientiousness 5 Extraversion 6 Neuroticism 7 Locus of Control 8 Overview 9 Stress and Coping: An Overview 10 Stress and Lazarus and Folkman’s (1984) Model 10 Overview of Coping Styles 12 Coping Styles Investigated in the Present Study 15 Coping Styles Specific to the Work-Family Domain 17 The Role of Personality in the Stress Process 18 Personality and Coping 18 Personality and Coping Relationships 21 The Relationship between Coping and Strain 27 Differential Effectiveness of Coping Styles 27 Hobfoll’s (1989) Conservati on of Resources Model 28 Spillover Model 30 Coping and Work-Family Conflict Relationships 31 Coping as a Mediator between Pers onality and Work-Family Conflict 35 Hypotheses 38 Chapter Two – Method 41 Participants 41 Measures 43 Big Five Personality Variables: Conscientiousness, Extraversion, and Neuroticism 43 Locus of Control 45 Coping Styles 46 Work-Family Conflict 51

PAGE 5

ii Demographics/Control Variables 51 Procedure 54 Chapter Three – Results 56 Preliminary Analyses 56 Hypothesis Testing 62 Personality and Work-Family Conflict 64 Personality and Coping Style 64 Coping Style and Work-Family Conflict 73 Coping Style as a Mediator be tween Personality and WorkFamily Conflict 80 Supplementary Analysis 83 Chapter Four – Discussion 92 Personality and Work-Family Conflict 93 Personality and Coping Style 95 Coping Style and Work-Family Conflict 98 Coping Style as a Mediator betw een Personality and Work-Family Conflict 100 Supplementary Analysis 100 Theoretical Implications 103 Practical Implications 107 Limitations 109 Future Directions 111 Conclusion 116 References 117 Appendices 132 Appendix A: Hypothesized Relationships 133 Appendix B: Big Five Personality Scale Items 134 Appendix C: Locus of Control Scale Items 136 Appendix D: Coping Scale Items 137 Appendix E: Work-Family Conflict Scale Items 139 Appendix F: Hypothesis Testing with Self-Reported Personality and with the Average of Selfand Signifi cant Other-Reported Personality 140 About the Author End Page

PAGE 6

iii List of Tables Table 1 Descriptive Statistics of Demographic Variables 43 Table 2 Confirmatory Factor Analys is for Coping Items: Goodness of Fit Indices 50 Table 3 Confirmatory Factor Analys is for Coping Items: Chi Square Difference Test 51 Table 4 Descriptive Statistics of Study Variables 57 Table 5 Intercorrelations among Study Variables 58 Table 6 Regression of WIF and FI W on Conscientiousness (Significant Other-Report) 65 Table 7 Regression of WIF and FIW on Extraversion (Significant OtherReport) 65 Table 8 Regression of WIF and FIW on Neuroticism (Significant OtherReport) 66 Table 9 Regression of WIF and FIW on Internal Locus of Control (SelfReport) 66 Table 10 Regression of Coping Style on Conscientiousness (Significant Other-Report) 71 Table 11 Regression of Coping Style on Extraversion (Significant OtherReport) 72 Table 12 Regression of Coping Style on Neuroticism (Significant OtherReport) 72 Table 13 Regression of WIF and FI W on Problem Solving for Work Stressors 74 Table 14 Regression of WIF and FI W on Problem Solving for Family Stressors 74

PAGE 7

iv Table 15 Regression of WIF and FIW on Instrumental Support Seeking for Work Stressors 75 Table 16 Regression of WIF and FI W on Emotional S upport Seeking for Work Stressors 75 Table 17 Regression of WIF and FIW on Instrumental Support Seeking for Family Stressors 76 Table 18 Regression of WIF and FI W on Emotional S upport Seeking for Family Stressors 76 Table 19 Regression of WIF and FIW on Positive Cognitive Restructuring for Work Stressors 77 Table 20 Regression of WIF and FIW on Positive Cognitive Restructuring for Family Stressors 77 Table 21 Regression of WIF and FIW on Rumination for Work Stressors 78 Table 22 Regression of WIF and FIW on Rumination for Family Stressors 78 Table 23 Regression of WIF and FI W on Escape for Work Stressors 79 Table 24 Regression of WIF and FIW on Escape for Family Stressors 79 Table 25 Mediated Regression of WI F on Rumination for Work Stressors and Neuroticism (Significant Other-Report) 83 Table 26 Moderated Regression of FI W on Conscientiousness (Significant Other-Report) and Problem So lving for Work Stressors 86 Table 27 Moderated Regression of FI W on Conscientiousness (Significant Other-Report) and Positive Cognitive Restructuring for Family Stressors 86 Table 28 Moderated Regression of FI W on Conscientiousness (Significant Other-Report) and Escape for Family Stressors 87 Table 29 Moderated Regression of WI F on Extraversion (Significant OtherReport) and Escape for Family Stressors 87 Table 30 Moderated Regression of FIW on Neuroticism (Significant OtherReport) and Rumination for Family Stressors 88

PAGE 8

v Table 31 Moderated Regression of WIF on Internal Locus of Control (SelfReport) and Instrumental Support Seeking for Family Stressors 88 Table 32 Regression of WIF and FIW on Conscientiousness (Self-Report) 140 Table 33 Regression of WIF and FIW on Conscientiousness (Average of Selfand Significant Other-Report) 140 Table 34 Regression of WIF and FI W on Extraversion (Self-Report) 141 Table 35 Regression of WIF and FIW on Extraversion (Average of Selfand Significant Other-Report) 141 Table 36 Regression of WIF and FI W on Neuroticism (Self-Report) 142 Table 37 Regression of WIF and FIW on Neuroticism (Average of Selfand Significant Other-Report) 142 Table 38 Regression of Coping Style on Conscientiousness (Self-Report) 143 Table 39 Regression of Coping Style on Conscientiousness (Average of Selfand Significant Other-Report) 143 Table 40 Regression of Coping Styl e on Extraversion (Self-Report) 144 Table 41 Regression of Coping Style on Extraversion (Average of Selfand Significant Other-Report) 144 Table 42 Regression of Coping Styl e on Neuroticism (Self-Report) 145 Table 43 Regression of Coping Style on Neuroticism (Average of Selfand Significant Other-Report) 145 Table 44 Mediated Regression of WI F on Rumination for Work Stressors and Neuroticism (Self-Report) 146 Table 45 Mediated Regression of WI F on Rumination for Work Stressors and Neuroticism (Average of Self and Significant Other-Report) 146

PAGE 9

vi List of Figures Figure 1. Problem solving for work stressors as a moderator between conscientiousness (significant other-report) and FIW. 89 Figure 2. Positive cognitive restructuring for family stressors as a moderator between conscientiousness (signifi cant other-report) and FIW. 89 Figure 3. Escape for family stressors as a moderator between conscientiousness (significant other-report) and FIW. 90 Figure 4. Escape for family stressors as a moderator between extraversion (significant other-report) and WIF. 90 Figure 5. Rumination for family st ressors as a moderator between neuroticism (significant other-report) and FIW. 91 Figure 6. Instrumental support seeking fo r family stressors as a moderator between internal locus of cont rol (self-report) and WIF. 91 Figure 7. Hypothesized relationships be tween personality variables, coping styles, and work-family conflict. 133

PAGE 10

vii Personality and Work-Family Conflict: Th e Mediational Role of Coping Styles Rebecca H. Bryant ABSTRACT Although an extensive body of literature exists on the consequences of workfamily conflict (WFC), comparatively little research has examined the construct’s antecedents. Research on two sets of anteceden t variables, personality and coping style, is particularly scarce. Thus, the present study expands the literature by examining four personality variables (conscien tiousness, extraversion, neurotic ism, and locus of control) and five coping styles (problem solving, s upport seeking, positive cognitive restructuring, rumination, and escape) in relation to work-i nterference-with-family (WIF) and familyinterference-with-work (FIW) conflict. Addi tionally, coping style, which was assessed separately for managing work stressors and for managing family stressors, was examined as a potential mediator between persona lity and both direct ions of WFC. Two hundred and four participants, r ecruited from a snowball approach, completed surveys. Additionally, significant others provided ratings of conscientiousness, extraversion, and neuroticism. Of the four pe rsonality variables, only neuroticism related to WIF and FIW. Furthermore, among the hypot hesized relationships between coping and WFC, only rumination and escape for work stressors related to WIF, though several cross-domain relationships were observed. Ov erall, the present st udy found little support for coping as a mediator between personality and WFC, though there was some evidence

PAGE 11

viii that rumination mediated the relationsh ip between neuroticism and WIF. As a supplementary analysis, coping was examined as a moderator between personality and WFC. Theoretical and practical implications, as well as future directions, are discussed.

PAGE 12

1 Chapter One Introduction Over the past two decades, an abundance of research on work-family conflict (WFC) has been conducted, cont ributing to an extensive body of knowledge in the field. Researchers have identified numerous conse quences of the construct, including workrelated outcomes, non work-rela ted outcomes, and stress-rel ated outcomes (see Allen, Herst, Bruck, & Sutton, 2000 and Mesme r-Magnus & Viswesvaran, 2005 for metaanalytic reviews), highlighting the potent ially deleterious eff ects of WFC for both individuals and organizations. In terms of antecedents, rese archers have found support for role-related variables as well as persona lity variables (see Byron, 2005 and Frone, 2003 for reviews), though comparatively less rese arch has been conducted on the latter. Another relatively neglected set of variables is coping style, with less than one percent of work-family research examining coping as a predictor of WFC (Eby, Casper, Lockwood, Bordeaux, & Brinley, 2005). Thus, the present st udy focuses on the role that dispositions and coping styles play in the work-family c onflict process. By identifying individual difference variables that relate to WFC, th e present study aims to expand the construct’s nomological network and allow for the identifi cation of individuals who may be at higher risk for experiencing conflict between the wo rk and family domains. Furthermore, the present study aims to incr ease our understandin g of how employees cope with the

PAGE 13

2 demands associated with work and family life, with the intention of providing insight into effective strategies for ma naging work-family conflict. In addition to examining personality traits and coping styles that relate to WFC, it is important to understand the mechanis ms underlying these relationships. Although several researchers have sugge sted coping style as a mediator between dispositions and WFC (e.g., Frone, 2003; Greenhaus, Allen, & Spector, 2006; Wayne, Musisca, & Fleeson, 2004), very little research has exam ined this proposition (see Andreassi, 2007 and Smoot, 2005 for exceptions). Furthermore, th e two dissertations th at did investigate the mediational role of coping used taxonom ies of coping that have been widely criticized in the l iterature (see Skinner, Edge, Altm an, & Sherwood, 2003 for a review). The present study addresses this limitation, providing a more thorough investigation of the role of coping styles in the relationship between personali ty and work-family conflict. Specifically, using a stress framework, personality is expected to relate to coping strategy both directly and indirectly via the apprai sal process. Additionally, consistent with Hobfoll’s (1989) conservation of resources model and a sp illover model of work and family, coping strategy is expected to relate to work-family conflict. By examining two sets of relatively neglected antecedents to WF C, dispositional traits and coping styles, as well as the processes underly ing these relationships, the present study represents an important contributi on to the work-family conflict field. Work-Family Conflict: An Overview Drawing from Kahn, Wolfe, Quinn, Snoe k, and Rosenthal’s (1964) research on role conflict, Greenhaus and Beutell (1985) defined work-family conflict as “a form of interrole conflict in which the role pressures from the work and family domains are

PAGE 14

3 mutually incompatible in some respect. That is participation in the work (family) role is made more difficult by virtue of participat ion in the family (work) role” (p. 77). Greenhaus and Beutell (1985) also identifie d multiple dimensions of WFC, stemming from the source of the conf lict. Specifically, time-base d WFC involves competing time requirements across the two role s; strain-based WFC occurs when pressures in one role impair performance in the second role; a nd behavior-based conflict involves an incompatibility of behaviors necessary for the two roles. Since Greenhaus and Beutell’s (1985) seminal article, researchers have iden tified other ways of conceptualizing the construct as well. For example, Greenhaus et al. (2006) supplemented the three forms of conflict with a fourth type: en ergy-based conflict, which results from a lack of energy for dealing with role demands. Conversely, Carl son and Frone (2003) conceptualized workfamily conflict as either external (represe nting outward behavioral interference) or internal (representing psychological preoccupa tion while in the other role), providing empirical support for the proposed factor stru cture. Despite these alternative ways of defining the WFC construct, the vast majority of research in the field has focused on only two forms: timeand strain-based conflict, often utilizing measur es that combine both forms into one overall dimension (e.g., Netemeyer, Boles, & McMurrian, 1996). In addition to the various forms of work-family conflict, researchers have acknowledged the bidirectionality of the construct; specifical ly, work can interfere with family (WIF), and family can interfere w ith work (FIW; Frone, Russell, & Cooper, 1992). Furthermore, according to the domain sp ecificity hypothesis, situational variables associated with a given domain relate to conflict originating from that domain (Frone, 2003; Frone et al., 1992). For example, the dom ain specificity hypothesi s posits that work

PAGE 15

4 stressors are related to WIF, while family stre ssors are related to FI W. Overall, empirical findings are consistent with these propos itions, though the research is not uniformly supportive (see Ford, Heinen, & Langkamer, 2007 and Mesmer-Magnus & Viswesvaran, 2005 for meta-analytic reviews). A great deal of research in the WFC field has focused on identifying its consequences, and much insight has been gain ed as a result. In a meta-analytic summary of the literature, Allen et al. (2000) reviewed 67 arti cles published between 1977 and 1998, describing results in term s of three broad categories of outcomes: work-related (e.g. job satisfaction), non work-related (e.g., life satisfaction), and stress-related (e.g., general psychological strain). Significant relationships were reported for an array of variables across the three domains, highlight ing the potentially damaging consequences of WFC to individuals, families, and orga nizations alike. For example, WFC was negatively related to such attitudes as j ob satisfaction, organizational commitment, and life satisfaction, and positively related to tur nover intentions. Additionally, relationships between WFC and stress-related outcomes were consistently positive, including assessments of strain (e.g., burnout, work -related stress, family-related stress), somatic/physical symptoms, and depression. Al though Allen et al.’s (2000) meta-analysis excluded studies examining FIW, other meta -analyses have reviewed research on both WIF and FIW, reporting similar findings (e.g., Mesmer-Magnus & Viswesvaran, 2005). Given that work-family conflict has been linked to a wide range of potentially deleterious consequences, it is important to understand its causes. To that end, researchers have investigated, and found s upport for, numerous antecedents of WFC, including those arising in the work domain (e.g., work time, role overload, schedule

PAGE 16

5 flexibility, work support) and those arising in the family domain (e.g., family conflict, family support; see Byron, 2005 and Ford et al., 2007 for meta-analytic reviews). Additionally, a comparatively small amount of research has examined the role of individual differences, such as personalit y traits (e.g., Bruck & Allen, 2003). Because personality traits have been relatively neglec ted in the study of work -family conflict, the present study contributes to the work-family literature by examining their role in the process. Specifically, four personality variables are inve stigated: conscientiousness, extraversion, neuroticism, and locus of c ontrol. In the following section, previous research examining the relationship between these dispositional traits and WFC is summarized. Personality and WFC: A Summary of Previous Research Conscientiousness. Conscientiousness represents one of the five major variables in the five-factor model of personality, al so known as the big five (Goldberg, 1990). The five-factor model has been em pirically substantiated across numerous samples, cultures, and studies, using both natural language ad jectives and theoretically developed personality surveys. Thus, the big five – comprised of conscientiousness, extraversion, neuroticism, agreeableness, and openness to e xperience – represents a successful attempt to conceptualize the personality domain into a small number of broad factors (McCrae & John, 1992). According to Costa and McCrae (1 992), each of the big five personality traits can be sub-divided into six facets, highlighting the hier archical nature of the fivefactor model. Conscientiousness can be defined in terms of its six facet scales, including competence, order, dutifulness, self-discip line, deliberation, and achievement striving

PAGE 17

6 (Costa & McCrae, 1992). Thus, highly consci entious individuals tend to be organized, responsible, and efficient, while individuals scoring low on conscientiousness tend to be lackadaisical, unreliable, and careless in working toward their goals (McCrae & John, 1992). Researchers examining conscientiousness as an antecedent to work-family conflict have generally reported a negative relations hip. For example, FIW and conscientiousness were negatively related among a sample of 164 employees at the University of South Florida (Bruck & Allen, 2003) a nd a sample of 296 employed fathers in the Netherlands (Kinnunen, Vermulst, Gerris, & Mkikangas, 2003). Although neither of these studies found a significant relationship between WIF and conscientiousness, other researchers have. For example, Wayne et al. (2004) found that conscientiousness was negatively related to both WIF and FIW using a na tional, random sample (N=2,130). Although the correlation between FIW and conscientiousness was slightly higher than that of WIF and conscientiousness, they were both significant ( r s=-.20 and -.17, respectively). Given that Wayne et al.’s (2004) study u tilized a large, random sample their results may reflect a more accurate estimate of these relations hips. Additionally, Smoot (2005) reported a negative relationship between a general meas ure of WFC and conscientiousness. Thus, while studies linking WIF and conscientious ness have found inconsistent results, the literature generally supports a negative re lationship between work-family conflict and conscientiousness. Extraversion Another one of the big five pers onality traits, extraversion can be defined by its six facet scales: warmth, greg ariousness, assertiveness, activity, positive emotionality, and excitement seeking (Costa & McCrae, 1992). Thus, high-extraverts tend to be outgoing and energetic, while individuals low on extraversion are more

PAGE 18

7 introverted and reserved (McCrae & John, 1992). Sim ilar to the research on conscientiousness, researchers have reported a negative relationship between extraversion and WFC, with some inconsistencies. Fo r example, using data from 1,986 employed adults participating in the National Survey of Midlife Development, Grzywacz and Marks (2000) found that extraversion was negatively related to both negative spillover from work to family (among men and women) and negative spillover from family to work (among women only). Although the authors used the term “negative spillover,” the items measuring the construct mirrored work-family conflict items. While Grzywacz and Marks (2000) found that family-to-work spillo ver only held for women, other researchers have reported a relationship between FIW a nd extraversion across the entire sample (e.g., Andreassi, 2007; Kinnunen et al., 2003). Add itionally, Smoot (2005) reported a negative relationship between extraversion and a genera l measure of WFC. T hough the majority of research in the field has supported a significant negative re lationship between extraversion and WFC, Bruck and Alle n (2003) reported null findings among their sample of 164 university employees. Still, ove rall the literature s upports the notion that individuals high on extraversion tend to report lower levels of work-family conflict. Neuroticism Individuals high on neuroticism te nd to be self-pitying, tense, and worrying, while individuals low on neuroticis m are described as emotionally stable, relaxed, and even-tempered (McCrae & John, 1992). Another one of the personality variables in the five-factor model, neurotic ism is comprised of the following six facets: anxiety, hostility, self-consciousness, de pression, vulnerability to stress, and impulsiveness (Costa & McCrae, 1992). Compar ed to other dispositional antecedents examined in the work-family conflict litera ture, neuroticism has received the most

PAGE 19

8 research attention and support. Across various samples and studies, neuroticism has been positively linked to negative spillover from work to family and family to work (Grzywacz & Marks, 2000); to WIF and FIW (Andreassi, 2007; Bruck & Allen, 2003; Wayne et al., 2004); to timeand strain-b ased conflict (Andreassi 2007; Bruck & Allen, 2003), and to a bidirectional measure of WF C (Smoot, 2005). Negative affectivity (NA), a variable conceptually similar to neuroticism, has been examined in the work-family literature as well. Individual s high on NA tend to focus on the negative aspects of the world and are more likely to report distre ss and discomfort (Watson & Tellegen, 1985). Several studies have found support for a pos itive relationship between NA and various forms/directions of WFC (e.g., Bruck & A llen, 2003; Carlson, 1999). Thus, the positive relationship between neuroticism and wo rk-family conflict is well-supported. Locus of Control Locus of control (LOC) is a stable individual difference trait representing the extent that an individual at tributes outcomes of ev ents to his/her own behavior and actions (internal LOC) or to external circumstances such as powerful others or chance (external LOC; Rotte r, 1966). Although Ro tter (1966) conceptu alized LOC as a unidimensional construct, with internal a nd external LOC repres enting opposite poles on a continuum, researchers have found support fo r anywhere from one to nine factors comprising Rotter’s (1966) locus of control scal e, with the majority of studies reporting a two-factor solution (Ferguson, 1993) As a result, some researchers have defined LOC as a multi-faceted construct. For example, Levenson (1974) proposed, and found support for, three independent dimensions of locus of control, including internal influences, influence of powerful others, and effects of o ccurrences of chance, fate, or luck. Paulhus and Christie (1981) also propos ed a three-factor solution, though they partitioned the

PAGE 20

9 construct into primary behavioral spheres. Sp ecifically, their Sphere s of Control Scale is comprised of three subscales: personal, interp ersonal, and socio-polit ical control (Paulhus & Van Selst, 1990), highlighting the notion that perceived control may vary across different aspects of the individual’s life space. Research examining locus of control in the work-family domain has generally found that individuals higher on internal LOC report less WFC, while individuals higher on external LOC experience higher levels of WFC. For example, Andreassi and Thompson (2007) reported a negative relationshi p between internal lo cus of control and both WIF and FIW using data from a nationally representative sample of employed adults (N=3,504). Similarly, Noor (2002) found that internal locus of control was significantly related to WFC among a sample of 310 ma rried women. While bo th Andreassi and Thompson (2007) and Noor (2002) used a unidi mensional measure of LOC, representing both internal and external LOC, Andreassi ( 2007) used only the internal LOC scale from Levenson’s (1974) measure. Among a sample of 291 employees in diverse industries, Andreassi (2007) found that in ternal LOC was negatively related to strain-based WFC and to FIW but unrelated to time-based WF C and to WIF. Although the null results are surprising, the divergent findings may reflect th e different measures of LOC. Perhaps the relationship between LOC and WFC is par tially driven by the positive relationship between external LOC and WFC, in addition to a negative relationship between internal LOC and WFC. Thus, including bo th internal and external LOC items may be important when assessing the relationship between LOC and WFC. Overview. Although relatively little research has examined the role of personality variables in relation to work-family conflict, th e literature does offer some insight into the

PAGE 21

10 nature of these relationships. The present st udy focuses on four dispos itional traits that have been empirically linked to WFC. Ov erall, the literature supports a negative relationship between WIF/FIW and conscientious ness, extraversion, and internal locus of control, and a positive relationship be tween WIF/FIW and neuroticism. Although identifying personality variables related to WFC is important, it is also necessary to understand the mechanisms that underlie the re lationships. Thus, in order to understand why these relationships exist, mediating va riables must be iden tified. The present study utilizes a stress and coping framework, arguing that the relationships between personality variables and work-family conf lict can be explained via choi ce of coping strategy. To the extent that individuals are pr edisposed to use effective coping mechanisms in managing work and family stressors, they are less likel y to experience work-family conflict. The following section provides a brief overview of the stress and coping pr ocess in order to provide a framework for these propositions. Stress and Coping: An Overview Stress and Lazarus and Folkman’s (1984) Model Stress has been studied in multiple domains, ranging from a medical approach, a clinical/psychological approach, an engineering psychology approach, and an organizational psychology approach (Jex, 2002). The present study views stress from an organizational psychology approach, which focuses on psychosocial sources of stre ss and how they are cognitively appraised (Jex, 2002). Accordingly, stress is defined as occurring when an i ndividual appraises an event as about to tax or exceed that pers on’s resources, thus endangering his/her wellbeing (Lazarus & Launier, 1978). Additionally, stress is often view ed in terms of a stimulus-response process, in which stress is a type of force acting on a person (Jex,

PAGE 22

11 2002), representing the interaction between the person and the environment (Folkman, Lazarus, Dunkel-Schetter, DeLongis, & Gruen, 1986). Stressors, or environmental factors eliciting a response from the individual, ma y lead to strain, defined as maladaptive responses to stress (Jex, 2002). In the present study, work-family conflict is conceptualized as a strain that occurs when job and/or family stressors are not effectively managed. This perspective is consistent with previous research (e.g., Carlson & Perrew, 1999). Though researchers have proposed numerous models of the stress process, the current study focuses on Lazarus and Folkman’s (1984) Cognitive Theory of Stress and Coping. Specifically, they posit that an indivi dual faced with a stre ssor(s) undergoes two appraisal processes: primary a ppraisal, to determine the imp lications of the stressor (e.g., whether the event is threatening or challeng ing), and secondary appr aisal, an evaluation of his/her ability to manage the even t by assessing demands, constraints, and environmental resources. How the individual ap praises the stressor affects his/her choice of coping strategy (Lazarus & Folkman, 1984). For example, several studies (e.g., Mikulincer & Victor, 1995) have found that when an event is appraised as threatening, individuals tend to respond with emotion-focu sed coping (efforts to manage the emotions arising from the stressor), while an event that is appraised as challe nging tends to elicit problem-focused coping (efforts to modify the source of the distress; Lazarus & Folkman, 1984). Thus, two processes are critic al to understanding the stress process: the appraisal process (the meaning the indivi dual attaches to the encounter) and coping (his/her attempts to manage the enco unter; Lazarus & Folkman, 1982). Research on coping is summarized in the following section.

PAGE 23

12 Overview of Coping Styles Coping is defined as “cogniti ve and behavioral efforts to manage specific external and/or internal demands that are a ppraised as taxing or exceeding the resources of the person” (Folkman & Lazarus, 1991, p. 210). Although an immense amount of research on coping has been conducted, researchers have conceptualized coping in many different ways making it difficult to integrate research (Thompson, Poelmans, Allen, & Andreassi, 2007). For example, modern coping research has been heavily influenced by Lazarus and Folkman’s (1984) distinction between problemand emotion-focused coping, defined in the previous section. However, other popular distinctions have been utilized as well, including appr oach (responses that bring the individual in closer contact with the st ressful encounter) versus avoidance (responses that allow the individual to withdraw; Roth & Cohen, 1986) and primary control coping (change the stressor or emo tions through problem solving or emotion regulation) versus secondary control coping (f acilitate adaptation to stre ss via acceptance or cognitive restructuring; Rothbaum, Weisz, & Snyder, 1982). A summary of the various conceptualizati ons of coping is presented by Skinner et al. (2003), who reviewed over 100 category systems of coping, identifying over 400 different labels of coping styles. Their ar ticle covered 20 years worth of research on coping category systems, reviewing cla ssification systems developed from both inductive/bottom-up and deductive/top-down approaches. The revi ew highlights the limitations of using broad distinctions of copi ng styles. Specifically, they criticized the emotionversus problem-focused taxonomy as lacking conceptual clarity and mutual exclusiveness. For example, emotion-focuse d coping includes such divergent behaviors as relaxation, seeking social support, and avoidance. More over, making a plan has been

PAGE 24

13 categorized into both types of coping. In fact, Lazarus (1996) commented that “distinguishing between the two functions, but tr eating them as if they were distinctive types of coping actions, has led to an overs imple conception of the way coping works and is measured in much research” (p. 292). Furt hermore, factor-analyti c investigations of problemand emotion-focused measures have reported anywhere from two to nine factors, with eight different factor structures being reporte d across eight different studies (Skinner et al., 2003). Other broad distinctions used in the coping literature have been criticized as well. For example, Skinner et al. (2003) co mmented that although support seeking is commonly classified as an approach stra tegy, it could also be categorized under avoidance coping, given that it orients the individual away from the stressor. Additionally, researchers tend to only include constructiv e behaviors under approachoriented coping and destructive behaviors unde r avoidance-oriented coping, but in reality, either may be beneficial depending on th e situation. Thus, the approach-avoidance distinction is overly simplistic. With rega rd to the distinction between primary and secondary control coping, Skinner et al. (2003) pointed to lack of clarity in the definitions and the fact that the distincti on is not exhaustive. Other res earchers have criticized these three coping distinctions as well, offe ring similar viewpoint s (e.g., Compas, ConnorSmith, Saltzman, Thomsen, & Wadsworth, 2001) Additionally, meta-analytic results examining the relationship between coping styl es and personality support the notion that broad distinctions do not tell the whole story, as results differed when analyzed at a specific versus global level of c oping (Connor-Smith & Flachsbart, 2007).

PAGE 25

14 Because of the limitations of such broad classifications of coping styles, Skinner et al. (2003) advocated the id entification of coping styles based on action types, defined as “higher order classes of actions,” di stinguished by a common “motivation underlying the actions” (p. 229). Similarly, Lazarus (1996) agreed that the development of an action typology is essential for identify ing a higher order structure of coping. In order to identify such action tendencies, Skinner et al. ( 2003) reviewed four category systems for classifying lower order ways of coping that have been empirically tested, representing “the state-of-the-art in the study of ca tegory systems” (p. 232). Additionally, they supplemented the review by examining two rationally-based taxonomies developed from action tendencies. Based on these six studies, they identified 13 potential families of coping styles. Five of the coping styles, problem solving, support seeking, escape, distraction, and positive cogniti ve restructuring, were descri bed as clearly “core,” given that they appeared in 25 to 50 percent of all systems review ed and were used with both children/adolescents and adults, as well as domain-specific and general stressors. Other “strong candidates” included rumination, helple ssness, social withdrawal, and emotional regulation. These four coping styles were selected base d on the following criteria: appeared in two of the four empirically-d erived systems and both of the rationallyderived systems; were in at le ast twenty of the systems revi ewed; and appeared in scales used with different ages and types of stressor s. The last four potential families of coping were used less frequently, including ne gotiation, information seeking, opposition, and delegation. Therefore, while the vast majority of re search in the coping l iterature has focused on such broad distinctions as emotionvers us problem-focused coping, the present study

PAGE 26

15 examines coping in terms of action tendenc ies. Although individuals may use a wide variety of coping strategies to deal with work and family stressors, the present paper focuses on five in particular – problem solving, support seeking, positive cognitive restructuring, escape, and rumination – to explain the relationship between personality and WFC. These five coping styles were c hosen from Skinner et al.’s (2003) list of “core” action tendencies, w ith the exception of rumination. Although rumination was identified as a “strong candidate ” rather than a “core” action te ndency, it is included here on the basis of its conceptual link with th e work-family domain in general and the spillover model of work and family in particul ar. The spillover model will be described in a subsequent section. First, it is important to clearly define the five coping strategies investigated in the present study. Coping Styles Investigated in the Present Study Given that the coping styles included in the present study were derived fr om Skinner et al.’s (2003) notion of action tendencies, it is appropriate to define each style in terms of the behaviors each one represents. The following descriptions stem from Skinner et al.’s (2003) comprehensive review of the literatu re. Problem solving includes such be haviors as taking direct action, planning, strategizing, and decision making. Th ese actions are intended to directly address the underlying problem or stressor. In relating problem solving behaviors to the work-family domain, sample actions include car efully planning one’s day to facilitate the accomplishment of role-related tasks in a timely manner, or deciding to work late in the evening one day in order to attend a ch ild’s sporting event the following day. Conversely, support seeking actions incl ude both comfort-seeking (approaching others for consolation; i.e., emotional s upport seeking) and help-seeking (approaching

PAGE 27

16 others for instrumental assistance with the pr oblem; i.e., instrument al support seeking). An individual may seek comfort from his/her spouse when work pressures are overwhelming (emotional support seeking) or ask a co-worker to switch shifts with him/her to prevent a potential work-family conflict (instrumental support seeking). Although the instrumental component of suppor t seeking overlaps with problem solving in the sense that they both attempt to direc tly tackle the problem, the distinction lies in whether the individual attempts to handle th e problem him/herself or he/she approaches another person for assistance. The third coping style inve stigated in the present study is positive cognitive restructuring. Individuals engaging in this co ping style attempt to change their view of the stressful situation, rec onceptualizing it in a pos itive light. Positive cognitive restructuring may involve positive thinking, se lf-encouragement, or a minimization of the negative consequences. Thus, while problem solving and support seeking represent behavioral components of coping, positive cognitive restructuring represents a cognitive approach to managing the stressor. In the work-family domain, an individual facing work/family stressors may engage in self-t alk, convincing him/herself that he/she can effectively manage work/family stressors. A more specific example of positive cognitive restructuring may occur when an employee must take off work to take care of his/her child during winter break ; rather than viewing the event as a negative stressor, he/she may perceive the event as an opportunity to spend more time with his/her child. Escape represents efforts to disengage or stay away from the stressful transaction. Sample escape-related actions include avoiding the situation or engaging in denial. For example, an individual may disengage from family stressors by visiting a friend’s house,

PAGE 28

17 or practice denial rather than admitting that work stressors are interfering with his/her ability to successfully acco mplish family-related tasks. Finally, rumination involves “a passive a nd repetitive focus on the negative and damaging features of a stressful transactio n” (Skinner et al., 2003, p. 242). Similar to positive cognitive restructuring, rumination represents a cognitive approach to dealing with stressors; however, this coping styl e emphasizes the negative aspects of the situation, rather than focusing on the positiv e. Individuals engaging in rumination may have intrusive or negative thoughts, engaging in self-blame and/or worry. In the workfamily domain, rumination may involve exce ssively worrying about family stressors while at work or vice versa. Coping Styles Specific to the Work-Family Domain The present study focuses on action tendencies that individuals may use to respond to a range of stressors. However, it is also important to note that research ers have developed taxonomies of coping mechanisms specific to managing work-fam ily conflict. Specifically, Hall (1972) developed a typology based on interviews with women, emphasizing three coping strategies: structural role redefinition, personal role re definition, and reactive role behavior. Structural role re definition involves altering exte rnal, structurally imposed expectations about one’s role, while persona l role redefinition involves changing one’s own attitudes and perceptions of role demands. Finally, i ndividuals who try to meet everyone’s expectations are engagi ng in reactive role behavior. Another taxonomy for coping with wo rk-family conflict was developed by Somech and Drach-Zahavy (2007). After in terviewing employed parents working in Israel, they identified eight coping strategies : super at home and super at work (based on

PAGE 29

18 the notion that one must achieve perfec tion in task accomplishment); good enough at home and good enough at work (lowering one’s pe rformance to a less-th an-perfect level); delegation at home and delegation at work (e nlisting the help of others to accomplish role-related tasks); and priorities at home a nd priorities at work (prioritizing tasks and responding accordingly). While coping taxonomies specific to the work-family domain are insightful and useful, the present study focuses on general copi ng strategies instead for several reasons. First, there is a great deal of research on the coping styles include d in the present study, while the research on work-family-specific co ping strategies is comparatively limited. Additionally, the action tendenc ies identified by Skinner et al. (2003) and used in the present study were chosen ba sed on their relevance to both general and domain-specific stressors. Thus, it is reasonabl e to assume that they can be appropriately applied to job and family stressors. Finally, researchers ha ve not examined pers onality variables in relation to Hall (1972) or to Somech and Drach-Zahavy’s (2007) taxonomies. Conversely, there is both empirical and theo retical support for the relationship between the personality variables and coping styles examined in the present study. The following section reviews such research, elucidating the role that personality variables play in the stress and coping process. The Role of Personality in the Stress Process Personality and Coping Throughout the history of c oping research, the presumed relationship between personality and copi ng has changed. Early coping researchers considered personality and coping to be the same thing, reflecting Freud and a psychoanalytic perspective (Suls & David, 1996). A second school of thought focused on

PAGE 30

19 the situational specificity inherent in copi ng, viewing choice of copi ng style as a function of the situation, rather than the indi vidual (Lazarus & Folkman, 1984). However, researchers have recently reconsidered the ro le of dispositions in the coping process, and modern researchers have generally taken th e viewpoint that pers onality and coping are distinct but overlapping (Sul s & David, 1996). Empirical re search has supported this notion, reporting relationships between pers onality and coping in the mid-range (e.g., Connor-Smith & Flachsbart, 2007; Vo llrath, Torgersen, & Alns, 1995). The relationship between personality and c oping can be explained via the role of personality on the appraisal process. As pr eviously mentioned, Lazarus and Folkman’s (1984) Cognitive Theory of Stress and C oping highlights the importance of an individual’s cognitive apprai sal of a stressor for unders tanding the stress process. Specifically, an individual’s appraisal of st ressor severity and his/ her perceived coping potential predicts the nature of the coping strategies he/she will use in managing the stressor. To the extent that stressors are pe rceived as positive or challenging rather than threatening (primary appraisal), and to the ex tent that the individual believes he/she has the resources to cope with the threat (sec ondary appraisal), the i ndividual is likely to engage in effective coping strategies. Thus, pe rsonality traits that affect these appraisal processes are likely to affect choice of coping strategy. The notion that personality affects how an individual appraises stressors has been suggested by numerous researchers (e .g., Bolger & Zuckerman, 1995; Lazarus & Folkman, 1984). For example, Bolger and Zu ckerman (1995) suggested that personality may lead to “differential exposure” to stress ors (the number or type of events that a person experiences and that can cause stre ss) and “differential reactivity” (individual

PAGE 31

20 differences in the felt intensity or reaction to stress). In a 14-day daily diary study of 94 students, they found support for the notion that participants high on neuroticism reported greater exposure and reactivity to the stress or of interpersonal conflicts. Furthermore, Kulenovic and Buško (2006) empirically tested the hypothesis that c ognitive appraisals mediate the relationship between stable dispos itional traits and coping style, offering support for the notion that perceived even t stressfulness mediated the relationship between personality and emotion-focused coping. While personality appears to affect c hoice of coping style indirectly via the appraisal process, it may also affect coping behavior dire ctly (Connor-Smith & Flachsbart, 2007; Lazarus & Folkman, 1984). Personality involves “a person’s characteristic patterns of behaviour, thought s, and feelings” (Carver & Scheier, 2000, p. 5). To the extent that the be havior underlying a given copi ng strategy coincides with an individual’s characteris tic patterns of behavior, he/she is likely to engage in that coping mechanism. For example, conscientious indi viduals tend to be planful (McCrae & John, 1992), so it is not surprising that they tend to engage in problem solving behaviors such as planning (Vollrath, 2001). Similarly, high neuroticism is associated with worry (McCrae & John, 1992), so high-ne urotics are likely to experience ruminating thoughts. Thus, the Cognitive Theory of Stress and Coping (Lazarus & Folkman, 1984) elucidates why personality affects choice of copi ng style, both directly and indirectly via the appraisal process. In the following s ection, the hypothesized re lationships between the investigated personality variables and coping styles ar e described, along with the theoretical and empirical support underlying the proposed relationships.

PAGE 32

21 Personality and Coping Relationships The present study examines four personality variables – consci entiousness, extraversion, neuro ticism, and locus of control – and five coping styles – problem so lving, support seeking, positive cognitive restructuring, rumination, and escape. Each pe rsonality trait is expected to exhibit a unique pattern of relationshi ps with the coping styles. High-conscientious individuals tend to be reliable, planful, and dutiful (McCrae & John, 1992). Furthermore, because of their pr eventative efforts, conscientiousness has been associated with reduced stress expos ure (Connor-Smith & Flachsbart, 2007). Given that high-conscientious indivi duals are exposed to fewer stressors and tend to be organized and planful, they are likely to appraise stressors as manageable and controllable. In support of this propositi on, Bouchard, Guillemette, and Landry-Leger (2004) found that conscientious individuals tended to appraise stressors as less threatening and were more like ly to report having the resources to manage them. This pattern of appraisal has been associated with an increased use of active coping (Lu & Chen, 1996). Specifically, because of the behaviors ch aracteristic of conscientiousness, such individuals are likely to e ngage in coping strategies th at involve planning and selfdiscipline, such as problem solving (C onnor-Smith & Flachsbart, 2007). Additionally, conscientiousness has been associated with the ability to focus, even when engaged in unpleasant tasks (Connor-Smith & Flachsbar t, 2007). Thus, high-conscientious individuals are less likely to experience the intrusive thoughts characteristic of rumination and more likely to engage in pos itive cognitive restructuring (Connor-Smith & Flachsbart, 2007). Conscientiousness is also expected to be negatively related to

PAGE 33

22 escape behaviors, given that such individuals have the self-disci pline to face problems head on, rather than avoid the stressor. C onversely, low-conscientious individuals are expected to engage in fewer problem so lving behaviors and less positive cognitive restructuring, resorting to rumina tion and escape coping styles instead. The proposed relationships have received empirical support. In a meta-analysis of 124 studies conducted between 1980 and 2004 (N=33,094), Connor-Smith and Flachsbart (2007) reported a positive relationship between conscientiousness and both problem solving and cognitive restructuring, and a negative relationship between conscientiousness and both negative emotion focus (similar to rumination) and denial (subsumed by the escape coping strategy) In fact, the relationship between conscientiousness and problem solving was am ong the highest correlations reported in the meta-analysis, with a mean correlati on of .30. Studies examining the relationship between conscientiousness and coping styles repeatedly report positive relationships with problem solving strategies that involve active, rational planning and negative relationships with passive maladaptive copi ng styles such as distancing-avoidance, disengagement, and venting emotions (e.g., B ouchard et al., 2004; Jelinek & Morf, 1995; Vollrath, Banholzer, Caviezel, Fischli, & Jungo, 1994). Thus, in the present study, conscientiousness is expected to be positiv ely related to problem solving and positive cognitive restructuring, and ne gatively related to ruminati on and escape. On the other hand, no relationship is expected between conscientiousness and support seeking, consistent with past research (Vollrath et al., 1994) and the nature of conscientiousness. The second personality variable inves tigated in the pres ent study, extraversion, includes facets such as warmth, gregariousne ss, activity, and positive emotionality (Costa

PAGE 34

23 & McCrae, 1992). Because of the positive emo tionality component, researchers have suggested that extraverted indi viduals are less likely to perc eive stressful encounters as threatening (Wayne et al., 2004). Additionally, extraversion has been associated with low stress-reactivity and positive appraisals of coping resources (Bouchard et al., 2004; Connor-Smith & Flachsbart, 2007); thus, they ar e likely to engage in active coping (Lu & Chen, 1996). Specifically, given that high-ex traverts experience positive moods, they may be more likely to engage in positive cogni tive restructuring and less likely to engage in rumination. Furthermore, high-extraverts engage in more activity and have higher energy levels, which they may utilize to enga ge in problem solving behaviors (Vollrath, 2001). On the other hand, as Connor-Smith a nd Flachsbart (2007) point out, increased use of engagement coping does not imply decr eased use of disengagement coping; thus, there is less reason to suspect a link between extraversion and escape behaviors. Finally, extraversion is expected to be positively related to support seeking behaviors, given that high-extraverts tend to engage in more so cial interactions (Eysenck & Eysenck, 1985). Low-extraverted individuals, or introverts, are likely to display the opposite pattern of coping. Connor-Smith and Flachsbart’s (2007) me ta-analysis reported that extraversion was positively related to problem solving, seeking social support, and cognitive restructuring. Although the relationship between extraversion and ne gative emotion focus was not significant in the full sample, it was significant in the “h igh confidence” data subset, characterized by studies with coping styles that could be unambiguously coded. (The authors were not always able to acquire the coping scale used in the original study, leading to concerns of coding errors; they therefore created the “high confidence” data

PAGE 35

24 subset for comparative purposes.) On the other hand, extraversion was not significantly related to avoidance, denial, or withdrawal behaviors analogous to the escape coping style. Thus, similar to the findings of the me ta-analysis, as well as the results reported by individual studies (e.g., Lu & Chen, 1996; Vo llrath et al., 1994; Vollrath et al., 1995; Watson & Hubbard, 1996), extraversion is expect ed to be positively related to problem solving, support seeking, and positive cogniti ve restructuring; negatively related to rumination; and unrelated to escape. High-neurotic individuals tend to be anxious, hostile, and vulnerable to stress (McCrae & John, 1992). It is ther efore no surprise that they have intense emotional and physical responses to stress (Connor-Smith & Fl achsbart, 2007) and are prone to anxiety and fear responses (Eysenck & Eysenc k, 1985). Moreover, participants high on neuroticism reported greater exposure and re activity to the stresso r of interpersonal conflicts (Bolger & Zuckerman, 1995). Because of their nature, high-neurotics are likely to appraise stressors as thr eatening, rather than challengi ng; Bouchard et al. (2004) found support for this proposition, along with the no tion that high-neurotics are less likely to report having the resources to manage stre ss. Such an appraisal pattern makes highneurotics less resistant to stre ss (Hobfoll, 1989) and less likely to engage in active coping (Lu & Chen, 1996). Thus, one would expect a negative relationship between neuroticism and problem solving behaviors. Conversely, b ecause high-neurotics ar e less able to deal with stress, they are more likely to exhi bit an escape coping style. Additionally, individuals high on neuroticism tend to experi ence such negative emotions as worry and anxiety, while emotionally stable individuals (low-neurotics) are described as even tempered and relaxed (Costa & McCrae, 1992; McCrae & John, 1992). As a result of the

PAGE 36

25 negative emotionality component associated with neuroticism, it is hypothesized that neuroticism will be positively related to ru mination and negatively related to positive cognitive restructuring. Neuroticism is one of the most studied personality variables in the coping literature, and research s upports the proposed relationshi ps. In particular, studies consistently report that highneurotics tend to exhibit deni al, escapist behaviors, selfblame, and disengagement (e.g., Bolger, 1990; Vo llrath et al., 1994; Vollrath et al., 1995; Watson & Hubbard, 1996). Additionally, in a longitudinal study conducted over six years, Vollrath et al. (1995) found that neuro tic traits (such as emo tional instability, selfdoubt, and insecurity) were negatively related to active coping and planning as well as positive reinterpretation. Meta-analytic findi ngs are consistent with these results; neuroticism was negatively related to problem solving and cognitive restructuring, and positively related to withdrawal, denial, a nd negative emotion focus (Connor-Smith & Flachsbart, 2007). While neuroticism is expected to be significantly related to problem solving, positive cognit ive restructuring, rumination, and escape in the present study, no relationship is expected between neuroticism and support seeking. The fourth and final personality variable investigated in the present study is locus of control, defined as the ex tent that an individual attrib utes outcomes of events to internal or external causes (Rotter, 1966). Se veral researchers have suggested that locus of control may affect whether a situation is perceived as threat ening in the primary appraisal; specifically, individuals with an internal lo cus of control are expected to appraise stressors as less harmful or threat ening (Folkman, 1984; Rotter, 1975; Schan & Abelson, 1977). Moreover, because individuals with an internal LOC are likely to

PAGE 37

26 perceive stressors as controllable, they are li kely to engage in active coping (Lu & Chen, 1996); conversely, external LOC individuals are likely to per ceive stressors as uncontrollable and therefore threatening, thereby relyi ng on less effective coping strategies. Thus, individuals with an intern al LOC are expected to engage in more problem solving and fewer escape behaviors, given their beliefs that their actions can affect the outcome, while external LOC indi viduals are expected to display the opposite pattern of behavior. Furthermore, because indi viduals with an external LOC are likely to feel helpless when dealing with stressors, they may experience more negative and worrisome thoughts, characteristic of rumina tion. On the other hand, because individuals with an external LOC believe that external circumstances are more influential than internal factors, they may be less apt to enga ge in self-blame, which is also included in the rumination coping style. For this reason, no hypothesis is specified between LOC and rumination. However, LOC is expected to be related to positive cognitive restructuring. Internal LOC individuals are likely to engage in self-encouragement, given their beliefs that their actions are likely to influence the situation, while external LOC individuals are unlikely to have such thoughts, given that th ey believe that environmental conditions will dictate the outcome. There is no reason to expect a relationship between LOC and support seeking. Compared to the big five personality va riables, far fewer studies have examined the relationship between LOC and coping styl e. Those that have reported a positive relationship between internal LOC and direct/active coping strategies and a negative relationship between internal LOC and avoidance/passive coping strategies (Andreassi, 2007; Parkes, 1984; Petrosky & Birkimer, 1991). Because these studies used broad

PAGE 38

27 categorizations of coping beha vior, they provide little in sight into the relationship between LOC and more specific coping styles. Still, they are generally supportive of the notion that internal LOC is likely to relate positively to problem so lving and to positive cognitive restructuring and negatively to escape, as proposed in the present study. The Relationship between Coping and Strain The previous section focused on the re lationship between personality variables and coping styles; this section addresses the relationship between coping and strain. Individuals may engage in a variety of coping behaviors; however, coping styles are likely to be differentially effective in mana ging stress. The first se ction addresses this notion, describing relevant em pirical research. The followi ng two sections focus on why certain coping styles may be more effective th an others, in terms of reducing strains in general and work-family conflict in particul ar, relying on the conservation of resources model and the spillover model. Finally, the last section reviews research on coping and WFC. Differential Effectiveness of Coping Styles Although no coping style is universally effective, resear ch tends to support the utility of problem-focused coping, engagement coping, and primary and secondary control coping in terms of improved physical and mental health outcomes (Com pas et al., 2001). Conversely, disengagement coping and emotion-focused coping tend to have a detrimental effect on one’s health. For example, in a study of 217 Swiss high school students, psychosomatic problems were negatively related to active coping and positively related to denial, disengagement, and venting emotions. Similarly, in a study of 233 French-Canadian university students, use of distancing-avoiding related to increased psychological distress, both concurrently and

PAGE 39

28 prospectively (10 weeks later; Bouchard et al., 2004). The viewpoi nt among researchers and laypersons alike is that active coping, planning, seeking social support, and positive reinterpretation are the most effective, while avoidance, denial, and disengagement are the least successful coping st rategies (Ben-Zur, 1999). Em pirical research tends to support this notion. However, because studies have examined coping styles from a broad perspective, one must infer the relationshi p between more specific coping styles and strain-related outcomes. Hobfoll’s (1989) Conservation of Resources Model Hobfoll’s (1989) conservation of resources model, which enco mpasses several stress theories, offers insight into why some coping styles are more effective than others. According to the model, stress is an environmental reaction to one of three circumst ances: the potential net loss of resources, the actual ne t loss of resources, or the lack of expected resource gain following resource expenditure. Thus, resour ces are the “single unit necessary for understanding stress” (Hobfoll, 1989, p. 515). H obfoll (1989) defines resources as “those objects, personal characteristics, conditions, or en ergies that are valued by the individual or that serve as means for attainment of these objects, personal characteristics, conditions, or energies” (p. 515). Additionally, a central co mponent of Hobfoll’s (1989) model is that individuals seek to maintain, protect, and acquire resources. Consequently, when an individual is under stress, he/she strives to minimize the net loss of resources by employing resources he/she already po ssesses or by calling upon any available environmental resources. To the extent that individuals are able to effectively utilize and/or gain resources, they are less likely to experience stra in following a stressful event.

PAGE 40

29 Conversely, individuals who are ill-equipped to gain resource s are particularly vulnerable to stress (Hobfoll, 1989). Thus, the conservation of resources mode l elucidates why some coping styles may be more or less effective in managing the st ress process and reducing the resultant strain. For instance, when applied to work and/or fa mily stressors, problem solving strategies allow the individual to efficiently fulfill his/her work/family demands, thereby providing him/her with more time and energy (Lapie rre & Allen, 2006). In this example, the individual gains the resources of time and en ergy, which he/she can use to effectively manage role obligations, thus minimizing wo rk-family conflict. Additionally, Hobfoll (1989) describes social relations hips as a resource, provided that they help the individual preserve his/her resources and fulfill situational needs. Accordingly, to the extent that individuals engaging in support seeking behaviors receive th e constructive help and/or comfort that they require, this coping strate gy is likely to be negatively related to WFC. In terms of positive cognitive restructuri ng, Hobfoll (1989) states that “one way individuals may conserve resources is by rein terpreting threat as challenge” (p. 519). By focusing on potential gains rather than lose s, the individual is expected to protect valuable resources, thereby reducing the resu ltant strain. Additionally, such a positive outlook may reduce the individual’s perceptions of WFC, given that the stressors are perceived as manageable (R otondo, Carlson, & Kincaid, 2003). Conversely, ruminating thoughts are lik ely to detract from an individual’s resources; such negative thoughts are counter productive to solving the problem and are likely taxing to the individual. Hence, ru mination may reduce the individual’s energy, taking valuable resources away from work a nd family role obligations. The implications

PAGE 41

30 of an escape coping strategy are less clear. On the one hand, individuals engaging in escape behaviors do nothing to address the st ressor at hand. While disengagement may not require resources, it does not necessarily allow for a resource gain. Because the stressful event itself is taxing to the individual’s resources (Lazarus & Launier, 1978), the result of avoidance may be a net loss of resources. Furthermore, Lapierre and Allen (2006) suggest that “failure to meet ro le demands may potentially threaten the individual’s ability to maintain or gain valued resources, su ch as close relationships at home or a promotion at work” (p. 172). Thus individuals who escap e or avoid work and family stressors, rather than tackling the problem head-on, may experience higher WFC. On the other hand, disengagement could provi de an opportunity for the individual to “recharge his/her batteries,” so to speak, th ereby generating the resources necessary to directly address the problem. Despite these c onflicting hypotheses, th e research generally supports the former proposition, with rese archers reporting a positive relationship between avoidance coping and WFC (e.g., Ro tondo et al., 2003). Such research is summarized in a subsequent section. Spillover Model In addition to the conservation of resources model, the spillover model provides further insight into the impact of various coping styles on strain, and specifically on work-family conflict. Work-f amily spillover involves the transfer of thoughts, affect, and behavior between the work and family domains; thus, experiences in one domain impact experiences in the othe r domain (Roehling, Moen, & Batt, 2003). In support of the spillover model, Williams and Alliger (1994) found that moods, stress, and thoughts generated in one domain influenced behavior and cogni tions in the other domain. The spillover model provides insight into the role of rumination and positive

PAGE 42

31 cognitive restructuring in the strain process. Specifically, the nega tive thoughts and affect associated with rumination are likely to spill over from one domain to the other. This is likely to exacerbate the extent to which work and family roles interfere with each other. Conversely, positive cognitive restructuring invo lves positive thoughts and affect; to the extent that this positive emotionality tran sfers from one domain to the other, the individual is likely to experience less WFC. Coping and Work-Family Conflict Relationships As suggested by the conservation of resources model and the spil lover model, individuals are likely to experience less work-family conflict when they utilize coping strategies that increase their resources and/or allow for the positive, rather than negative, spillover of thoughts and emotions from one domain to the othe r. Although relatively few researchers have examined the relationship between coping and WFC, results are generally supportive of these propositions, though ther e are some inconsistent fi ndings. Specifically, active, problem-focused coping tends to be related to lower levels of WF C, and passive coping tends to be related to higher levels of WFC (see Thompson et al., 2007 for a review). However, most research on coping and WF C has used Hall’s (1972) typology (i.e., structural role redefinition, personal role rede finition, and reactive role definition), rather than the coping styles examined in the pres ent study. For example, Matsui, Ohsawa, and Onglatco (1995) found that work role redefi nition was positively re lated to FIW, while family role redefinition was positively rela ted to WIF among a sample of 131 Japanese full-time married women. Additionally, Ki rchmeyer (1993) surveyed Canadian managers, finding that active coping (t he dominant coping factor, which was a combination of strategies from Hall’s (1972) three coping types) negatively related to

PAGE 43

32 FIW and positively related to positive spillover. (WIF was not examined in Kirchmeyer’s 1993 study.) Furthermore, coping strategies that involved altering one’s own attitude and increasing personal efficiency were rated as mo st effective for coping with multiple roles. A few studies have examined the re lationship between WFC and coping using coping measures not specific to the work -family domain. Lapierre and Allen (2006) reported a negative relationship between problem-focused coping (including such behaviors as prioritizing, aski ng for assistance, and following a plan of action) and strainbased FIW. Problem-focused coping was unrelated to the other types of WFC investigated, including time-based FIW a nd timeand strain-based WIF. Similarly, Andreassi (2007) found no relati onship between active coping and work-family conflict, though passive coping was positively related to strain-based WFC, time-based WFC, WIF, and FIW. Conversely, Aryee, Luk, Leung, and Lo (1999) found that problemfocused coping, but not emotion-focused copi ng, was significantly related to WIF and FIW. Furthermore, both problemand emotion-focused coping were positively related to WFC among a sample of university faculty (Smoot, 2005). The inconsistent findings between coping and WFC are surprising, though the divergent results may be a function of the broad nature of the coping constr ucts. Researchers often vary in their conceptualizations of acti ve/problem-focused and pa ssive/emotion-focused coping, which is likely to affect wh ether/how coping relates to othe r variables, such as workfamily conflict. In one of the few studies to examine work-family conflict in relation to more specific coping styles, Rotondo et al. (2003) examined four t ypes of coping: direct action, help-seeking, positive thinking, and avoidanc e/resignation, along with four types of

PAGE 44

33 WFC: timeand strain based WIF and timeand strain-based FIW. Additionally, participants responded to the coping scale twice: once for managing work stressors and once for managing family stressors. Rotondo et al. (2003) reported a negative relationship between help seeking used for family st ressors and all types of WFC; a negative relationship between help seeking for work st ressors and timeand strain-based WIF; a negative relationship between direct action us ed for family stressors and all types of WFC; and a negative relationship between dir ect action for work stressors and timeand strain-based FIW. In terms of positive thinking, when used for family stressors, individuals reported lower timeand strain-b ased FIW, as well as lower strain-based WIF; conversely, positive thinking for work st ressors was negatively related to strainbased WIF only. Finally, avoida nce/resignation for work stre ssors was positively related to strain-based WIF, while avoidance/resignation for family stressors was positively related to time-based and strain-based FIW. Overall, Rotondo et al.’s (2003) result s are consistent with what one would expect, given the conservation of resources m odel, the spillover model, and research on coping and other forms of strain. Consiste nt with the domain specificity hypothesis (Frone et al., 1992), coping stra tegies used for work stressors tended to relate to WIF, while coping strategies used for family stre ssors tended to relate to FIW, with some cross-domain relationships. Additionally, di rect action, positive thinking, and support seeking were negatively related to WFC, while avoidance/resignation was positively related to WFC. These findings mirror the relationships hypothesi zed in the present study, though Rotondo et al. (2003) did not include ruminatio n. Another notable finding is that, while coping styles e xhibited relationships with both forms of conflict (i.e., time

PAGE 45

34 and strain-based), there were more significan t findings for strain-based WFC. This was true for both avoidance/resignation and positiv e thinking. Conceptually, it makes sense that positive thinking would be more eff ective in reducing strain-based conflict, compared to time-based conflict, given that positive thinking doe s not involve active steps to reduce time conflicts. The null findi ng between avoidance/resignation for work stressors and time-based WIF is more su rprising, though again, the conceptual link between this coping style and strain-based conflict is str onger than that of time-based conflict. Aside from these two exceptions, th e results between coping styles and workfamily conflict were very similar for timea nd strain-based conflict. For this reason, and to keep the number of statis tical tests at a reasonable num ber, the present study does not distinguish between the various forms of work-family conflict. Instead, only WIF and FIW are examined. Finally, a great deal of research has examined the relationship between WFC and social support, and a few studies have examin ed the role of behavioral strategies to reduce WFC. In terms of social support, research consistently supports a negative relationship between support at home and work and both WIF and FIW. In a metaanalysis of 61 studies, Byron (2005) reporte d weighted means ranging from -.11 (family support and WIF) to -.19 (work support and WIF). While receipt of social support and seeking support are distinct, these findings loosely support the noti on that individuals who engage in support seeking coping styles are likely to experience reduced WFC. In terms of behavioral strategies, Baltes a nd Heydens-Gahir (2003) found that use of selection (identifying/setting goals), optimization (using means to achieve a goal), and compensation (anticipating a loss of resources and responding with al ternative means) at

PAGE 46

35 work and home was negatively related to FI W and WIF, respectively. Similarly, use of time management behaviors had a negative re lationship with both WIF and FIW (Adams & Jex, 1999). Selection, optimization, comp ensation, and time management behaviors overlap with problem solving coping strategi es, so the findings in these two studies support the notion that problem solvi ng and WFC are negatively related. Thus, even though relatively few studies have examined coping and WFC, those that have provide important insight. Rot ondo et al.’s (2003) study, in particular, is relevant to the present study. Consistent with their findings, as well as the conservation of resources and spillover models, problem solving, support seeking, and positive cognitive restructuring are expected to decrease WFC, while rumination and escape are expected to increase WFC. Additionally, in accordance with the domain specificity model (Frone et al., 1992), coping strategies used to manage work stressors are expected to relate to WIF, and coping strategies used to manage family stressors are expected to relate to FIW. Asking participants how they manage work stressors separately from how they handle family stressors is consistent with the notion that a situationally specific assessment of coping skills provides a more direct and accu rate portrayal of individual behavior (Lazarus & Folkman, 1984). Coping as a Mediator between Pe rsonality and Work-Family Conflict The previous sections underscore the rationale underlying the expected relationships between personality and WIF/FI W; between personality and coping styles; and between coping styles and WIF/FIW. The present study goes a step further by contending that coping styles mediate the relationship between personality and workfamily conflict.

PAGE 47

36 Although very little resear ch has examined the mechanisms by which personality and WFC are related, two dissertations have addressed this resear ch question (Andreassi, 2007; Smoot, 2005). Similar to the present st udy, both researchers proposed coping styles as the mediating variable. Using a sample of university faculty, Smoot (2005) examined emotionand problem-focused coping as mediator s between the big five personality traits and WFC. Although personalit y, coping, and work-family conflict were related, no support was found for mediation. Similarly, A ndreassi (2007) proposed that active and passive coping would mediate the relationshi p between extraversion, neuroticism, and locus of control and various forms of work-f amily conflict (timeand strain-based WFC, WIF, and FIW). The positive relationship between neuroticism and both strain-based WFC and FIW was mediated by passive copi ng, but none of the other mediating tests were significant. Although such findings are discourag ing, both studies had two noteworthy limitations. First, in both dissertations, coping was examined in terms of broad categorizations, either problemversus emo tion-focused coping (Smoot, 2005) or active versus passive coping (Andreassi, 2007). As previously discussed, such broad distinctions of coping are problematic (see Sk inner et al., 2003 for a review). Thus, the present study contributes to the literatu re by examining specific coping styles, as advocated by Skinner et al. (2003). Additio nally, in both dissertations, general coping behaviors were assessed, rather than asking pa rticipants how they cope with job-related stressors independently from how they c ope with family-related stressors. This methodology ignores the possibility that coping styles vary by situation, as proposed by numerous researchers (e.g., Folkman et al., 1986; Lazarus & Folkman, 1984).

PAGE 48

37 Additionally, it precludes an examination of domain-specif ic relationships (i.e., coping with work stressors relates to WIF, and coping with family stressors relates to FIW). The present study addresses these limitations, th ereby contributing to the work-family field. It is also worth noting that a few st udies have examined coping as a mediator between personality and other types of st rain, offering some support. For example, emotion-focused coping partially mediated the relationship between extraversion and physical symptoms, and avoidance coping partially mediated the link between neuroticism and physical symptoms (Hudek-Kneevi Kardum, & Maglica, 2005). Additionally, in a daily diary study of prem edical students preparing for the medical school entrance examination, self-blame a nd wishful thinking mediated the positive relationship between neurotic ism and distress (Bolger, 1990) Similarly, using a sample of adult male prisoners, Ireland, Brown, and Ba llarini (2006) offered tentative support for the mediating role of maladaptive coping in the relationship between asocial and anxious/dramatic personality and psychologica l distress. Finally, though coping has been studied as both a mediator and a moderator in the relationship be tween personality and strain, Vollrath et al. (1994) compared both models, finding stronger support for a mediation model. Thus, the present study examines five coping styles (problem solving, support seeking, positive cognitive restructuring, rumi nation, and escape) as mediators between four personality variables (conscientiousness, extraversion, neuroticism, and locus of control) and two dimensions of work-family conflict (WIF and FIW). By examining the mechanisms by which personality relates to work-family conflict, including specific

PAGE 49

38 measures of coping styles, and examini ng the process from a domain-specific perspective, the present study offers a unique contribution to the work-family literature. Hypotheses Hypothesized relationships are listed be low. Additionally, Appendix A presents a graphical depiction of hypotheses, with green arrows representing proposed positive relationships, and red arrows deno ting proposed negative relationships. 1. Conscientiousness (1a), extrav ersion (1b), and internal lo cus of control (1c) will negatively relate to WIF and to FIW, and neuroticism (1d) will positively relate to WIF and to FIW. 2. Conscientiousness will positively relate to problem solving (2a) and to positive cognitive restructuring (2b) and negatively relate to rumination (2c) and to escape (2d). 3. Extraversion will positively relate to probl em solving (3a), to support seeking (3b), and to positive cognitive restructuring (3c) and negatively relate to rumination (3d). 4. Neuroticism will negatively rela te to problem solving (4a) and to positive cognitive restructuring (4b) and positively relate to rumination (4c) and to escape (4d). 5. Internal locus of control will positively rela te to problem solving (5a) and to positive cognitive restructuring (5b) and ne gatively relate to escape (5c). 6. Problem solving for work stressors (6a), s upport seeking for work stressors (6b), and positive cognitive restructuring for work stressors (6c) will negatively relate to WIF; and rumination for work stressors (6d) a nd escape for work stressors (6e) will positively relate to WIF.

PAGE 50

39 7. Problem solving for family stressors (7a), support seeking for family stressors (7b), and positive cognitive restructuring for family stressors (7c) will negatively relate to FIW; and rumination for family stressors (7 d) and escape for family stressors (7e) will positively relate to FIW. 8. The relationship between conscientiousness and WIF will be mediated by coping styles used for managing work stressors, including: problem solving (8a), positive cognitive restructuring (8b), ru mination (8c), and escape (8d). 9. The relationship between conscientiousne ss and FIW will be mediated by coping styles used for managing family stressors, including: problem solving (9a), positive cognitive restructuring (9b), ru mination (9c), and escape (9d). 10. The relationship between extraversion and WIF will be mediated by coping styles used for managing work stressors, including : problem solving (10a), support seeking (10b), positive cognitive restructur ing (10c), and rumination (10d). 11. The relationship between extraversion and FIW will be mediated by coping styles used for managing family stressors, includi ng: problem solving ( 11a), support seeking (11b), positive cognitive restructur ing (11c), and rumination (11d). 12. The relationship between neuroticism and WIF will be mediated by coping styles used for managing work stressors, incl uding: problem solving (12a), positive cognitive restructuring (12b), ru mination (12c), and escape (12d). 13. The relationship between neuroticism and FIW will be mediated by coping styles used for managing family stressors, including: problem solving (13a), positive cognitive restructuring (13b), ru mination (13c), and escape (13d).

PAGE 51

40 14. The relationship between locus of control and WIF will be mediated by coping styles used for managing work stressors, incl uding: problem solving (14a), positive cognitive restructuring (14b), and escape (14c). 15. The relationship between locus of control and FIW will be mediated by coping styles used for managing family stressors, including: problem solving (15a), positive cognitive restructuring (15b), and escape (15c).

PAGE 52

41 Chapter Two Method Participants In work-family conflict research, partic ipants are typically only included if they are working at least 20 hours a week and are either married, living w ith a partner, or a parent with a child living at home. In the present study, the inclusion criteria were modified slightly. Because personality ra tings were provided by spouses/significant others, participants had to be either marri ed or living with a significant other, whom he/she had been dating for at least one year, in order to participate. Participants also had to be employed for pay for at least 20 hours a week. Participants were recruited via a sn owball approach in which potential participants were contacted via email and aske d to participate in the study and/or forward the survey link to interested individuals. Sp ecifically, recruitment emails were sent to friends, family members, and acquaintances of the principal inve stigator, as well as members of various organizations, including Boy Scouts, Girl Scouts, PTA, and the Society for Industrial-Organizational Psychol ogy (SIOP). In total, 1,999 emails were sent, the vast majority of which were to members of SIOP (N = 1,837). Fifty-two emails were returned as undeliverable, and 27 indivi duals responded that th ey did not meet the study’s inclusion criteria. Three hundred and fort y-nine participants filled out part or all of the survey, resulting in an estimated response rate of 19.0 percent. However, a

PAGE 53

42 determination of the “true” res ponse rate is difficult, given the uncertainty in the number of emails that were never received, the numbe r of participants who were ineligible to participate, and the number of emails that were forwarded to potential participants. Of the 349 individuals who filled out the survey, 32 had a substantial amount of missing data (more than 30 percent), and 19 di d not meet the inclusion criteria. These 51 individuals were dropped from analysis, resu lting in 298 participants. Significant otherratings of personality were provided for 204 of the 298 participants (68.5%). Thus, the sample size used to test study hypotheses was 204. Of the 204 primary participants, 62.3 per cent were female, with a race/ethnicity breakdown of 90.2 percent White/Caucasian, 4.9 percent Asian or Pacific Islander, 2.5 percent Hispanic or Mexican American, and 2.0 percent Black or Af rican American. One participant indicated a race of “other.” In terms of e ducation, 1.0 percent had a high school degree, 2.5 percent had attended some college, 13.7 percent had a college degree, 14.7 percent had a master’s degree, and 67.6 perc ent had a doctoral degree. The majority of the sample was married (91.2%), did not have primary care giving responsibilities for an elderly relative (93.6%), and report ed a household income of $100,000 or higher (73.6%). Approximately 60 percent of the sample had at least one child, with 46.6 percent having at least one ch ild living at home at the time the survey was completed. Of the 204 spouse/significant other partic ipants, 37.7 percent were female, with a race/ethnicity breakdown of 89.2 percent White /Caucasian, 4.4 percent Asian or Pacific Islander, 2.0 percent Hispanic or Mexican American, 1.5 percent Black or African American, and .5 percent American Indian or Alaskan Native. Four participants indicated a race of “other.” While 85.8 percent of the significant other sample was employed for

PAGE 54

43 pay, 14.2 percent were unemployed or fulltime homemakers. Additional demographic information for both the primary participants and the significant othe rs is presented in Table 1. Table 1. Descriptive Statistics of Demographic Variables Variable N M SD Obs. Min. Obs. Max. Primary Participants Age 203 40.93 10.51 22 68 Tenure (years) 204 6.90 7.73 .08 40 Work Time (hrs/wk) 204 46.96 10.17 20 80 # Children at Home 201 .85 1.11 0 6 RFD Index1 204 4.52 5.76 0 24 Relationship Length (years) 204 14.95 10.41 1.5 46.5 Significant Other Participants Age 200 41.09 10.32 24 68 1Responsibility for Dependents Index (Rothausen, 1999) Measures Scores were calculated by averaging item responses. Scales are provided in Appendices B, C, D, and E, and alpha coeffici ents for each scale are provided in Table 4. Big Five Personality Variables: Conscientiousness, Extraversion, and Neuroticism The Big Five Inventory (BFI) was used to measure conscientiousness, extraversion, and neuroticism. Developed by John, Donahue, and Kentle (1991), the BFI is a brief inventory that relie s on short phrases, rather than single adjective items. Thus, BFI items “retain the advantages of adject ival items (brevity and simplicity) while

PAGE 55

44 avoiding some of their pitfalls (ambiguous or multiple meanings and salient desirability)” (John & Srivastava, 1999, p. 115). The BFI scales demonstrate convergent and discriminant validity, high alpha reliabilities, and three-month test -retest reliabilities ranging from .80 to .90 (John & Srivastava, 1999). Although another set of persona lity scales, the NEO questio nnaires, “represent the best-validated Big Five measures in the questionnaire tradition” (John & Srivastava, 1999, p. 115), the BFI is optimal when partic ipant time is limited. Thus, in order to minimize survey length, the BFI was selected for the present study. It is also worth noting another readily available personality invent ory – the 50-item measure of the big five personality factors available from the Inte rnational Personality Item Pool (IPIP; Goldberg, Johnson, Eber, Hogan, Ashton, Cloninger, & Gough, 2006). Although this scale has been used frequently in the fiel d (Goldberg et al., 2006) the items do not tap several facets of interest. For example, th e extraversion items assess an individual’s sociability, but items measur ing one’s positive emotionali ty and activity levels are lacking. Given that these aspects of extrav ersion are important in the present study, the BFI was chosen over the IPIP scale. The BFI conscientiousness scale includes nine items, while the extraversion and neuroticism scales are each comprised of eight items. Sample items are as follows: “Does things efficiently” (conscientiousness), “Is full of energy” (extraversion), and “Worries a lot” (neuroticism). Responses to the personality items were on a five-point Likert-type scale that ranged from 1 (strongly disagree) to 5 (s trongly agree). To minimize common method variance, both selfand significant other-ratings of conscientiousness, extraversion, and neurotic ism were collected. Correlations between

PAGE 56

45 selfand partner-report s of personality were moderate to high in magnitude, ranging from .46 for conscientiousness to .68 for extrav ersion. The correlation between selfand partner-reported neuroticis m was .62. While there was no significant mean difference between selfand significant othe r-ratings of conscientiousness ( t (203) = .32) or extraversion ( t (203) = -1.20), significant other-ratings of neuroticism tended to be higher than self-ratings ( t (203) = -3.14, p < .01; Ms = 2.86 and 2.71). Previous researchers have reported simila r correlations between selfand partnerratings of personality. For example, McCrae, Stone, Fagan, and Costa (1998) compared selfand spouse-ratings of personality, re porting uncorrected correlations of .49 for conscientiousness, .74 for extraversion, and .46 for neuroticism. A nother study reported correlations of .70, .57, and .63 between selfand spouse-ratings for conscientiousness, extraversion, and neuroticism, respectively (Mut n, 1991). In the present study, hypotheses were tested with partner-ratings of conscientiousne ss, extraversion, and neuroticism. Additionally, self-ra tings of personality, as well as the average of selfand significant other-ratings, were examined fo r comparative purposes. Alpha coefficients ranged from .84 to .87 for significant other-re port and from .79 to .91 for self-ratings. Locus of Control To assess locus of control, Paulhus and Van Selst’s (1990) Personal Control Scale was used. The Personal C ontrol Scale is a subscale of the Spheres of Control Scale, which is designed to a ssess components of perceived control. The original scale was published by Paulhus and Ch ristie (1981) but has since been revised. The revised scale, published in Paulhus and Van Selst (1990), addr essed limitations of the original scale, such as low internal consistency of the Personal Control subscale. Paulhus and Van Selst (1990) report an alpha re liability of .80 for the revised subscale, as

PAGE 57

46 compared to .59 for the original subscale. Furthermore, the Personal Control Scale is moderately correlated with other measur es of LOC (e.g., Levenson, 1981; Rotter, 1966, as reported in Paulhus & Van Selst, 1990), and the test-retest reliabi lity at four weeks is .90 (Paulhus, 1983). Moreover, the scale has been used numerous times in the literature to assess locus of control (e.g., Burns, Dittmann, Nguyen, Mitchelson, 2000; Parkes & Razavi, 2004). Although Rotter’s (1966) I-E Scale is used more frequently in the literature, the scale is significantly longer, w ith 29 items. Thus, for the pur poses of brevity, the Paulhus and Van Selst (1990) scale was used in the present study. A sample item is “My major accomplishments are entirely due to my hard work and ability.” Participants responded on a five-point Likert-type scale that ranged from 1 (totally inaccu rate) to 5 (totally accurate). The alpha coefficient was .70. Unlik e the other three personality variables investigated in the present st udy, significant others did not provide ratings of locus of control. Locus of control represents an indi vidual’s internal beliefs, and it is unclear whether significant others can accurately gauge the extent that their husbands/wives would endorse an internal versus external LOC. Coping Styles The coping styles examined in the present study were selected based on Skinner et al.’s (2003) review of the literature, rather than an existing taxonomy. As such, there is no existing copi ng scale that includes the five specific dimensions of coping examined here, though several scales measure similar and/or overlapping constructs. For example, the c oping scale used in Ro tondo et al.’s (2003) study had four dimensions: direct acti on, positive thinking, help seeking, and avoidance/resignation (based on Havlovic & Keenan’s, 1991 scale). Similarly, the Ways

PAGE 58

47 of Coping scale (Folkman & Lazarus, 1980; Folkman et al., 1986) includes scales for planful problem solving, confrontive coping, s eeking social support, positive reappraisal, and escape-avoidance. Another popular coping measure, the COPE (Carver, Scheier, & Weintraub, 1989) is comprised of 15 scales including active coping, planning, positive reinterpretation and growth, use of instrumental social suppo rt, use of emotional social support, focus on and venting of emotions behavioral disenga gement, and mental disengagement. Thus, numerous coping scales have con ceptualized coping in slightly different ways. Although existing scales include taxonom ies similar to the one examined in the present study, none provide an exact match. Th erefore, the present study used items from existing coping scales (Carver et al., 1989; Connor-Smith, Compas, & Wadsworth, 2000; Folkman & Lazarus, 1980; Folkman et al ., 1986; Havlovic & Keenan, 1991; Saffrey & Ehrenberg, 2007) to develop a new measure. To guide the final selection of items, a pilot study was conducted with eighty -six undergraduate students who were employed for pay and were either married, living wi th a significant other, or a parent with a child living at home. The initial coping scales included betw een eight and 10 items per coping style, and participants were asked to respond to each item twice: once in terms of coping with work stressors and once in terms of coping with family/home stressors. Participants also responded to personality and work-family conf lict items. A subset of the initial coping items were selected for the present study on th e basis of a factor analysis, reliability analysis, and converg ent/discriminant validity evidence. The selected coping items are provided in Appendix D. Sample items are as follows: “I make a plan of action” for problem solving; “I talk to someone about how I

PAGE 59

48 feel” for support seeking; “I tell myself that everything will be alright” for positive cognitive restructuring; “I think about every single deta il of the event over and over again” for rumination; and “I try to forget the whole thing” for escape. Responses were on a five-point Likert-type scale that range d from 1 (very rarely) to 5 (very often). Participants were asked to respond to each c oping item twice: once for coping with work stressors and once for coping with family stressors. The coping scales used in the present study initially included five items per coping style, with the exception of escape, wh ich included four items. However, two of the items, including the first escape item (“ I hope a miracle will happen”) and the fourth problem solving item (“I try to work harder and more efficiently”) exhibited poor itemtotal statistics. Specifically, the initial alpha coefficients increased by deleting the two items (escape: .51 versus .64 for coping with work and .64 versus .77 for coping at family; problem solving: .76 versus .80 fo r coping with work and .79 versus .82 for coping with family). Additionally, the corrected item-total correlations were in the low to moderate range (escape: .09 for coping w ith work and .17 for coping with family; problem solving: .33 for coping with work and .44 for coping with family). Thus, these two items were dropped from analysis, resul ting in a four-item scale for problem solving and a three-item scale for escape. After dele ting these two items, the alpha coefficients ranged from .63 to .86 for coping with work stressors and .72 to .90 for coping with family stressors. Because the coping scales were not ps ychometrically established and wellvalidated measures, confirmatory factor an alysis (CFA) was conducted to assess model fit, using the PROC CALIS program within SAS 9.1. Because of the relatively small

PAGE 60

49 sample size given the large number of items a nd parameters to be estimated, coping with work items were analyzed separately from c oping with family items. A five-factor model was examined, with problem solving, support seeking, positive cognitive restructuring, rumination, and escape being specified as f actors. Additionally, because support seeking is frequently divided into tw o factors (i.e., instrumental and emotional), a six-factor model was tested as well. The first thr ee support seeking items were considered instrumental support seeking (e.g., I talk to someone who could do something concrete about the problem), while the last two items were considered emotional support seeking (e.g., I talk to someone about how I feel). Fi nally, a one-factor model was also examined for comparative purposes. Across all test ed models, the convergence criterion was satisfied. Goodness of fit indices are provided in Tabl e 2, and chi-square difference tests are presented in Table 3. Small, non significant chi-square values indicate th at the model fits the data. However, chi-square values are in fluenced by several factors other than model fit, including sample size and model complex ity. Thus, a model may fit the data well yet have a significant chi-sq uare value, and it is important to consider other factors as well when assessing model fit. One criteria is the chi-square/df ratio; values less than 2 are often considered acceptable. Other goodness of fit indices include the comparative fit index (CFI), the non-normed fit index (NNFI), the adju sted goodness-of-fit index (AGFI), and the root mean square error of approximation (RMSEA). CFI, NNFI, and AGFI values of .90 or greater are considered acc eptable fit, while valu es of .95 or greater indicate good fit. Moreover, RMSEA values around .05 or less indicate a good fit for the model, while values between .05 and .08 are generally considered acceptable (Browne &

PAGE 61

50 Cudeck, 1993; Hu & Bentler, 1998, 1999; Kline, 1998). As Tables 2 and 3 indicate, the six-factor model fit the data best for both coping with work and coping with family. The improvement in fit between the fiveand si x-factor model was particularly strong for coping with work, indicating the importance of differentiating between instrumental and emotional support seeking for copi ng with work-related stressors. T -values and standardized estimates of each path were also examined. For both coping with work and coping with family, all t -values were significant ( p < .01; range: 5.40 to 16.36 for coping with work; 7.32 to 17.21 for coping with family). The standardized loadings were moderate to high as well. For coping with work, only two values were less than .50 (cognitive restructuring #3 and #4 = .37 and .47, respectively); and only one value was less than .50 for coping with family (cognitive restructuring #2 = .47). The average standard loading was .67 fo r coping with work and .72 for coping with family. Overall, the confirmatory factor an alysis results provide support for the validity of the coping scale used in the present study. Table 2. Confirmatory Factor Analysis for Coping Items: Goodness of Fit Indices Model 2 df 2/df p < RMSEACFI NNFI AGFI Coping with Work Stressors 1 Factor 1030.31 209 4.93 0.01 0.14 0.41 0.35 0.50 5 Factor 544.83 199 2.74 0.01 0.08 0.83 0.81 0.80 6 Factor 382.76 194 1.97 0.01 0.06 0.91 0.89 0.85 Coping with Family Stressors 1 Factor 1885.41 209 9.02 0.01 0.17 0.36 0.29 0.43 5 Factor 435.28 199 2.19 0.01 0.07 0.91 0.90 0.84 6 Factor 361.13 194 1.86 0.01 0.06 0.94 0.92 0.86

PAGE 62

51 Table 3. Confirmatory Factor Analysis for Coping Items: Chi Square Difference Test Model Comparison 2 df p < Coping with Work Stressors 1 vs. 5 Factor 485.48 10 0.01 1 vs. 6 Factor 647.55 15 0.01 5 vs. 6 Factor 162.07 5 0.01 Coping with Family Stressors 1 vs. 5 Factor 1450.14 10 0.01 1 vs. 6 Factor 1524.28 15 0.01 5 vs. 6 Factor 74.14 5 0.01 Work-Family Conflict Netemeyer et al.’s (1996) WIF and FIW scales were used, each of which is composed of five items. A sample WIF item includes “The demands of my work interfere with my home and family life,” while a sample FIW item is “Familyrelated strain interferes with my ability to perform job-related duties.” The items in this scale represent a mixture of tim eand strain-based conflict, consistent with the majority of research on work-family conflict. Netemeye r et al. (1996) offer support for the validity of the two scales, including the scales’ dime nsionality, internal consistency, measurement invariance, and convergent/discriminant validi ty. In the present study, responses were on a five-point Likert-type scale that ranged from 1 (very rare ly) to 5 (very often). Alpha coefficients were .90 for WIF and .89 for FIW. Demographics/Control Variables Potential control variab les that were collected include gender, age, significant other’s empl oyment status, number of children living at home, work hours, household income, and wh ether the participan t had primary care giving responsibilities for an elderly relative. Gender and age were included as potential

PAGE 63

52 control variables for hypotheses in which c oping is the dependent variable, given that some studies have found differences in choice of coping style across gender and age. For example, Endler and Parker (1990) found th at women tended to use emotion-focused coping more often than men, and Carver et al. (1989) found that women reported greater use of focusing on/venting emotions and s eeking social support, while men reported increased likelihood of using al cohol and drugs as a coping mechanism. Age differences in coping style have been reported as well (e.g., Blanchard-Fields, Stein, & Watson, 2004; Heiman, 2004). Specifically, Blanchard-Fields et al. ( 2004) found that middle-aged adults used more proactive emotion-regula tion strategies and fewer passive emotionregulation strategies than olde r adults, while Heiman (2004) found that older participants reported more task-oriented coping and less emotional coping as compared to younger participants. Other studies have reported ag e differences as well (e.g., Haren & Mitchell, 2003), though findings are inconsiste nt (e.g., Kato & Pedersen, 2005). In terms of work-family conflict, control variables typically include factors likely to affect the individual’s wo rk and/or family demands. Speci fically, individuals who have more children living at home, have primary caregiving responsibilities for an elder relative, and/or work more hours each week are likely to experience higher levels of WFC than those who have fewer children/car egiving responsibilities and/or work fewer hours. In addition to number of children living at home, Responsibility for Dependents (RFD), an index that statistically combines the number and age of dependents according to level of responsibility (Rothausen, 1999), wa s calculated as well. Specifically, the RFD index is a weighted sum of the number of children living at home, whereby younger children are weighted more heavily, and older children ar e given a smaller weight.

PAGE 64

53 Although the majority of work-family research focuses solely on the number of children living at home and/or the age of the youngest child living at home, the Responsibility for Dependents index provides a more thorough a ssessment by considering the age of each child. Household income and significant other’s em ployment status are also likely to affect work/family demands indirectly by pr oviding the individual w ith the resources to manage WFC. For example, those with a high household income may hire a cleaning service, thereby reducing demands in the fa mily realm. Similarly, having a full-time homemaker as a significant other may reduce family demands as well. Gender is also frequently included as a c ontrol variable, given that familyand household-related responsibilities often disproportionately fall on the woman’s shoulders, creating a “second shift” for employed women (Hochs child & Machung, 2003). Although research linking demographic variables to WFC has been inconsistent, with researchers frequently reporting null findings (e.g., Byron, 2005), other studies have found significant relationships (e.g., Gutek, Searle, & Kl epa, 1991; Kinnunen & Mauno, 1998), and the control variables listed above are freque ntly included in work-family studies. Thus, gender, age, number of childre n living at home, work hours, household income, and significant other’s employment stat us were included in the present study as potential control variables. However, each vari able was only included if it related to the dependent variable in question. Additional demographic information was collected as well. Participants were asked to identify themselves as American I ndian or Alaskan Nativ e; Asian or Pacific Islander; Black or African American; White/Caucasian, Non Hispanic; Hispanic or

PAGE 65

54 Mexican American; or other. Education, organi zational tenure, and leng th of relationship with significant other were also included. Significant others were asked basic demographic questions, including ag e, race/ethnicity, and gender. Procedure Participants were contacted via email a nd asked to participate voluntarily in the study. Additionally, they were asked to forward the email to other interested individuals, if willing. Emails included a link to a website with survey instructions and informed consent, assuring participants that their re sponses would be confidential and anonymous. Those who chose to continue were directed to the webpage with survey items. Each participant was asked to provi de his/her significant other’s email address so that the primary investigator could send him/her a similar survey (for another source of personality ratings), as well as a unique si x to eight digit code (generated by the participant) to link the se lf and significant other res ponses. Upon completion of the survey, participants were instructed to submit their responses electronically, and a debriefing page was displayed. Significant ot hers were contacted via email as well, instructed that participation is voluntary, and provided with a link to their version of the survey. Each significant other was provided with the six to eight digit code generated by his/her significant othe r and instructed to enter the code in the beginning of the survey. Significant others were sent reminder emails approximately two weeks after the initial email if they had not yet completed the survey. Participants were also provided with the option of receiving a paper version of the survey via postal mail, and two participan ts elected this option. These participants followed a similar procedure, though they filled out hard copies of study materials, were

PAGE 66

55 asked to manually give their significant ot her the survey, and were provided with an envelope/prepaid postage for mailing studying materials back. In order to link the two sources of data (self and signi ficant other), hard copies of surveys were marked with a code number. The identification numbers were unrelated to particip ant data but rather generated randomly to protect the anonymity of the participants.

PAGE 67

56 Chapter Three Results Preliminary Analyses Descriptive statistics for each study va riable, including sample size, mean, standard deviation, observed minimum/maximum, and alpha coefficient, are provided in Table 4. Intercorrelations between study variables ar e provided in Table 5. Consistent with past res earch, participants reporte d higher levels of WIF ( M = 2.90) than FIW ( M = 2.01; t (202) = 15.61, p < .01). Additionally, participants tended to use problem solving ( t (203) = 2.12, p < .05; M s = 4.04 and 3.95), instrumental support seeking ( t (203) = 2.60, p < .05; M s = 3.45 and 3.29), and positive cognitive restructuring ( t (202) = 4.77, p < .01; M s = 3.37 and 3.18) more with work than with family stressors, while they reported similar use of emotional su pport seeking ( t (203) = -.64), rumination ( t (203) = -.79), and escape ( t (203) = -1.55) across domains. Before conducting the primary analyses, the data were inspected to determine whether any regression assumptions had b een violated. The first assumption of regression, independence, is a methodologi cal question. The study design provides no reason to suppose that the responses of the primary participants depended upon each other; thus, this assumption was assumed to be met. Scatterplots of variable pairs were inspected to test the assumptions of linear ity and homoscedasticity. The graphs did not indicate that non-linear relationships were present. Additionally, the variance of the

PAGE 68

57 Table 4. Descriptive Statis tics of Study Variables Variable N # of Items M SD Obs. Min. Obs. Max. Personality: Self-Report Conscientiousness 204 9 .79 4.07 .50 2.56 5.00 Extraversion 204 8 .91 3.35 .82 1.38 5.00 Neuroticism 204 8 .87 2.71 .77 1.00 4.75 Internal Locus of Control 204 10 .70 4.13 .43 2.40 5.00 Personality: Significant Other-Report Conscientiousness 204 9 .84 4.06 .60 2.11 5.00 Extraversion 204 8 .87 3.40 .78 1.13 5.00 Neuroticism 204 8 .87 2.86 .77 1.13 5.00 Coping Style with Work Stressors Problem Solving 204 4 .80 4.04 .59 2.75 5.00 Support Seeking (Instrumental)204 3 .74 3.45 .78 1.00 5.00 Support Seeking (Emotional) 204 2 .78 3.43 .99 1.00 5.00 Cognitive Restructuring 203 5 .63 3.37 .60 1.40 5.00 Rumination 204 5 .86 2.80 .80 1.00 4.80 Escape 204 3 .64 1.86 .67 1.00 4.00 Coping Style with Family Stressors Problem Solving 204 4 .82 3.95 .65 2.25 5.00 Support Seeking (Instrumental)204 3 .77 3.29 .88 1.00 5.00 Support Seeking (Emotional) 204 2 .84 3.47 1.05 1.00 5.00 Cognitive Restructuring 204 5 .72 3.18 .70 1.00 5.00 Rumination 204 5 .90 2.83 .89 1.00 5.00 Escape 204 3 .77 1.93 .78 1.00 4.00 Work-Family Conflict WIF 204 5 .90 2.90 .86 1.00 5.00 FIW 203 5 .89 2.01 .79 1.00 4.80 Note : All variables are measured on a five-point scale

PAGE 69

58 Table 5. Intercorrelations among Study Variables Variable 1 2 3 4 5 6 7 1 Conscientiousness (self) -2 Extraversion (self) .30** -3 Neuroticism (self) -.24** -.31** -4 Internal LOC (self) .55** .29** -.29** -5 Conscientiousness (SO)1 .46** .05 -.09 .28** -6 Extraversion (SO)1 .23** .68** -.19** .21** .19** -7 Neuroticism (SO)1 -.13 -.06 .62** -.21** -.28** -.16* -8 Problem Solving (w) .31** .18* -.29** .36** .22** .15* -.26** 9 Support Seeking I (w) .19** .24** -.01 .19** -.01 .18* .03 10 Support Seeking E (w) .06 .17* .07 .06 .01 .10 .05 11 Cog. Restructure (w) .09 .16* -.23** .24** -.03 .12 -.17* 12 Rumination (w) -.26** -.12 .58** -.30** -.17* -.09 .39** 13 Escape (w) -.09 -.05 .02 -.05 .08 -.01 .02 14 Problem Solving (f) .39** .24** -.20** .31** .06 .20** -.05 15 Support Seeking I (f) .13 .23** -.07 .09 -.19** .13 .04 16 Support Seeking E (f) .14* .22** .01 .14* -.04 .16* .05 17 Cog. Restructure (f) .14* .23** -.15* .21** -.10 .24** -.04 18 Rumination (f) -.21** -.06 .49** -.15* -.18* -.09 .31** 19 Escape (f) -.16* -.11 .02 -.12 .05 -.01 -.05 20 WIF -.10 -.11 .22** .00 -.01 -.09 .24** 21 FIW -.06 -.03 .18* .01 .00 -.05 .23** 22 Gender2 -.12 -.16* -.15* -.05 -.04 -.10 -.24** 23 Age -.09 .00 -.21** -.16* -.02 .03 -.14* 24 Work Time .07 -.11 .00 .06 .11 -.12 -.05 25 Household Income .05 .08 .04 -.03 -.01 .04 .09 26 SO Employment1,3 .00 -.03 .18** -.04 -.02 -.05 .20** 27 # Children at Home .06 .09 -.07 .02 -.02 .11 .08 28 RFD4 .07 .07 -.04 .04 .00 .14* .07 29 Elder Care3 -.08 -.05 -.04 -.07 .00 -.02 -.03 p <.05; ** p <.01; Ns ranged from 191 to 204 1SO=Significant other; 2Male=1; Female=0; 3Yes=1; No=0; 4Responsibility for Dependents Index

PAGE 70

59 Table 5 (Continued) Variable 8 9 10 11 12 13 14 1 Conscientiousness (self) 2 Extraversion (self) 3 Neuroticism (self) 4 Internal LOC (self) 5 Conscientiousness (SO)1 6 Extraversion (SO)1 7 Neuroticism (SO)1 8 Problem Solving (w) -9 Support Seeking I (w) .34** -10 Support Seeking E (w) -.09 .36** -11 Cog. Restructure (w) .37** .32** .10 -12 Rumination (w) -.33** -.05 .20** -.15* -13 Escape (w) -.31** -.21* .00 .05 .15* -14 Problem Solving (f) .56** .24** -.03 .25** -.22** -.19** -15 Support Seeking I (f) .17* .42** .36** .28** -.09 -.24** .37** 16 Support Seeking E (f) .10 .43** .63** .21** .03 -.14 .24** 17 Cog. Restructure (f) .26** .28** .11 .64** -.08 -.01 .45** 18 Rumination (f) -.21** .09 .25** -.05 .70** .09 -.10 19 Escape (f) -.24** -.10 .04 .10 .13 .59** -.33** 20 WIF -.08 -.11 -.14 -.07 .21** .16* -.16* 21 FIW -.12 -.11 -.12 -.14 .11 .14* -.07 22 Gender2 .10 -.21** -.45** .02 -.15* -.06 .02 23 Age .02 -.02 -.06 -.09 -.09 .12 .03 24 Work Time -.04 -.16* -.13 -.06 .01 .20** -.16* 25 Household Income .06 .03 .02 -.18* .01 .00 -.03 26 SO Employment1,3 -.05 .03 .16* -.03 .06 -.02 -.11 27 # Children at Home .00 .07 -.05 .00 .01 .03 .10 28 RFD4 .03 .09 -.04 .01 .01 -.00 .11 29 Elder Care3 -.03 .07 .11 .09 .12 .18* -.04 p <.05; ** p <.01; Ns ranged from 191 to 204 1SO=Significant other; 2Male=1; Female=0; 3Yes=1; No=0; 4Responsibility for Dependents Index

PAGE 71

60 Table 5 (Continued) Variable 15 16 17 18 19 20 21 1 Conscientiousness (self) 2 Extraversion (self) 3 Neuroticism (self) 4 Internal LOC (self) 5 Conscientiousness (SO)1 6 Extraversion (SO)1 7 Neuroticism (SO)1 8 Problem Solving (w) 9 Support Seeking I (w) 10 Support Seeking E (w) 11 Cog. Restructure (w) 12 Rumination (w) 13 Escape (w) 14 Problem Solving (f) 15 Support Seeking I (f) -16 Support Seeking E (f) .59** -17 Cog. Restructure (f) .47** .29** -18 Rumination (f) .07 .20** -.09 -19 Escape (f) -.27** -.11 -.09 .12 -20 WIF -.17* -.19** -.11 .13 .12 -21 FIW -.05 -.02 -.05 .12 .06 .52** -22 Gender2 -.27** -.39** -.10 -.17* -.02 .02 -.07 23 Age -.10 -.07 -.07 -.09 .09 -.18* -.13 24 Work Time -.15* -.23** -.14* -.05 .09 .37** .02 25 Household Income .01 .04 -.14* -.01 .03 -.03 .06 26 SO Employment1,3 .19** .26** -.01 .06 -.03 .07 .17* 27 # Children at Home .02 -.02 .18* .01 .00 .05 .31** 28 RFD4 .02 .00 .17* .01 -.02 .08 .33** 29 Elder Care3 .07 .08 .08 .13 .13 .08 .05 p <.05; ** p <.01; Ns ranged from 191 to 204 1SO=Significant other; 2Male=1; Female=0; 3Yes=1; No=0; 4Responsibility for Dependents Index

PAGE 72

61 Table 5 (Continued) Variable 22 23 24 25 26 27 28 1 Conscientiousness (self) 2 Extraversion (self) 3 Neuroticism (self) 4 Internal LOC (self) 5 Conscientiousness (SO)1 6 Extraversion (SO)1 7 Neuroticism (SO)1 8 Problem Solving (w) 9 Support Seeking I (w) 10 Support Seeking E (w) 11 Cog. Restructure (w) 12 Rumination (w) 13 Escape (w) 14 Problem Solving (f) 15 Support Seeking I (f) 16 Support Seeking E (f) 17 Cog. Restructure (f) 18 Rumination (f) 19 Escape (f) 20 WIF 21 FIW 22 Gender2 -23 Age .16* -24 Work Time .14 -.11 -25 Household Income -.14 .21** .17* -26 SO Employment1,3 -.27** -.25** -.02 .16* -27 # Children at Home -.01 .03 -.18* .07 -.18** -28 RFD4 -.01 -.03 -.16* .07 -.15* .98** -29 Elder Care3 -.08 .16* .01 -.18* .05 -.02 -.06 p <.05; ** p <.01; Ns ranged from 191 to 204 1SO=Significant other; 2Male=1; Female=0; 3Yes=1; No=0; 4Responsibility for Dependents Index

PAGE 73

62 independent variables appeared relatively constant across all levels of the dependent variable, supporting the assu mption of homoscedasticity. The data were also assessed for normali ty and outliers. Specifically, skewness and kurtosis values were computed, and histogram s and box plots were graphed. For the most part, the skewness and kurtosis values were small (less than one). A few variables, including locus of control and number of ch ildren living at home, were kurtotic, and number of children living at home and the Responsibility for Dependents index were positively skewed. However, given that linear regression is fairly robust with respect to the normality assumption, it was deemed unnecessary to normalize the data. Box plots were examined for extreme outlier s, defined as data points falling more than three times the interquartile range aw ay from the upper or lower quartile. Two individuals had extreme scores for work hour s, reporting working an average of 80 hours per week. These values are plausible a nd were thus kept in the data set. Hypothesis Testing Hypotheses 1 through 7 were tested with zer o-order correlations as well as with regression to determine whether relationships di ffered with and without control variables. For each equation, independent variables were entered first, followed by the control variable(s) related to the dependent variab le. Hypotheses 8 through 15 involve mediation and were tested following Baron and Ke nny’s (1986) three-step procedure. When sufficient support was found, the Sobel test was also used to test the mediation hypotheses. An alpha level of .05 was used for all analyses. To preserve power, control variables were chosen based on thei r relationship with the dependent variable (DV) of interest. Of the potential control variables for work-

PAGE 74

63 family conflict, only work time related to WIF ( r = .37, p < .01). Number of children at home ( r = .31, p < .01), the Responsibility for Dependents (RFD) index ( r = .33, p < .01), and significant other’s employment status related to FIW ( r = .17, p < .05). Specifically, participants with more children living at home, a higher RFD index, and/or an employed significant other tended to experi ence higher levels of FIW. Given the nature of the RFD index, there was a high correlation between RFD and number of children at home ( r = .98). The RFD index was chosen as a control va riable over number of children at home, given that it takes into account the age of each child living at home, in addition to the overall number. Age was included as a potentia l control variable for the coping variables, but it was not significantly related to any of the coping styles. Gender, on the other hand, was related to instrumental (work: r = -.21, p < .01; family: r = -.27, p < .01) and to emotional (work: r = -.45, p < .01; family: r = -.39, p < .01) support seeking and to rumination (work: r = -.15, p < .05; family: r = -.17, p < .05); females reported using these coping styles more frequently than men, both for work and for family stressors. Thus, gender was entered as a control variable for analyses in wh ich support seeking or rumination was the dependent variable. Ne ither potential control variable was significantly related to the other coping styl es, so hypotheses involving problem solving, cognitive restructuring, and escape were te sted with zero-order correlations only. For hypotheses involving personalit y, significant othe r-ratings of conscientiousness, extraversion, and neuroticis m were used, unless otherwise specified. However, for comparative purposes, self-rati ngs, as well as the average of selfand significant other-ratings, were examined as we ll. These results are provided in Appendix F. Only self-ratings of internal locus of control were collected.

PAGE 75

64 Personality and Work-Family Conflict The first hypothesis pr edicted that workfamily conflict would relate to the four personality variables under investigation. Specifically, it was expected that conscienti ousness (1a), extraversi on (1b), and internal locus of control (1c) would ne gatively relate to WIF and FIW, while neuroticism (1d) would positively relate to WIF and FIW. C ontrary to Hypothesis 1, conscientiousness, extraversion, and internal locus of control were not significantly correlated with WIF ( r s = -.01, -.09, and .00, respectiv ely) or with FIW ( r s = .00, -.05, and .01, respectively). Additionally, as presen ted in Tables 6, 7, and 9, the beta weights for conscientiousness, extraversion, and internal lo cus of control were not significant after accounting for control variables ( s = -.05, -.05, and -.02, respectively for WIF; s = .01, -.09, and .00, respectively for FIW). On the other hand, neur oticism was significantly related to WIF ( r = .24, p < .01; = .26, p < .01) and to FIW ( r = .23, p < .01; = .17, p < .01). Regression results for neuroticism are presented in Table 8. For comparative purposes, self-ratings a nd the average of selfand significant other-ratings were examined for conscienti ousness, extraversion, and neuroticism. As demonstrated by both correlations and regres sion analysis (presented in Tables 32 through 37), the results for se lf-ratings and average-rati ngs mirrored those found for significant other-ratings. Thus across both sources of rati ngs, Hypotheses 1a, 1b, and 1c were not supported wh ile Hypothesis 1d was. Personality and Coping Style Hypotheses 2 through 5 predicted that conscientiousness, extraversion, neuroticism, and internal locus of control would each relate to a subset of the fi ve coping styles. For example, Hypothesis 2 predicted that conscientiousness would positively relate to problem solving (2a) and to positive

PAGE 76

65 Table 6. Regression of WIF and FIW on Cons cientiousness (Significant Other-Report) Dependent Variable WIF FIW Independent Variable Conscientiousness -.05 .01 Control Variables Work Time .38** SO1 Employment .23** RFD2 .37** in R2 .14 .16 Overall R2 .14 .16 Adjusted R2 .13 .15 Overall F 16.52** 12.51** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index Table 7. Regression of WIF and FIW on Ex traversion (Significant Other-Report) Dependent Variable WIF FIW Independent Variable Extraversion -.05 -.09 Control Variables Work Time .37** SO1 Employment .22** RFD2 .38** in R2 .13 .16 Overall R2 .14 .17 Adjusted R2 .13 .15 Overall F 16.47** 13.22** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index

PAGE 77

66 Table 8. Regression of WIF and FIW on Neuroticism (Significant Other-Report) Dependent Variable WIF FIW Independent Variable Neuroticism .26** .17** Control Variables Work Time .39** SO1 Employment .19** RFD2 .35** in R2 .15 .14 Overall R2 .21 .19 Adjusted R2 .20 .18 Overall F 25.89** 15.30** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index Table 9. Regression of WIF and FIW on In ternal Locus of Control (Self-Report) Dependent Variable WIF FIW Independent Variable Internal Locus of Control -.02 .00 Control Variables Work Time .37** SO1 Employment .23** RFD2 .37** in R2 .14 .16 Overall R2 .14 .16 Adjusted R2 .13 .15 Overall F 16.27** 12.51** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index

PAGE 78

67 cognitive restructuring (2b) and negatively rela te to rumination (2c) and to escape (2d). In terms of coping with work stressors, consci entiousness significantly related to problem solving ( r = .22, p < .01) and to rumination ( r = -.17, p < .05) but not to positive cognitive restructuring ( r = -.03) or to escape ( r = .08). On the other hand, for coping with family stressors, only rumination was signifi cantly related to conscientiousness ( r = -.18, p < .05). The multiple regression results for rumination, presented in Table 10, provide support for a relationship between consci entiousness and rumination for work ( = -.18, p < .05) and for family ( = -.18, p < .01) stressors after controlling for gender. Hypothesis 2 was also examined us ing selfand aver age-ratings of conscientiousness. The results were similar, with a couple of not eworthy differences. Consistent with the significant other-ratings, rumination for work and for family stressors were negatively related to selfand to aver age-ratings of conscientiousness, both with and without control variables (see Tables 38 and 39 for regression results). Additionally, selfand average-ratings of conscientiousne ss were significantly related to problem solving for work stressors. However, unlike the significant other-rat ings, problem solving for family stressors significantly relate d to selfand to average-ratings of conscientiousness ( r = .39, p < .01 and r = .30, p < .01, respectively). Additionally, selfratings (though not average-ratings) of consci entiousness significantl y related to positive cognitive restructuring ( r = .14, p < .05) and to escape ( r = -.16, p < .05) for family stressors. Thus, limited support was found for Hypothesis 2, though results differed by coping style, rating source, a nd coping style context. Sp ecifically Hypothesis 2a was supported for coping with work stressors (bot h rating sources of c onscientiousness) and for coping with family stressors (selfand average-ratings of personality only);

PAGE 79

68 Hypothesis 2c was fully supported; and H ypotheses 2b and 2d were largely unsupported, though self-ratings of conscien tiousness related to coping with family stressors for both coping styles. Additionally, though not hypothe sized, significant other-ratings of conscientiousness negatively related to instrumental suppor t seeking for family stressors ( r = -.19, p < .01), while self-ratings of conscientiousness positively related to instrumental support seeking for work stressors ( r = .19, p < .01) and to emotional support seeking for family stressors ( r = .14, p < .05). Hypothesis 3 predicted that extraversi on would positively relate to problem solving (3a), to support seeking (3b), and to positive cognitive restructuring (3c) and negatively relate to rumination (3d). Regressi on results for Hypothesis 3 are presented in Table 11. Hypotheses 3a was supported; extrav ersion was significantly related to problem solving for work ( r = .15, p < .05) and for family stressors ( r = .20, p < .01). Additionally, extraversion was significantly related to inst rumental support seeking for work stressors ( r = .18, p < .05) and to emotional support seeking for family stressors ( r = .16, p < .05) but not to emotional suppor t seeking for work stressors ( r = .10) or to instrumental support seeking for family stressors ( r = .13). While the relationship between instrumental support seeking for wo rk and extraversion remained significant after controlling for gender ( = .16, p < .05), the relationship between emotional support seeking for family and extraversion did not ( = .12). Extraversion was also significantly related to positive cognitive restructuring for family stressors ( r = .24, p < .01), but there was no support for the relationship between extraversion and positive cognitive restructuring for work ( r = .12), rumination for work ( r = -.09, = -.11), or rumination

PAGE 80

69 for family (r = -.09, = -.11). As expected, there was no significant relationship between extraversion and escape for work ( r = -.01) or for family ( r = -.01) stressors. The results using selfand average-ratings of extraversion were very similar. However, unlike the significant other-rati ngs, both selfand average-ratings of extraversion significantl y related to all forms of social support seeking (instrumental for work: r = .24, p < .01 for self, r = .23, p < .01 for average; emotional for work: r = .17, p < .05 for self, r = .15, p < .05 for average; instrumental for family: r = .23, p < .01 for self, r = .19, p < .01 for average; emotional for family: r = .22, p < .01 for self, r = .21, p < .01 for average), as well as to cogni tive restructuring for work stressors ( r = .16, p < .05 for self, r = .15, p < .05 for average). As shown in Tables 40 and 41 (see Appendix F for regression results using selfand average-rati ngs of personality), after controlling for gender, the relationships between selfa nd average-ratings of extraversion and instrumental support seeking for work stressors ( = .21, p < .01 and = .20, p < .01, respectively), instrumental support seeking for family stressors ( = .19, p < .01 and = .16, p < .05, respectively), and emotional support seeking for family stressors ( = .16, p < .05 and = .15, p < .05, respectively) remained si gnificant. Additionally, selfand average-ratings of extraversion were signi ficantly related to rumination for work stressors, after controlling for gender ( = -.15, p < .05 and = -.14, p < .05, respectively). Overall, full support was f ound for Hypothesis 3a; partial support was found for Hypothesis 3b (depending on type of support seeking and source of personality ratings); and Hypothesis 3c was supported for coping with family st ressors (both rating sources of extraversion) and for coping with work stressors (selfand average-ratings only). Hypothesis 3d was largely unsupported.

PAGE 81

70 Hypothesis 4 predicted that neuroticis m would negatively relate to problem solving (4a) and to positive cognitive rest ructuring (4b) and positively relate to rumination (4c) and to escape (4d). For copi ng with work stressors, Hypotheses 4a, 4b, and 4c were fully supported; neuroticism significantly related to problem solving ( r = .26, p < .01), to positive cognitive restructuring ( r = -.17, p < .05), and to rumination ( r = .39, p < .01), though not to escape ( r = .02). However, for coping with family stressors, only rumination was significantl y related to neuroticism ( r = .31, p < .01), while neuroticism was not significantl y related to problem solving ( r = -.05), to positive cognitive restructuring ( r = -.04), or to escape ( r = -.05). As presented in Table 12, the relationship between neuroticism and rumina tion remained significant after controlling for gender ( = .38, p < .01 for work; = .29, p < .01 for family). As expected, the relationships between neuroticism and s upport seeking for work (instrumental: r = .03; emotional: r = .05) and for family (instrumental: r = .04; emotional: r = .05) stressors were not significant. In terms of selfand average-ratings of ne uroticism, the only differences were that selfand average-ratings of neuroticism significantly rela ted to problem solving for family stressors ( r = -.20, p < .01 and r = -.14, p < .05, respectively), and self-ratings (though not average-ratings) of neuroticism significantly rela ted to positive cognitive restructuring for family stressors ( r = -.15, p < .05). (Regression results for selfand average-ratings of neuroticism and coping ar e presented in Tables 42 and 43.) Overall, full support was found for Hypothesis 4c, whil e Hypotheses 4a and 4b were supported for coping with work stressors ( both rating sources of neuro ticism) and for coping with family stressors (self-ratings only). Hypothesis 4d received no support.

PAGE 82

71 Hypothesis 5 predicted that internal locu s of control would positively relate to problem solving (5a) and to positive cognitive restructuring (5b) and negatively relate to escape (5c). For both work and family stre ssors, internal locus of control was significantly related to problem solving ( r = .36, p < .01 for work; r = .31, p < .01 for family) and to positive cognitive restructuring ( r = .24, p < .01 for work; r = .21, p < .01 for family) but unrelated to escape ( r = -.05 for work; r = -.12 for family). Thus, Hypotheses 5a and 5b were fully supporte d, while no support was found for Hypothesis 5c. Although not hypothesized, internal locus of control was also related to instrumental support seeking for work ( r = .19, p < .01) and emoti onal support seeking for family ( r = .14, p < .05). Moreover, the relationship be tween internal locus of control and rumination was significant for both work ( r = -.30, p < .01) and family ( r = -.15, p < .05) stressors. Table 10. Regression of Coping Style on C onscientiousness (Significant Other-Report) Dependent Variable Rumination (work) Rumination (family) Independent Variable Conscientiousness -.18* -.18** Control Variable Gender -.16* -.18* in R2 .02 .03 Overall R2 .05 .06 Adjusted R2 .04 .05 Overall F 5.70** 6.60** p <.05; ** p <.01; s are standardized regression we ights from the final equation

PAGE 83

72 Table 11. Regression of Coping Style on Extraversion (Significant Other-Report) p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Instrumental Support Seeking; 2Emotional Support Seeking Table 12. Regression of Coping Style on Neuroticism (Significant Other-Report) Dependent Variable Rumination (work) Rumination (family) Independent Variable Neuroticism .38** .29** Control Variable Gender -.06 -.10 in R2 .00 .01 Overall R2 .16 .11 Adjusted R2 .15 .10 Overall F 18.46** 11.87** p <.05; ** p <.01; s are standardized regression we ights from the final equation Dependent Variable Support Seeking – I1 (w) Support Seeking – E2 (w) Support Seeking – I1 (f) Support Seeking – E2 (f) Rumination (w) Rumination (f) Independent Variable Extraversion .16* .06 .10 .12 -.11 -.11 Control Variable Gender -.20** -.45** -.26** -.38** -.16* -.18* in R2 .04 .20 .07 .14 .03 .03 Overall R2 .07 .21 .08 .17 .03 .04 Adjusted R2 .06 .20 .07 .16 .02 .03 Overall F 7.39** 26.45** 8.80** 20.21** 3.51* 4.18*

PAGE 84

73 Coping Style and Work-Family Conflict The next two hypotheses predicted that coping for work stressors would relate to WIF (Hypothesis 6), while coping for family stressors would relate to FI W (Hypothesis 7). Specifically, problem solving (6a and 7a), support seeking (6b and 7b), and positive cogni tive restructuring (6c and 7c) were expected to negatively relate to work-family conflict, while rumi nation (6d and 7d) and escape (6e and 7e) were expected to positively relate to work-family conflict. Regression results for Hypotheses 6 and 7 are provide d in Tables 13 through 24. Although domainspecific relationships (i.e., coping with work and WIF; coping with family and FIW) were hypothesized, cross-domain relationships (i.e., coping with work and FIW; coping with family and WIF) are also presented in these tables for exploratory purposes. In terms of coping with work stress ors, WIF was significantly related to rumination ( r = .21, p < .01) and to escape ( r = .16, p < .05) but not to problem solving ( r = -.08), to support seeking (instrumental: r = -.11; emotional: r = -.14), or to positive cognitive restructuring ( r = -.07). Additionally, only rumination for work stressors remained significant after c ontrolling for work time ( = .21, p < .01); the beta weights for problem solving ( = -.06), support seeking (instrumental : = -.05; emotional: = .09), cognitive restructuring ( = -.05), and escape ( =.09) were not significant. Thus, Hypothesis 6d was fully suppor ted; Hypothesis 6e was partially supported; and Hypotheses 6a, 6b, and 6c were unsupported. While Hypothesis 6 received some suppor t, Hypothesis 7 did not. None of the coping styles for family stressors were sign ificantly related to FI W, with or without accounting for the control variables: r = -.07, = -.09 for problem solving; r = -.05, = .09 for instrumental support seeking; r = -.02, = -.08 for emoti onal support seeking; r =

PAGE 85

74 Table 13. Regression of WIF and FIW on Problem Solving for Work Stressors Dependent Variable WIF FIW Independent Variable Problem Solving (work) -.06 -.11 Control Variables Work Time .37** SO1 Employment .22** RFD2 .37** in R2 .14 .16 Overall R2 .14 .17 Adjusted R2 .13 .16 Overall F 16.74** 13.75** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index Table 14. Regression of WIF and FIW on Problem Solving for Family Stressors Dependent Variable WIF FIW Independent Variable Problem Solving (family) -.11 -.09 Control Variables Work Time .36** SO1 Employment .22** RFD2 .38** in R2 .12 .16 Overall R2 .15 .17 Adjusted R2 .14 .15 Overall F 17.68** 13.27** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index

PAGE 86

75 Table 15. Regression of WIF and FIW on In strumental Support Seeking for Work Stressors Dependent Variable WIF FIW Independent Variable Instrumental Support Seeking (work) -.05 -.15* Control Variables Work Time .37** SO1 Employment .23** RFD2 .38** in R2 .13 .17 Overall R2 .14 .18 Adjusted R2 .13 .17 Overall F 16.51** 14.51** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index Table 16. Regression of WIF and FIW on Emotional Support Seeking for Work Stressors Dependent Variable WIF FIW Independent Variable Emotional Support Seeking (work) -.09 -.14* Control Variables Work Time .36** SO1 Employment .25** RFD2 .36** in R2 .13 .16 Overall R2 .15 .18 Adjusted R2 .14 .17 Overall F 17.29** 14.34** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index

PAGE 87

76 Table 17. Regression of WIF and FIW on In strumental Support Seeking for Family Stressors Dependent Variable WIF FIW Independent Variable Instrumental Support Seeking (family) -.12 -.09 Control Variables Work Time .36** SO1 Employment .24** RFD2 .37** in R2 .12 .16 Overall R2 .15 .17 Adjusted R2 .14 .15 Overall F 18.05** 13.22** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index Table 18. Regression of WIF and FIW on Emotional Support Seeking for Family Stressors Dependent Variable WIF FIW Independent Variable Emotional Support Seeking (family) -.10 -.08 Control Variables Work Time .35** SO1 Employment .25** RFD2 .37** in R2 .11 .16 Overall R2 .15 .17 Adjusted R2 .14 .15 Overall F 17.72** 13.09** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index

PAGE 88

77 Table 19. Regression of WIF and FIW on Pos itive Cognitive Restructuring for Work Stressors Dependent Variable WIF FIW Independent Variable Cognitive Restructuring (work) -.05 -.13* Control Variables Work Time .37** SO1 Employment .22** RFD2 .37** in R2 .14 .16 Overall R2 .14 .18 Adjusted R2 .13 .16 Overall F 16.48** 14.17** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index Table 20. Regression of WIF and FIW on Positive Cognitive Restructuring for Family Stressors Dependent Variable WIF FIW Independent Variable Cognitive Restructuring (family) -.06 -.12 Control Variables Work Time .36** SO1 Employment .23** RFD2 .39** in R2 .13 .17 Overall R2 .14 .17 Adjusted R2 .13 .16 Overall F 16.69** 13.76** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index

PAGE 89

78 Table 21. Regression of WIF and FIW on Rumination for Work Stressors Dependent Variable WIF FIW Independent Variable Rumination (work) .21** .09 Control Variables Work Time .37** SO1 Employment .22** RFD2 .36** in R2 .14 .16 Overall R2 .18 .17 Adjusted R2 .18 .16 Overall F 22.50** 13.34** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index Table 22. Regression of WIF and FIW on Rumination for Family Stressors Dependent Variable WIF FIW Independent Variable Rumination (family) .15* .11 Control Variables Work Time .38** SO1 Employment .22** RFD2 .36** in R2 .15 .16 Overall R2 .16 .17 Adjusted R2 .15 .16 Overall F 19.39** 13.56** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index

PAGE 90

79 Table 23. Regression of WIF and FIW on Escape for Work Stressors Dependent Variable WIF FIW Independent Variable Escape (work) .09 .14* Control Variables Work Time .35** SO1 Employment .22** RFD2 .36** in R2 .12 .16 Overall R2 .15 .18 Adjusted R2 .14 .16 Overall F 17.37** 14.24** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index Table 24. Regression of WIF and FIW on Escape for Family Stressors Dependent Variable WIF FIW Independent Variable Escape (family) .09 .06 Control Variables Work Time .37** SO1 Employment .23** RFD2 .37** in R2 .13 .16 Overall R2 .15 .16 Adjusted R2 .14 .15 Overall F 17.34** 12.88** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index

PAGE 91

80 -.05, = -.12 for positive cognitive restructuring; r = .12, = .11 for rumination; and r = .06, = .06 for escape. Although not hypothesized, cross-domain rela tionships were also examined for exploratory purposes. The relationship between FIW and escape for work stressors was significant ( r = .14, p < .05). Additionally, after controlling for significant other’s employment status and for the RFD index, instrumental support seeking ( = -.15, p < .05), emotional support seeking ( = -.14, p < .05), positive cognitive restructuring ( = .13, p < .05), and escape ( = .14, p < .05) for work stressors si gnificantly related to FIW. Furthermore, problem solving ( r = -.16, p < .05) and support seeking (instrumental: r = .17, p < .05; emotional: r = -.19, p < .01) for family stressors significantly related to WIF, and the relationship between WIF and rumina tion for family stressors was significant after controlling for work time ( = .15, p < .05). Coping Style as a Mediator betwee n Personality and Work-Family Conflict Hypotheses 8 through 15 predic ted that coping style woul d mediate the relationship between personality and work-family conflic t. Following the strate gy outlined in Baron and Kenny (1986), three regression equations are required to demonstrate support for mediation: (1) the mediator (coping style) is regressed onto the independent variable (personality); (2) the dependent variable (W IF/FIW) is regressed onto the IV; and (3) the dependent variable is regressed onto both th e IV and the mediator. Support for mediation requires four conditions: (1) a significant rela tionship between the IV and the DV; (2) a significant relationship between the IV and th e mediator, (3) a sign ificant relationship between the mediator and the DV, and (4) afte r controlling for the mediator (step 3), the effect of the independent variable on the dependent variable becomes non-significant

PAGE 92

81 (full mediation) or decreases in streng th (partial mediati on; Baron & Kenny, 1986). Additionally, in step three, the relationship between the me diator and the DV should be established after controlling for the effect of the IV. Showing a co rrelation between the mediator and the DV is not sufficient in and of itself, as the correla tion could be a result of the IV causing both the mediator and the outcome (thus indicating a spurious relationship). As previously mentioned, conscientiousne ss, extraversion, and internal locus of control were not related to WIF or to FIW. Thus, the first condition was not satisfied for Hypotheses 8 through 11 and 14 through 15, and no subsequent analyses were performed. On the other hand, neuroticism was significan tly related to WIF and to FIW, thereby satisfying condition one for Hypotheses 12 a nd 13, which predicte d that coping style would mediate the relationship between neuroticism and work-family conflict. Specifically, problem solving for work (12a) and for family (13a), positive cognitive restructuring for work (12b) and for family (13b), rumination for work (12c) and for family (13c), and escape for work (12d) and for family (13d) were hypothesized to function as mediators between neurotic ism and WIF (Hypothesis 12) or between neuroticism and FIW (Hypothesis 13). For both work and family stressors and acr oss all sources of personality ratings, escape was not related to neuroticism. Theref ore, the second condition was not satisfied for Hypotheses 12d and 13d, and no subseque nt analyses were performed. Although problem solving, positive cognitive restruct uring, and rumination for work stressors significantly related to neuroticism, only ru mination was significantly related to WIF. Thus, the third condition was not satisfied fo r Hypotheses 12a or 12b. Similarly, none of

PAGE 93

82 the coping styles for family stressors related to FIW, so no additional analyses were conducted for Hypothesis 13. Thus, the only mediation hypothesis that met Baron and Kenny’s (1986) first two conditions and was significantly related to th e DV (indicating initial support for condition three) was Hypothesis 12c, that ruminati on for work stressors would mediate the relationship between neuroticism and WIF. Hier archical regression was performed to test conditions three and four, with control va riables entered in step one, neuroticism (significant other-ratings) ente red in step two, and ruminati on for work stressors entered in step three. The hypothesis was also tested without control variab les, but the results were the same, so they are not presented here. As shown in Table 25, neuroticism remain ed significant in step three, after rumination was added, but the beta we ight decreased from .26 to .21 ( p s < .01). However, rumination for work stressors was not significan tly related to WIF, after controlling for neuroticism ( = .13). Thus, Baron and Kenny’s (1986) third condition was not met. As an additional test of mediation, the Sobel test was performed (Preacher & Hayes, 2004; Sobel, 1982). This method provides a single test of the significance of the indirect effect and is recommended as an alternative and/ or supplement to Baron and Kenny’s (1986) method (Kenny, 2008). The Sobel test provi ded support that rumination for work stressors mediated the relationshi p between neuroticism and WIF ( z = 2.74, p < .01), thus providing partial support for Hypothesis 12c. Hypothesis 12c was also examined usi ng both selfand average-ratings of neuroticism, as presented in Tables 44 a nd 45. Unlike the results for significant otherratings of neuroticism, the relationship be tween self-reported neuroticism and WIF was

PAGE 94

83 non-significant in step three of the regr ession equation, after rumination was added, thereby meeting Baron and Kenny’s (1986) four th condition for mediation. However, the beta weights for rumination were not significant for either the self( =.12) or for the average-ratings ( = .09) of neuroticism. Thus, consis tent with the results for significant other-ratings of neuroticism, condition three was not met. On the other hand, the Sobel test was significant for both self( z = 2.98, p < .01) and for average-ratings ( z = 2.90, p < .01) of neuroticism, also consistent with the results using significant other-reported neuroticism. Table 25. Mediated Regression of WIF on Rumination for Work Stressors and Neuroticism (Significant Other-Report) Dependent Variable: WIF Step 1 Step 2 Step 3 Control Variable Work Time .37** .39** .38** Independent Variable Neuroticism .26** .21** Mediator Rumination (work) .13 in R2 -.07 .01 Overall R2 .14 .21 .22 Adjusted R2 .14 .20 .21 Overall F 32.60** 25.89** 18.70** p <.05; ** p <.01 Supplementary Analysis Although the present study hypothesized that coping style would mediate the relationship between personality and work-f amily conflict, coping style can also be conceptualized as a moderator. Thus, as a s upplementary analysis, the five coping styles,

PAGE 95

84 for both work and for family stressors, were examined as potential moderators between the four personality variables and both WI F and FIW. Moderated regression was used, with the independent variables en tered in step one and the inte raction term entered in step two. With the exception of internal locus of control, significan t other-ratings of personality were used. Of the 96 possible regres sion equations, only six had a significant interaction term. The moderated regression results for thes e six equations are pr esented in Tables 26 through 31, and the interactions are depicted graphically in Figur es 1 through 6. (Note that the figures do not include control variables.) As pres ented in Table 26, problem solving for work stressors moderated the relationship between conscientiousness and FIW ( = -1.44, p < .05). Specifically, as shown in Figure 1, at low levels of conscientiousness, use of problem solving for wo rk stressors was not related to the degree of FIW reported; however, at high levels of conscientiousness, individuals who did not use problem solving to manage work stressor s tended to report higher levels of FIW, while individuals who used problem solving te nded to report lower levels of FIW. A similar pattern emerged for positive cognitive restructuring used to cope with family stressors, which interacted with consci entiousness to explain variance in FIW ( = -1.09, p < .05; see Table 27 and Figure 2). Additionally escape for family stressors moderated the relationship between c onscientiousness and FIW ( = -1.41, p < .01; see Table 28). As shown in Figure 3, at low levels of cons cientiousness, individuals who used escape to cope with family stressors tended to experience higher leve ls of FIW than individuals who did not. On the other hand, at high leve ls of conscientiousness, there was little difference in levels of FIW, regardless of th e use of escape to cope with family stressors.

PAGE 96

85 Extraversion, neuroticism, and internal locus of control were each involved in one significant interaction. Extraversion interacted with escape for family stressors to explain variance in WIF ( = -.74, p < .05; see Table 29). The natu re of this interaction is depicted in Figure 4 and is consistent w ith that of conscientiousness and escape for family stressors. Additionally, as shown in Table 30, for individuals low on neuroticism, low rumination for family stressors was associ ated with lower levels of FIW, while high rumination was associated with higher levels of FIW. Conversely, at high levels of neuroticism, there was little relationship between rumination for family stressors and FIW. This interaction is shown graphically in Figure 5. Finally, internal locus of control interacted with instrumental support seeking for family stre ssors to explain variance in WIF ( = 1.53, p < .05; see Table 31). As depicted in Figure 6, individuals with an internal locus of control expe rienced similar levels of WIF, irrespective of the amount of instrumental support seeking used to cope w ith family stressors. On the other hand, for individuals with an external lo cus of control, high use of instrumental support seeking for family stressors was associated with lower le vels of WIF, while low use of instrumental support seeking for family stressors was a ssociated with higher levels of WIF.

PAGE 97

86 Table 26. Moderated Regression of FIW on Co nscientiousness (Significant Other-Report) and Problem Solving for Work Stressors Dependent Variable: FIW Step 1 Step 2 Independent Variables Conscientiousness .03 .95* Problem Solving (work) -.12 .78 Interaction Term Conscientiousness X Problem Solving (work) -1.44* in R2 -.02 Overall R2 .02 .03 Adjusted R2 .01 .02 Overall F 1.48 2.32 p <.05 Table 27. Moderated Regression of FIW on Co nscientiousness (Significant Other-Report) and Positive Cognitive Restructuring for Family Stressors Dependent Variable: FIW Step 1 Step 2 Independent Variables Conscientiousness -.00 .64* Cognitive Restructuring (family) -.05 .89* Interaction Term Conscientiousness X Cognitive Restructuring (family) -1.09* in R2 -.02 Overall R2 .00 .03 Adjusted R2 -.01 .01 Overall F .255 1.71 p <.05

PAGE 98

87 Table 28. Moderated Regression of FIW on Co nscientiousness (Significant Other-Report) and Escape for Family Stressors Dependent Variable: FIW Step 1 Step 2 Independent Variables Conscientiousness -.00 .46* Escape (family) .06 1.35** Interaction Term Conscientiousness X Escape (family) -1.41** in R2 -.04 Overall R2 .00 .04 Adjusted R2 -.01 .02 Overall F .36 2.66* p <.05; ** p <.01 Table 29. Moderated Regression of WIF on Ex traversion (Significant Other-Report) and Escape for Family Stressors Dependent Variable: WIF Step 1 Step 2 Independent Variables Extraversion -.09 .27 Escape (family) .12 .75* Interaction Term Extraversion X Escape (family) -.74* in R2 -.02 Overall R2 .02 .04 Adjusted R2 .01 .03 Overall F 2.34 2.98* p <.05

PAGE 99

88 Table 30. Moderated Regression of FIW on Ne uroticism (Significant Other-Report) and Rumination for Family Stressors Dependent Variable: FIW Step 1 Step 2 Independent Variables Neuroticism .21** .65** Rumination (family) .05 .59* Interaction Term Neuroticism X Rumination (family) -.80* in R2 -.02 Overall R2 .06 .07 Adjusted R2 .05 .06 Overall F 5.86** 5.32** p <.05; ** p <.01 Table 31. Moderated Regression of WIF on Inte rnal Locus of Control (Self-Report) and Instrumental Support Seeking for Family Stressors Dependent Variable: WIF Step 1 Step 2 Independent Variables Internal Locus of Control .02 -.48* Instrumental Support Seeking (family) -.17* -1.56* Interaction Term Internal Locus of Control X Instrumental Support Seeking (family) 1.53* in R2 -.03 Overall R2 .03 .05 Adjusted R2 .02 .04 Overall F 3.04* 3.82* p <.05; ** p <.01

PAGE 100

89 1 2 3 4 12 ConscientiousnessFIW Low Problem Solving for Work Stressors High Problem Solving for Work Stressors Figure 1. Problem solving for work stressor s as a moderator between conscientiousness (significant other-report) and FIW. 1 2 3 4 12 ConscientiousnessFIW Low Cognitive Restructuring for Family Stressors High Cognitive Restructuring for Family Stressors Figure 2. Positive cognitive restructuring for family stressors as a moderator between conscientiousness (significan t other-report) and FIW.

PAGE 101

90 1 2 3 4 12 ConscientiousnessFIW Low Escape for Family Stressors High Escape for Family Stressors Figure 3. Escape for family stressors as a moderator between conscientiousness (significant other-report) and FIW. 1 2 3 4 12 ExtraversionWIF Low Escape for Family Stressors High Escape for Family Stressors Figure 4. Escape for family stressors as a moderator between extraversion (significant other-report) and WIF.

PAGE 102

91 1 2 3 4 12 NeuroticismFIW Low Rumination for Family Stressors High Rumination fo r Family Stressors Figure 5. Rumination for family stressors as a moderator between neuroticism (significant other-report) and FIW. 1 2 3 4 12 Internal Locus of ControlWIF Low Instrumental Support Seeking for Family Stressors High Instrumental Support Seeking for Family Stressors Figure 6. Instrumental support seeking for fa mily stressors as a moderator between internal locus of contro l (self-report) and WIF.

PAGE 103

92 Chapter Four Discussion The present study aimed to contribute to the literature by examining two relatively neglected sets of variables in the work-family literature, pe rsonality and coping styles, as well as the mechanisms by which these variab les are related. Using a stress and coping framework, as well as Hobfoll’s (1989) cons ervation of resources model and a spillover perspective of WFC, coping was proposed as a mediator between personality and workfamily conflict. Although two dissertati ons recently examined this proposition (Andreassi, 2007; Smoot, 2005), the present study aimed to expa nd this line of research by examining coping styles as action tende ncies, differentiating between coping with work and coping with family stresso rs, and collecting dual-source data. The present study offers several key fi ndings. Consistent with past research, neuroticism related to both directions of work-family c onflict, offering support for the notion that certain dispositional variables do play a role in the WFC experience. There was also some support for the role of coping in the process, as problem solving, support seeking, rumination, and escape each related to at least one direc tion of work-family conflict. Furthermore, the pr esent study provides some insi ght into the process underlying the relationship between neuroticism and WI F, in that some support was found for the mediating role of rumination. Finally, limite d support was found for the moderating role

PAGE 104

93 of coping in the relationship between personality and WFC, an analysis performed for exploratory purposes only. Four sets of study hypotheses were propos ed, involving the relationship between personality and WFC, the relationship betw een personality and coping, the relationship between coping and WFC, and the mediating ro le of coping in the relationship between personality and WFC. In the following s ections, the results for each set of study hypotheses are examined in more detail, along with the results from the supplementary analysis. Theoretical and pr actical implications are then discussed, followed by study limitations and future directions. Personality and Work-Family Conflict While work-family conflict was expected to relate to conscientiousness, extraversion, neuroticism and lo cus of control, only neuro ticism was significantly related to WIF and FIW. Although research examining personality as an antecedent to WFC is limited, the relationship between neuroticis m and WFC has received the most support (e.g., Andreassi, 2007; Bruck & Allen, 2003; Smoot, 2005; Wayne et al., 2004), consistent with the present study’s findings. Research findings linki ng conscientiousness, extraversion, and locus of control with work-family conflict are less consistent, with results varying across study, gender, and type of work-family conf lict (e.g., Andreassi, 2007; Andreassi & Thompson, 2007; Bruc k & Allen, 2003; Kinnunen et al., 2003; Wayne et al., 2004). The inconsistent findings may suggest the pres ence of moderators. Potential moderators of the personality-WFC relationship are discussed subsequently in the Future Directions section.

PAGE 105

94 Despite some inconsistencies in the lite rature, several resear chers have reported significant associations between conscientiousness, extraversion, and locus of control and work-family conflict (e.g., Andreassi & Thompson, 2007; Grzywacz & Marks, 2000; Kinnunen et al., 2003; Wayne et al., 2004). Thus, the null relationships in the present study are surprising. One possible explanation for the null findi ngs is range restriction of some of the personality variables. For example, the mean conscientiousness rating (significant other-report) was 4.06 ( SD = .60) on a five-point scale. Similarly, the mean rating for locus of control was 4.13 ( SD = .43). Given the highly educated nature of the sample (67.6% had a doctoral degree), it is perhaps not surprising that the conscientiousness ratings were so high. Additionally, highly driven, motivated individuals may be particularly likely to have an internal locus of control and perceive that they are in control of their own destiny. The high means and small standard deviations for these variables may help e xplain the null findings in the present study. Another potential explanation for the null fi ndings is the personality measure used in the present study to assess conscientiousne ss, extraversion, and neuroticism: the Big Five Inventory (John et al., 1991). This measur e was selected because of its brevity and simplicity (John & Srivastava, 1999), along with th e fact that the item s tap several facets of interest, including the positive emotionality and activity components of extraversion. However, to the researcher’s knowledge, no ot her studies have examined the relationship between personality and WFC using the BF I. Thus, the relationship between both conscientiousness and extraversion and work -family conflict may only hold for certain personality measures. It is unclear why the BFI would be less apt to yield significant relationships in these cases, as compared to other measures of the Big Five. The BFI

PAGE 106

95 items are similar to other personality scales but there are slight differences in item content; thus, the BFI may be tapping aspect s of conscientiousness and extraversion that are less related to WFC, as compared to other scales. Personality and Coping Style The hypothesized relationships between personality and coping style received mixed support, with results differing by s ource of personality ratings, by context of coping (i.e., work or family), and, to a lesser degree, depending on whether background/demographic variables were in cluded as controls. For example, conscientiousness was positively related to problem solving for work stressors (across all ratings sources of conscientiousness) and for family stressors (selfand average-ratings of personality only); unrelated to positive cognitive restructuring for work stressors and positively related to positive cognitive restruct uring for family stressors (self-ratings of personality only); negatively related to rumi nation for both work and family stressors (across all ratings sources of personality); and unrelated to escape for work stressors (across all ratings sources of personality) and negatively related to escape for family stressors (self-ratings of pers onality only). The other three pe rsonality variables exhibited similar results, in that so urce of personality ratings a nd the context of coping style affected the pattern of findings. While mixed support was found for hypothe ses linking personality and coping style, a few general trends emerged. For exam ple, self-reported personality yielded the highest number of significant relationship s, followed by the average of selfand significant other-reported personality, follo wed by partner-reported personality. This trend may reflect common method variance. For example, impression management may

PAGE 107

96 play a role in self-ratings of both persona lity and coping style, given the socially desirable (and undesirable) natu re of some of the variable s. Alternatively, individuals may be in a position to provide more accurate ratings of their own personalities, as compared to significant otherratings, and the higher correlations may reflect this. Another general trend was that some rela tionships held for one domain but not the other (e.g., significant other-ratings of consci entiousness related to problem solving for work but not for family stressors; significan t other-ratings of extraversion related to positive cognitive restructuring for family but not for work stressors). Thus, personality and situation may interact to determine whic h coping style an individual will choose. For example, individuals high on extraversion may be predisposed to use positive cognitive restructuring in stressful situations, but they may perceive that such thoughts/behaviors will be more effective for managing family stressors than for managing work stressors, thereby explaining the divergent results. The notion that coping style is, at least in part, situationally driven has been suggested by numerous researchers (e.g., Lazarus & Folkman, 1984; McCrae, 1984). In the present study, participants tended to use problem solving, instrumental support seeking, and positive cognitive restructuring more for work than for family stressors, while use of emotional support seeking, rumination, and escape were similar across domains. Thus, some coping styles may lend th emselves for use in certain situations, as they may be considered particularly appropria te and/or effective. For example, given the task-oriented nature of work demands, problem solving be haviors, such as coming up with multiple solutions to the problem and taking action to fix the problem, may be

PAGE 108

97 particularly effective. The same behaviors may be less effective when managing family demands. While the effectiveness of each coping styl e may vary across different situations, thereby affecting choice of coping style, it is also possible th at individuals simply rely on different behavioral responses within each coping style as a function of the situation. Accordingly, personality may pl ay a role in choice of copi ng style, but the relationship may manifest itself differe ntly across various types of situations. For example, extraversion was significantly related to support seeking for both work and family stressors, but the type of support seeking differed across domains. Specifically, extraversion related to instrumental support seeking for managing work stressors and to emotional support seeking for managing family stressors. Thus, personality may dictate how an individual responds to stressors at a broad level, bu t the specific behaviors he/she chooses may depend on what he/she considers mo st effective for the given situation. This speaks to the importance of tapping a wide range of behaviors within each coping style, in order to adequately measur e the full construct, as well as the importance of ensuring that our dimensionality of coping is accurate and at the appropriate level, in order to reveal such behavioral differences. A third general trend among the observed personality-coping relationships was that some coping styles were consistently re lated to personality, while others were not. For example, problem solving for work st ressors significantly related to all four personality traits across bot h sources of ratings. On the other hand, escape for work stressors was not significantly related to any of the personali ty variables, and escape for family stressors was only related to self -reported conscientiousness. Certain coping

PAGE 109

98 styles, such as problem solving, may be particularly influenced by dispositional differences, while other coping st yles, such as escape, may be more situationally driven. Alternatively, escape may be influenced by pe rsonality in some domains but not those assessed in the present study (i.e., work and family demands). Another noteworthy finding was that the mean ratings for escape were substantially lower than the mean ratings for the other coping styles. The low ra tings may reflect reluctance by participants to admit to using escape as a coping style; the fa ct that escape is, in fact, rarely used to manage these types of stressors; and/or inadeq uacies in the measurement of escape, such that the three items did not sufficiently tap the full construct. Additional research is needed to determine which reason(s) is applicable. Coping Style and Work-Family Conflict The third set of hypotheses predicted that coping styles for managing work stressors would relate to WIF and that coping styles for ma naging family stressors would relate to FIW. Contrary to expectations, none of the coping styles for family stressors related to FIW, and only rumination and escape for work stressors related to WIF. Moreover, rumination for work stressors wa s the only coping styl e that significantly related to WIF after controlling for work tim e. Although very little research has examined the relationship between work-family conflict and coping style, particularly as it was conceptualized in the present study, these re sults differ from that of Rotondo et al. (2003), which found relationships between f our types of coping (direct action, helpseeking, positive thinking, and avoidance/resignation), differentiated by work and family stressors, and WIF/FIW.

PAGE 110

99 The divergent findings between the pres ent study and that of Rotondo et al. (2003) may reflect the different coping scales used. Rotondo et al. (2003) used Havlovic and Keenan’s (1991) measure of coping, whic h was more psychometrically established than the measure used in the present study. B ecause the coping styles examined in the present study were selected based on Skinne r et al.’s (2003) review of the literature, rather than an existing taxonomy, there wa s no existing scale that included the five specific dimensions of coping examined here. Thus, a new measure was developed. Although the measure was pilot tested before us e, there is relatively little support for its validity. Additionally, because a few items were dropped and because support seeking was divided into two sub-scales, a few of th e coping styles (i.e., instrumental support seeking, emotional support seeking, and escape) were only measured with two or three items. Given the small number of items, these scales may have been deficient in assessing the intended constructs, which may have affected the study results. Another notable finding was the mixed s upport for a relationship between support seeking and WFC. Although both types of support seeking for family stressors related to WIF, they did not relate to FIW; additionall y, support seeking for work stressors was not significantly correlated with either direction of work-family conflict, though both types of support seeking for family stressors related to FIW after controlling for significant other’s employment status and the Responsibility fo r Dependents index. The inconsistent results between support seeking and WFC are surprisi ng, given that severa l studies have found a link between WFC and support in both the work (e.g., Allen, 2001; Behson, 2005; Lapierre & Allen, 2006) and family (Lapie rre & Allen, 2006) domains. The divergent results most likely reflect the difference betw een support seeking (assessed in the present

PAGE 111

100 study) and receipt of support (assessed in ot her studies) and highlight the notion that individuals who seek support may not always receive it. Coping Style as a Mediator betwee n Personality and Work-Family Conflict The final set of hypotheses predicted th at coping would mediate the relationship between personality and work-family c onflict. However, only one hypothesized relationship received support. Specificall y, based on the Sobel test, there was some evidence that rumination for work stresso rs mediated the relationship between neuroticism and WIF. On the other hand, Baron and Kenny’s (1986) third condition of mediation was not met for self-, significant ot her-, or average-ratings of personality, in that the relationship between rumination for work stressors and WIF was not significant after controlling for neuroticism. The dive rgent results between the two methods of testing for mediation may reflect the fact that the Sobel test tends to have higher power than Baron and Kenny’s (1986) method. Base d on a Monte Carlo study comparing 14 methods used to test mediation, MacK innon, Lockwood, Hoffman, West, and Sheets (2002) found that, for a sample size of 200, empi rical estimates of statistical power were .0100 for small effect sizes and .5200 for medi um effect sizes us ing Baron and Kenny’s (1986) approach. Conversely, empirical estimates of statistical power for the Sobel test were .1220 and 1.000 for small and medium effe ct sizes, respectively. Given that each test of mediation has unique advantages and di sadvantages, it is beneficial to use multiple methods to test for mediation. Supplementary Analysis For exploratory purposes, coping style was examined as a moderator between personality and work-family conflict. Six of the 96 possible interactio ns were significant,

PAGE 112

101 three of which involved conscientiousne ss. Specifically, at low levels of conscientiousness, neither problem solving for work stressors nor positive cognitive restructuring were related to FIW. On th e other hand, highly cons cientious individuals who tended to use either of these coping st yles experienced lower WIF than those who did not. One possible explanation for this findin g is that highly conscientious individuals have the skills to effectively apply these coping styles. Individuals low on conscientiousness may try to use problem solving and/or positive cognitive restructuring, but they may not be as successful in implem entation, given their le ss planful nature and the fact that they may be less apt to view stressors as a challenge (versus a threat). On the other hand, individuals low on conscientiousness may be particularly effective at escaping from their stressors, which may explain th e interaction between conscientiousness and escape for family stressors in explaining variance in FIW. For this interaction, individuals high on conscientiousne ss experienced similar levels of FIW, while at low levels of conscientiousness, hi gh use of escape was associated with higher FIW than low use of escape. While highly conscientious individuals may report using escape to cope with stressors, they may ha ve difficulty implementing this strategy, given their desire to persevere and follow through. Low-conscientious individuals may not have this problem, making them particularly adep t at escaping from stressors. Because escape may be a less effective strategy for mini mizing subsequent work-family conflict, successful implementation of escape may actua lly result in higher levels of FIW. A similar argument can be made for the fourth interaction, between extraversion and escape for family stressors, which followed a similar pattern. Just as individuals high on conscientiousness may be unsuccessful at es caping from stressors due to their driven

PAGE 113

102 nature, individuals high on extr aversion may be unsuccessful at escaping from stressors due to their desire to be surrounded by ot her people. Conversely, low-extraverts (or introverts) may be particularly effective at escaping, which results in higher levels of WIF. The fifth interaction involved neuroticism and rumination for family stressors. At high levels of neuroticism, rumination explai ned little variance in FIW, while at low levels of neuroticism, high use of ruminati on was associated with higher FIW than low use of rumination. Individuals high on neur oticism may be predisposed to experience WFC regardless of their choice of coping strategy, as evidenced by the positive, significant relationship between neuroticism and both WIF and FIW in the present study. This predisposition may reflect the fact th at neuroticism is partially defined by an individual’s ability (or inability) to handle st ress, given that neuroticism items include “Is relaxed, handles stress well” and “Remains calm in tense situa tions” (both reverse coded). On the other hand, at low levels of neuroticism, rumination may play more of a role in the experience of WF C, with high use of ruminati on resulting in more conflict. Finally, the last interaction involved locu s of control and instrumental support seeking for family stressors. For individuals high on internal locus of control, use of instrumental support seeking did not relate to WIF. On the other hand, for individuals with an external locus of cont rol, use of instrumental support seeking was associated with lower levels of WIF. Individuals with an in ternal LOC may be less reliant on the support of others for handling stressors, given their beli ef that they have control over outcomes of events. Thus, even though they may choose to use support seeking for managing stress, this coping strategy may be less influential to the WFC experience as it is for someone

PAGE 114

103 with an external LOC, who believes that out comes are caused by external circumstances. Given external LOC individuals ’ perceived lack of control for handling situations in general, and stressors in part icular, use of instrumental s upport seeking may play a larger role in their ability to manage stress than it does for individuals with an internal LOC. The observed interactions are theoreti cally and practically interesting and highlight that certain personality and coping styles may interact to explain variance in work-family conflict. However, given the explor atory nature of these analyses, as well as the potential for type one error given the large number of anal yses, these results should be interpreted with caution. Theoretical Implications Theoretically, the present study aimed to c ontribute to the l iterature by expanding the nomological network of work-family conf lict and by shedding light on the processes by which individual differences relate to the WFC experience. However, with the exception of neuroticism, no support was found fo r a relationship betw een the personality variables examined in the present study a nd either direction wo rk-family conflict. Additionally, rumination and escape for work st ressors related to WIF, but none of the other hypothesized relationships between coping and work-family conflict were significant. As previously di scussed, the conservation of re sources and spillover models would suggest that individuals are likely to experience less work-family conflict when they utilize coping strategies th at increase their resources and/or allow for the positive, rather than negative, spillover of thoughts a nd emotions from one domain to the other. Because resources, thoughts, and emotions we re not directly measured in the present study, it is unclear whether the lack of findings has any implications for these models.

PAGE 115

104 For example, while problem solving, positive cognitive restructuring, and support seeking were conceptualized as coping styles th at provide the individu al with resources to manage his/her work-family conflict, and ru mination and escape were conceptualized as coping styles that serve to deplete the individual’s res ources, the present study cannot speak to the veracity of thes e assertions. Similarly, because the present study did not assess thoughts and emotions, it is unclear whether positive cognitive restructuring and rumination affect the spillover process by leading to the tran sfer of positiv e or negative affect. On the other hand, the present study doe s have implications for the domain specificity hypothesis, which states that si tuational variables associated with a given domain relate to conflict or iginating from that domain (Frone, 2003; Frone et al., 1992). Based on this hypothesis, coping with work st ressors was expected to relate to WIF, while coping with family stressors was exp ected to relate to FIW. However, crossdomain relationships were observed in se veral cases. For example, while escape for family stressors did not relate to FIW, escape for work stressors was negatively related to FIW. Similarly, after contro lling for partner employment status and the RFD index, instrumental support seeking, emotional suppor t seeking, positive c ognitive restructuring, and escape for work stressors related to FIW; none of these coping styles related to WIF. In terms of coping with family stressors, problem solving, instru mental support seeking, and emotional support seeking related to WI F, and rumination related for WIF after controlling for work time. The reason for the observed cross-domain re lationships is unclea r. One possibility is reverse causality, such that an individual experien cing work-family conflict may

PAGE 116

105 respond by modifying his/her coping style acco rdingly. For example, if an individual’s work is interfering with his/her home life, he /she may use problem solving to proactively manage his/her home stressors in order to pr event, or minimize the extent of, future conflict. Conversely, after experiencing FIW, an individual may mentally escape from his/her work stressors, in or der to deal with the experienced conflict. Given the crosssectional, non-experimental nature of the present study’s design, there is no way to determine the direction of the relationships. Although support was only found for one of the proposed mediated relationships, that of rumination for work stressors between neuroticism and WIF, the results are still theoretically meaningf ul. Very little research has exam ined the underlying mechanisms linking personality and work-family conflict, despite the fact that several researchers have suggested potential mediati ng variables and discussed the importance of this line of research (e.g., Frone, 2003; Greenhaus et al., 2006; Wayne et al., 2004). The present study provides support for the idea that, in cert ain cases, coping style may play a role in the process by which personality and WFC ar e related, though this may be limited to certain personality-coping pair ings. Future research is necessary to examine other potential mediators of thes e relationships as well. The present study also has implications for the stress and coping literature. For example, the results provide general support for a relationship between personality and coping style, which is consistent with Lazarus and Folkman’s (1984) Cognitive Theory of Stress and Coping. Because the present study di d not assess how i ndividuals appraise stressors, we cannot ascertain whether the relationship between personality and coping can be explained directly and/or indirectly via the appraisal process. However, the results

PAGE 117

106 are consistent with this noti on, particularly when considered in conjunction with studies that did directly measure individual apprai sals of stress (e.g., Bolger & Zuckerman, 1995; Kulenovic & Buško, 2006). In terms of the coping construct in partic ular, the present stud y provides important insight. First of all, the results emphasize the importance of differentiating between instrumental and emotional support seeking, as each style exhibite d a unique pattern of relationships with other variables. Additionall y, the CFA results demonstrated superior fit for a six-factor, versus five -factor, solution. Still, an in teresting finding was that the improved fit between the fiveand six-factor solutions was more evident for coping with work stressors as compared to coping with fa mily stressors. Furthermore, the relationship between instrumental and emotional support seeking was much stronger for coping with family stressors than for c oping with work stressors ( r = .59 versus .34). This observation highlights the situational na ture of coping. The distinc tion between emotional and instrumental support seeking may be more im portant in the work domain, given the taskoriented nature of work and the fact that there are likely norms regarding whom the individual may approach for instrumental he lp and to whom he/she should turn for sympathy and understanding. On the other hand, in the family domain, this distinction may be less clear cut. Moreover, while cert ain coping styles, such as support seeking, may operate differently across domains, ot hers may be less context-dependent. For example, rumination for work stressors hi ghly correlated with rumination for family stressors ( r = .70). Overall, the stud y’s findings point to the importance of considering both dispositional and contextual factors wh en studying coping; for some coping styles, the dimensionality of the construct, as we ll as the relationship be tween coping and other

PAGE 118

107 variables, seems to vary across situations while for other coping styles, contextual factors may play a smaller role. From a methodological standpoi nt, the present study high lights the importance of collecting personality ratings from multiple so urces. While there was a substantial degree of overlap between selfand significant otherratings of personality, correlations were far from unity, and mean ratings of self-reporte d neuroticism were signi ficantly lower than mean ratings of significant other-reported ne uroticism. Additionally, in several cases, the results differed, depending on whether selfor significant other-ratings were analyzed. The discrepancies may reflect numerous factor s, including the differential interpretation of items, the consideration of different spec ific behaviors, and m odesty in self-report (McCrae et al., 1998). Other po tential explanations includ e differences between one’s self-concept and his/her observable behavior impression management motives, and the fact that some personality traits are more readily observable than others. Regardless of the reason, the present study underscores the value of having multiple raters provide personality ratings. Practical Implications The present study has practi cal implications as well. Dispositional variables explained little variance in wo rk-family conflict in the pr esent study. On the other hand, consistent with past research, the relations hips between work-family conflict and rolerelated variables were in the moderate range, with work time relating to WIF and significant other’s employment status, number of childre n living at home, and the Responsibility for Dependents index relating to FIW. From a practical standpoint, these findings highlight the role-related nature of work-family conflic t. That is not to say that

PAGE 119

108 individual difference variables are unimporta nt, but it does reinfo rce the notion that individuals (and organizations) should be mindful of the demands that they place on themselves (and on their employees) and the implications of these demands for employees’ levels of WFC. Although dispositional variables were larg ely unrelated to WFC, the relationship between neuroticism and work-family c onflict was significant. Highly neurotic individuals may be predisposed to experien ce WFC, and they may benefit from this knowledge. By understanding their predispositio n to experience WFC, high-neurotics can take active steps to mitigate future conflict. For example, the present study found some support for the hypothesis that rumination medi ates the relationship between neuroticism and WIF. Individuals high on neuroticism can be trained to avoid, or minimize, such cognitions by engaging in self-talk and ot her emotion/cognition re gulation techniques. Though highly neurotic indivi duals may particularly be nefit from such training, interventions that target certain coping styles may be helpful to other individuals as well. The present study found that WIF was positiv ely related to rumination, to escape for work stressors, and to rumination for family stressors (after controlling for work time) and negatively related to problem solving, to instrumental support seeking, and to emotional support seeking for family stressor s. Similarly, FIW was positively related to escape for work stressors and negatively re lated to instrumental support seeking, to emotional support seeking, and to positive cognitive restructuring for work stressors (after controlling for significant other’s em ployment status and the RFD index). Thus, individuals may benefit from training progr ams that elucidate wh ich coping styles are particularly effective for managing work a nd family demands and that provide tips on

PAGE 120

109 how to effectively implement certain styles while minimizing the use of other styles. Organizations could develop and implement such interventions to help employees effectively manage work and family stress ors. Given the deleterious individual (e.g., burnout, stress, depression) and organizati onal (e.g., job satisfaction, organizational commitment, turnover intentions) consequen ces of WFC (Allen et al., 2000; MesmerMagnus & Viswesvaran, 2005), such training programs have the potential for widereaching effects. Limitations The present study has several limitations, each of which has implications for the generalizability and interpretation of the study’s findings. First, participants were recruited via a snowball appro ach, rather than th rough random sampling, so the sample is not necessarily representative of the larger population. Additionally, the vast majority of participants were identified based on th eir affiliation with SIOP, a professional organization for I/O psychologists. As a resu lt, the sample was highly educated, with 67.6 percent having a doctoral degree, and of hi gh socio-economic status, with 73.6 percent reporting a household income of $100,000 or higher. Moreover, 90.2 percent of the sample was White/Caucasian, and, given that pa rticipants were larg ely recruited from a professional organization, the majority likely held professional, wh ite collar positions. The type of work stressors associated with su ch jobs are clearly very different than those experienced by a blue collar wo rker, which may have implica tions for the type of coping strategy that will be selected and effective for managing work demands. Similarly, given that participants with a high household inco me may have the means to utilize quality childcare, housecleaning services, lawn care services, etc., the type of family/home

PAGE 121

110 stressors that they experience may differ as well. Given the non-repr esentativeness of the sample, it is unclear the extent that these findings generalize to the broader population. Another limitation is the cr oss-sectional, correlationa l nature of the study design. While the present study hypothesized that person ality leads to coping style, which leads to work-family conflict, th e study design precludes any statements of causality. Additionally, retrospective, self-reports of coping style are subject to memory and reporting biases (Smith, Leffingwell, & Ptace k, 1999; Stone, Schwartz Neale, Shiffman, Marco, Hickcox, et al., 1998). Another c oncern is common method variance. Even though significant other-ratings of personal ity were collected, several of the study hypotheses relied solely on self -reported variables (i.e., relationships between coping style and work-family conflict). Common met hod bias is of particular concern for variables that are considered socially desi rable or undesirable. To the extent that participants were motivated to use impressi on management techniques for reporting both coping style and work-family conflict, the obs erved relationship between these variables may be inflated. On the other hand, as eluc idated by Spector (2006), the influence of common method bias may be smaller than is often suggested. While the majority of constructs were measured with existing, psychometrically established scales, no such measure was av ailable for coping style, based on the taxonomy used in the present study. Thus, a ne w measure was developed to assess coping style. Although most of the coping items came from existi ng scales, and a pilot study was used to guide item selection, the coping m easure used in the present study was not without problems. Specifically, a few items were removed, based on poor item-level statistics, resulting in a three-item measure fo r escape. While the alpha coefficients for

PAGE 122

111 the escape scale were acceptable (.64 for wo rk stressors and .77 for family stressors), a longer measure would have been preferable, in order to adequately assess the construct. Similarly, though the initial in tention was to use a singul ar scale to assess support seeking, confirmatory factor analysis resu lts indicated improved fit for differentiating between instrumental and emotional support seeking. The result of the breakdown was a two-item measure for emo tional support seeking and a three-item measure for instrumental support seeking. Finally, though th e CFA results revealed acceptable fit for a six-factor solution, the AGFI indices were not optimal. T hus, additional research is necessary to refine the scale and provid e more psychometric support and construct validity evidence. Future Directions There are several theoretically importa nt and practically relevant avenues available for future research. Given the inc onsistent results in the literature between several personality variables and work-fam ily conflict, research investigating the presence of moderators is warranted. As th e present study demonstrates, there is some support for the notion that certain coping styl es interact with personality to explain variance in WFC. However, these analyses we re done post hoc, for exploratory purposes. Future researchers should use a theoretically-driven approach to explore the notion that coping style moderates the relationship betw een personality and work-family conflict. Another possible moderator is situ ational factors, such as the availability of resources to manage work and family demands and/or th e level of role dema nds experienced by the individual. Individual differences such as pers onality may play less of a role in explaining variance in work-family conflict when the indi vidual is able to rely on other types of

PAGE 123

112 resources, such as quality childcare and hous ecleaning services, to help manage his/her work and family demands. Similarly, when work and/or family demands are at extreme levels, personality may be less influential to one’s level of work-fam ily conflict; in other words, individual differences in characteris tic patterns of behavior may do little to alleviate (or intensify) the felt experience of WFC when one’s objective workload is extremely high or low. On the other hand, when such resources are unavailable and/or demands are at a moderate level, the individual’s personal dispositions may play a larger role in the work-family conflict experien ce. Future research should examine this supposition, in an attempt to unde rstand the inconsistent findings in the literature between work-family conflict and personality, and pa rticularly conscientiousness, extraversion, and locus of control. In terms of future research directions involving the link between personality and coping style, it is important to further ex amine why personality and coping style are related in some situations but not others To fully understand the nature of these relationships, researchers mu st simultaneously consider individual differences and situational variables, taking an interactionist approach. Specifically, the coping literature would benefit from an enhanced understan ding of the circumstances in which coping styles do and do not generalize across situatio ns. One important variab le to consider in this line of research is situational strengt h. Personality differences are likely to play a smaller role in responding to “strong” problem situations, whereby responses may be relatively similar across individuals, wh ile the role of personality may become increasingly important in “w eaker” situations (Suls & Davi d, 1996). Future research is necessary to explor e this supposition.

PAGE 124

113 It may also be fruitful to assess person ality using a context-specific approach in order to more accurately estim ate the relationship between personality and coping with work and family stressors. Wright and Mi schel (1987) suggested that although people may demonstrate stable patterns of behavior s, such behaviors may be contingent on situational conditions, or condi tional dispositions. In support of this notion, researchers have found significant increases in the validity of personality measures, simply by adding “at work” to several of the items (Hunthausen, Truxillo, Bauer, & Hammer, 2003; Schmit, Ryan, Stierwalt, & Powell, 1995). T hus, when assessing the relationship between personality and coping style for managing wo rk and family demands, future research should examine the relationship between person ality in the work c ontext and coping with work demands, as well as the relationshi p between personality in the family/home context and coping with family demands. Su ch research may yield more significant findings than using a general measure of personality. Additional research examining the relati onship between coping and WFC is also warranted. As demonstrated by Rotondo et al (2003), the relationship between coping and work-family conflict may vary by type of conflict. For example, coping styles that rely on affective responses to stress (e.g., pos itive cognitive restructuring, rumination) may relate to strain-based conflict, while be haviors aimed at direc tly tackling the problem (e.g., problem solving, instrumental support se eking) may have a stronger association with time-based conflict. Rot ondo et al. (2003) found some suppor t for this notion, in that positive thinking for work stressors was negatively related to strain-based, but not timebased, WIF. However, more research is necessa ry to fully explore these relationships, and it may be fruitful to use a measure of WFC th at assesses behavior-based conflict as well,

PAGE 125

114 such as Carlson, Kacmar, and Williams’s ( 2000) six-dimension measure. Alternatively, future research could examine the link be tween coping and the WFC experience using Carlson and Frone’s (2003) conceptualiz ation of work-family conflict, which differentiates between external (represent ing outward behavioral interference) and internal (representing psychological preoccupati on while in the other role) conflict. The internal-external distinction aligns well with Lazarus and Folkman’s (1984) problemfocused and emotion-focused coping. Specifical ly, problem-focused c oping is likely to relate to external WFC, while emotion-focused coping is likely to relate to internal WFC. Despite the limitations of the problemve rsus emotion-focused coping taxonomy, the majority of coping research still uses thes e broad conceptualizations, and researchers have found some support for a link between coping and WFC using this distinction (Aryee et al., 1999; Lapierre & Allen, 2006, Smoot, 2005). Another viable and important future direction is to examine coping effectiveness in relation to work-family c onflict. In the present study, participants were asked how often they engage in each coping behavior and/or experience each affective/cognitive response. However, it is possible that ther e are also individual differences in how effectively each coping style is carried out. For example, while several individuals may try to work harder and more efficiently or attempt to see the problem in a positive light, some individuals may be more successful than others in implementing these behaviors/cognitions. Similarly, some individuals may be more effective than others in making a plan of action, seeking out individu als who are particular ly supportive and/or knowledgeable in the problem at hand, and/or pretending that the event has not occurred. To some extent, this issue gets at the meas urement of coping style, in that some items

PAGE 126

115 represent an attempt to act (e.g., I try to forg et the whole thing), while others are more descriptive of what the i ndividual actually experiences (e.g., I spend too much time focusing on the stressful situation). Future re search should tease apart these differences and examine how both frequency and effectiven ess of coping style re late to work-family conflict and other types of strain. Coping represents one strategy for managing work and family demands; in reality, individuals, families, and organizations often utilize multiple strategies to reduce workfamily conflict. Frone (2003) distinguished between two types of strategies: personal initiatives (e.g., coping, reducing the psyc hological importance of one role) and organizational initiatives (e.g., family-friendly organizational policies and benefits, such as flexible work arrangements and dependent care assistance). Si milarly, Lapierre and Allen (2006) described three types of conflict avoidance tactics, including those in the work domain (family supportive supervision; use of telework and flextime); those in the family domain (emotional sustenance and inst rumental assistance); and individual tactics (use of problem-focused coping). The present study focused on coping in particular, but it is important to examine a variety of strategi es simultaneously in or der to understand the full picture. With few exceptions (e.g., Lapi erre & Allen, 2006), re searchers have not examined coping strategies in conjunction wi th other types of init iatives to determine their simultaneous effect on work-family conf lict. Additional research in this area is clearly warranted. Researchers may benefit from applying a systems approach in order to fully consider how the actions of the indivi dual, family, supervisor, and coworkers, as well as organizational norms and policies, interact to affect the work-family conflict experience.

PAGE 127

116 The field would also benefit from an e xploration of the study’ s relationships, and the coping domain in particular, using alte rnative methodologies, such as qualitative, longitudinal, and experience sampling appro aches. Qualitative research would shed additional light on the particular coping styles individuals use to manage work and family demands and the extent to which they overlap and diverge from coping strategies used in other domains. Such research would help br idge the gap between coping research in general and the few studies that have deve loped coping taxonomies specific to the workfamily domain. Longitudinal and experi ence sampling methodologies would provide insight into how individuals cope with work and family stressors on a daily basis and over an extended period of time, allowing fo r a more dynamic assessment of the coping process and how coping relates to work-family conflict over time. Finally, future research should rep licate the present study with a more representative sample and a psychometrically established, well-validated coping measure. Conclusion By examining two relatively neglected t ypes of antecedents to WFC, personality and coping style, as well as the processe s underlying these relationships, the present study makes an important contribution to the work-family literature. Results support the notion that certain personality traits and coping styles relate to work-family conflict, and coping may help explain the relationship between personality and WFC for certain variables. While the present study has importa nt theoretical and pr actical implications, additional research is clearly needed to gain a better understanding of how individual differences relate to work-family conflict a nd, in particular, how individuals, families, organizations, and society work together to mitigate, or exacerbate, the WFC experience.

PAGE 128

117 References Adams, G. A., & Jex, S. M. (1999). Relatio nships between time management, control, work-family conflict, and strain. Journal of Occupational Health Psychology, 4 72-77. Allen, T. D. (2001). Family-supportive work environments: The role of organizational perceptions. Journal of Vocational Behavior, 58 414-435. Allen, T. D., Herst, D., Bruck, C. S., & Sutt on, M. (2000). Consequences associated with work-to-family conflict: A review and agenda for future research. Journal of Occupational Health Psychology, 5 278-308. Andreassi, J. K. (2007). The role of personality and coping in work-family conflict: New directions. Unpublished doctoral dissertation, Ci ty University of New York. Andreassi, J. K., & Thompson, C. A. (2007). Dispositional and situational sources of control: Relative impact on work-family conflict and positive spillover. Journal of Managerial Psychology, 22 (8), 722-740. Aryee, S., Luk, V., Leung, A., & Lo, S. (1999) Role stressors, inte rrole conflict, and well-being: The moderating influence of spousal support and coping behaviors among employed parents in Hong Kong. Journal of Vocational Behavior, 54, 259-278.

PAGE 129

118 Baltes, B. B., & Heydens-Gahir, H. A. (2003) Reduction of work-fam ily conflict through the use of selection, optimization, and compensation behaviors. Journal of Applied Psychology, 88 1005-1018. Baron, R. M., & Kenny, D. A. (1986). The m oderator-mediator variab le distinction in social psychological research: Conceptual, strategic and statistic al considerations. Journal of Personality and Social Psychology, 51 1173-1182. Behson, S. J. (2005). The rela tive contribution of formal and informal organizational work-family support. Journal of Vocational Behavior, 66 487-500. Ben-Zur, H. (1999). The effectiveness of c oping meta-strategies: Perceived efficiency, emotional correlates and cognitive performance. Personality and Individual Differences, 26, 923-939. Blanchard-Fields, F., Stein, R., & Watson, T. L. (2004). Age differences in emotionregulation strategies in handling everyday problems. The Journals of Gerontology Series B: Psychological Sc iences and Social Sciences 59 261-269. Bolger, N. (1990). Coping as a persona lity process. A prospective study. Journal of Personality and Social Psychology, 59, 525-537. Bolger, N., & Zuckerman, A. (1995). A fram ework for studying personality in the stress process. Journal of Personality and Social Psychology, 69 890-902. Bouchard, G., Guillemette, A., & Landry-Leger, N. (2004). Situational and dispositional coping: An examination of their relations hip to personality, cognitive appraisals, and psychological distress. European Journal of Personality, 18, 221-238.

PAGE 130

119 Browne, M. W., & Cudeck, R. (1993). Alternativ e ways of assessing model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 136–142). Newbury Park, CA: Sage. Bruck, C. S, & Allen, T. D. (2003). The rela tionship between big five personality traits, negative affectivity, type A behavi or, and work-family conflict. Journal of Vocational Behavior, 63 457-472. Burns, L. R., Dittman, K., Nguyen, N., & Mitchelson, J. K. (2000). Academic procrastination, perf ectionism, and control: Associ ations with vigilant and avoidant coping. Journal of Social Behavior and Personality, 15 35-46. Byron, K. (2005). A meta-analytic review of work-family conflict and its antecedents. Journal of Vocational Behavior, 67 169-198. Carlson, D. (1999). Personality and role variab les as predictors of three forms of workfamily conflict. J ournal of Vocational Behavior, 55, 236-253. Carlson, D. S., & Frone, M. R. (2003). Re lation of behavioral and psychological involvement to a new four-factor conceptu alization of work-family interference. Journal of Business and Psychology, 17 212-535. Carlson, D. S., Kacmar, K. M., & Williams, L. J. (2000). Construction and initial validation of a multidimensional measure of work family conflict. Journal of Vocational Behavior, 56 249-276. Carlson, D. S., & Perrew, P. L. (1999). The ro le of social support in the stressor-strain relationship: An examinati on of work-family conflict. Journal of Management, 25 513-540.

PAGE 131

120 Carver, C. S., & Scheier, M. F. (2000). Perspectives on personality (4th ed.). Boston: Allyn and Bacon. Carver, C. S., Scheier, M. F., & Weintra ub, J. K. (1989). Assessing coping strategies: A theoretically ba sed approach. Journal of Personality and Social Psychology, 56 (2), 267-283. Compas, B. E., Connor-Smith, J. K., Saltzman, H., Thomsen, A. H., & Wadsworth, M. E. (2001). Coping with stress during childhood and adolescence: Progress, problems, and potential in theory and research. Psychological Bulletin, 127, 87-127. Connor-Smith, J. K., Compas, B. E., & Wadsworth, M. E. (2000). Responses to stress in adolescence: Measurement of copi ng and involuntary stress responses. Journal of Consulting and Clinical Psychology, 68 (6), 976-992. Connor-Smith, J. K., & Flachsbart, C. (2007). Relations between personality and coping: A meta-analysis. Journal of Personality and Social Psychology, 93 (6), 10801107. Costa, P. T., Jr., & McCrae, R. R. (1992). NEO PI-R professional manual Odessa, FL: Psychological Assessment Resources, Inc. Eby, L. T., Casper, W. J., Lockwood, A., Bordeaux, C., & Brinley, A. (2005). Work and family research in IO/OB: Content anal ysis and review of the literature (19802002). Journal of Vocational Behavior, 66, 124-197. Endler, N. S., & Parker, J. D. A. (1990) Multidimensional assessment of coping: A critical evaluation. Journal of Personality and Social Psychology, 58 (5), 844-854. Eysenck, H. J., & Eysenck, M. W. (1985). Personality and indivi dual differences: A natural science approach. New York: Plenum Press.

PAGE 132

121 Ferguson, E. (1993). Rotter’s locus of contro l scale: A ten-item two-factor model. Psychological Reports, 73 (3), 1267-1278. Folkman, S. (1984). Personal control and st ress and coping proce sses: A theoretical analysis. Journal of Personality and Social Psychology, 46 839-852. Folkman, S., & Lazarus, R. S. (1980). An anal ysis of coping in a middle-aged community sample. Journal of Health and Social Behavior, 21, 219-239. Folkman, S., & Lazarus, R. S. (1991). Coping and emotion. In A. Monet & R. S. Lazarus (Eds.), Stress and coping (pp. 207-227). New York: Columbia University Press. Folkman, S., Lazarus, R. S., Dunkel-Schetter, C., DeLongis, A., & Gruen, R. (1986). The dynamics of a stressful encounter: Cogn itive appraisals, coping, and encounter outcomes. Journal of Personality and Social Psychology, 50, 992-1003. Ford, M. T., Heinen, B. A., & Langkamer, K. L. (2007). Work and family satisfaction and conflict: A meta-analysis of cross-domain relations. Journal of Applied Psychology, 92 57-80. Frone, M. R. (2003). Work-family balance. In J. C. Quick & L. E. Tetrick (Eds.), Handbook of Occupational Health Psychology (pp. 143-162). Washington, DC: American Psychological Association. Frone, M. R., Russell, M., & Cooper, M. L. (1992). Antecedents and outcomes of workfamily conflict: Testing a mode l of the work-family interface. Journal of Applied Psychology, 77 (1), 65-78. Goldberg, L. R. (1990). An alternative “descr iption of personality”: The big-five factor structure. Journal of Personality and Social Psychology 59 1216-1229.

PAGE 133

122 Goldberg, L. R., Johnson, J. A., Eber, H. W ., Hogan, R., Ashton, M. C., Cloninger, C. R., & Gough, H. C. (2006). The International Pers onality Item Pool and the future of public-domain personality measures. Journal of Research in Personality, 40 8496. Greenhaus, J. H., Allen, T. D., & Spector, P. E. (2006). Health consequences of work– family conflict: The dark side of the work–family interface. Research in Occupational Stress and Well Being, 5, 61–99. Greenhaus, J. H., & Beutell, N. J. (1985). S ources of conflict between work and family roles. The Academy of Management Review, 10 76-88. Grzywacz, J. G., & Marks, N. F. (2000). R econceptualizing the wo rk-family interface: An ecological perspective on the correl ates of positive and negative spillover between work and family. Journal of Occupational Health Psychology, 5 111126. Gutek, B. A., Searle, S., & Klep a, L. (1991). Rational versus gender role expectations for work-family conflict. Journal of Applied Psychology, 76 560-568. Hall, D. T. (1972). A model of coping with role conflict: The role behavior of college educated women. Administrative Science Quarterly, 17 471-486. Haren, E. G., & Mitchell, C. W. (2003). Rela tionships between the five-factor personality model and coping styles. Psychology and Education: An Interdisciplinary Journal, 40 (1), 38-49. Havlovic, S. J., & Keenan, J. P. (1991). C oping with work stress: The influence of individual differences. Journal of Social Behavior & Personality, 6 199-212.

PAGE 134

123 Heiman, T. (2004). Examination of the sa lutogenic model, support resources, coping style, and stressors among Is raeli university students. The Journal of Psychology Retrieved March 7, 2008 from, http://www.redorbit.com/news/health/ 112504/examination_of_the_salutogenic_m odel_support_resources_coping_style_and/ Hobfoll, S. E. (1989). Conservation of re sources: A new attempt at conceptualizing stress. American Psychologist, 44, 513-524. Hochschild, A. R., & Machung, A. (2003). The second shift (2nd ed.) New York: Penguin Books. Hu, L., & Bentler, P. M. (1998) Fit indices in co variance structure modeling: Sensitivity to underparametized model misspecification. Psychological Methods, 3, 424–453. Hu, L., & Bentler, P. M. (1999). Cutoff criteri a for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equations Modeling, 6 1–55. Hudek-Kneevi J., & Kardum, I., & Maglica, B. K. (2005). The sources of stress and coping styles as mediators and moderators of the relationship between personality traits and physical symptoms. Review of Psychology, 12 (2), 91-101. Hunthausen, J. M., Truxillo, D. M., Bauer, T. N., & Hammer, L. B. (2003). A field study of frame of reference effect s on personality test validity. Journal of Applied Psychology, 88, 545–551. Ireland, J. L., Brown, S. L., & Ballarini, S. (2006). Maladaptive pers onality traits, coping styles and psychological distress: A study of adult male prisoners. Personality and Individual Differences, 41 (3), 561-573.

PAGE 135

124 Jelinek, J., & Morf, M. E. (1995). Accoun ting for variance shared by measures of personality and stress-re lated variables: A canoni cal correlation analysis. Psychological Reports, 76, 959-962. Jex, S. M. (2002). Organizational psychology: A scie ntist practitioner apporach. New York: John Wiley & Sons. John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five Inventory – Versions 4a and 54. Berkeley: University of California, Berkeley, Institute of Personality and Social Research. John, O. P., & Srivastava, S. (1999). The bigfive trait taxonomy: History, measurement, and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.), Handbook of personality: Theory and research (2nd ed., pp. 102-138). New York: Guilford. Kahn, R. L., Wolfe, D. M., Quinn, R., S noek, J. D., & Rosenthal, R. A. (1964). Organizational stress. New York: John Wiley & Sons. Kato, K., & Pedersen, N. L. (2005). Personality and coping: A study of twins reared apart and twins reared together. Behavior Genetics, 35 (2), 147-158. Kenny, D. A. (2008). Mediation Retrieved December 28, 2008, from http://davidakenny.net/cm/mediate.htm Kinnunen, U., & Mauno, S. (1998). Antecedents and outcomes of work-family conflict among employed women and men in Finland. Human Relations, 51 (2), 157-177. Kinnunen, U., Vermulst, A., Gerris, J., & M kikangas, A. (2003). Work–family conflict and its relations to well-being: The role of personality as a moderating factor. Personality and Individual Differences, 35 (7), 1669-1683.

PAGE 136

125 Kirchmeyer, C. (1993). Nonwork-to-work spil lover: A more balanced view of the experiences and coping of professional women and men. Sex Roles, 28 531-552. Kline, R. B. (1998). Principles and practices of structural equation modeling. New York: Guilford. Kulenovic, A., & Buško, V. (2006). Structur al equation analyses of personality, appraisals, and coping relationships. Review of Psychology, 13 (2), 103-112. Lapierre, L. M., & Allen, T. D. ( 2006) Work-supportive family, family-supportive supervision, use of organizational be nefits, and problem-focused coping: Implications for work-family conflict and employee well-being. Journal of Occupational Health Psychology, 11 (2), 169-181. Lazarus, R. S. (1996). The role of coping in the emotions and how coping changes over the life course. In C. Maletesta-Magni & S. H. McFadden (Eds.), Handbook of emotion, adult development, and aging (pp. 289-306). New York: Academic Press. Lazarus, R. S., & Folkman, S. (1982). Coping and adaptation. In W. D. Gentry (Ed.), The handbook of behavioural medicine. New York: Guilford. Lazarus, R. S., & Folkman, S. (1984). Stress, Appraisal and Coping New York: Springer. Lazarus, R. S., & Launier, R. (1978). Stress -related transactions between person and environment. In L. A. Pervin & M. Lewis (Eds), Perspectives in interactional psychology. New York: Plenum. Levenson, H. I. (1974). Activism and powerful others: Distinctions within the concept internal-external control. Journal of Personality Assessment, 38 377-385.

PAGE 137

126 Levenson, H. (1981). Differentiating among inte rnality, powerful others, and chance. In H. M. Lefcourt (Ed.), Research with the locus of control construct (Vol. I): Assessment methods (pp. 15-63). New York: Academic Press. Lu, L., & Chen, C. S. (1996). Correlates of coping behaviours: Internal and external resources. Counseling Psychology Quarterly, 9 (3), 297-307. MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002). A comparison of methods to test mediation and other intervening variable effects. Psychological Methods, 7 (1), 83-104. Matsui., T., Ohsawa, T., & Onglatco, M. ( 1995). Work-family conf lict and the stressbuffering effects of husband support a nd coping behavior among Japanese married working women. Journal of Vocational Behavior, 47, 178-192. McCrae, R. R. (1984). Situational determinan ts of coping responses: Loss, threat, and challenge. Journal of Personality and Social Psychology, 46, 919–928. McCrae, R. R., & John, O. P. (1992). An intr oduction to the five-factor model and its applications. Journal of Personality, 60, 175-215. McCrae, R. R., Stone, S. V., Fagan, P. J., & Costa, P. T., Jr. (1998). Identifying causes of disagreement between self-reports and spouse ratings of personality. Journal of Personality, 66 (3), 285-313. Mesmer-Magnus, J. R., & Viswesvaran, C. ( 2005). Convergence be tween measures of work-to-family and family-to-work c onflict: A meta-analytic examination. Journal of Vocational Behavior, 67 215-232.

PAGE 138

127 Mikulincer, M., & Victor, F. (1995). Appraisal of and copi ng with real-life stressful situation: The contribution of attachment styles. Personality and Social Psychology Bulletin, 21 (4), 406-415. Mut n, E. (1991). Self-reports, spouse ratings, and psychophysiological assessment in a behavioral medicine program: An a pplication of the five-factor model. Journal of personality assessment, 57 (3), 449-464. Netemeyer, R. G., Boles, J. S., & McMurri an, R. (1996). Development and validation of work-family conflict and family-work conflict scales. Journal of Applied Psychology, 81 (4), 400-410. Noor, N. M. (2002). Work–family conflict, lo cus of control, and women’s well-being: Tests of alternative pathways. The Journal of Social Psychology, 142 (5), 645– 662. Parkes, K. R. (1984). Locus of control, c ognitive appraisal, and coping in stressful situations. Journal of Personality and Social Psychology, 46, 655-668. Parkes, K. R., & Razavi, T. D. B. (2004) Personality and attitudinal variables as predictors of voluntary union membership. Personality and Individual Differences, 37 333-347. Paulhus, D. L. (1983). Sphere-specifi c measures of perceived control. Journal of Personality and Social Psychology, 44, 1253–1265. Paulhus, D. L. & Christie, R. (1981). Spheres of control: An interac tionist approach to assessment of perceived control. In H. M. Lefcourt (Ed.), Research with the locus of control construct (Vol. I.): Assessment methods (pp. 161-188). New York: Academic Press.

PAGE 139

128 Paulhus, D. L., & Van Selst, M. (1990). Th e Spheres of Control Scale: 10 years of research. Personality and Individual Differences, 11 (10), 1029-1036. Petrosky, M. J., & Birkimer, J. C. (1991). The relationship among locus of control, coping styles, and psycho logical symptom reporting. Journal of Clinical Psychology, 47, 336-345. Preacher, K. J., & Hayes, A. F. (2004). SPSS a nd SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36 (4), 717-731. Roehling, P., Moen, P., & Batt, R. (2003). Spillover. In P. Moen (Ed.), It’s about time (pp. 101-121). Ithaca, NY: Corn ell University Press. Roth, S., & Cohen, L. (1986). Approach, avoidance, and coping with stress. American Psychologist, 41 813-819. Rothausen, T. J. (1999). ‘Family’ in organiza tional research: A revi ew and comparison of definitions and measures. Journal of Organizational Behavior, 20 (6), 817-836. Rothbaum, F., Weisz, J. R., & Snyder, S. S. (1982). Changing the world and changing the self: A two process model of perceived control. Journal of Personality and Social Psychology, 42, 5-37. Rotondo, D. M., Carlson, D. S., & Kincaid, J. F. (2003). Coping with multiple dimensions of work-family conflict. Personnel Review, 32 275-296. Rotter, J. B. (1966). General expectancies for internal versus external control of reinforcement. Psychological Monographs General and Applied, 80 1-28.

PAGE 140

129 Rotter, J. B. (1975). Some problems and mi sconceptions related to the construct of internal versus external control of reinforcement. Journal of Consulting and Clinical Psychology, 43, 56-67. Saffrey, C., & Ehrenberg, M. (2007). When thinking hurts: Attachment, rumination, and postrelationship adjustment. Personal Relationships, 14, 351-368. Schan, R., & Abelson, R. (1977). Scripts, plans, goals, and understanding: An inquiry into human knowledge structures. Hillsdale, NJ: Lawrence Erlbaum. Schmit, M. J., Ryan, A. M., Stierwalt, S. L., & Powell, A. B. (1995). Frame-of-reference effects in personality scale scores and criterion-related validity. Journal of Applied Psychology, 80, 607–620. Skinner, E. A., Edge, K., Altman, J., & Sh erwood, H. (2003). Searching for the structure of coping: A review and critique of ca tegory systems for classifying ways of coping. Psychological Bulletin, 129 (2), 216-269. Smith, R. E., Leffingwell, T. R., & Ptacek, J. T. (1999). Can people remember how they coped? Factors associated with discorda nce between same-day and retrospective reports. Journal of Personality and Social Psychology, 76 1050–1061. Smoot, S. M. (2005). The mediational role of copi ng in the relationship between personality and work-family conflict Unpublished doctoral dissertation. Auburn University. Sobel, M. E. (1982). Asymptotic confidence in tervals for indirect e ffects in structural equation models. In S. Leinhart (Ed.), Sociological methodology 1982 (pp. 290312). San Francisco: Jossey-Bass.

PAGE 141

130 Somech, A., & Drach-Zahavy, A. (2007). Strategies for coping with work-family conflict: The distinctive relations hips of gender role ideology. Journal of Occupational Health Psychology, 12 (1), 1-19. Spector, P. E. (2006). Method va riance in organizational resear ch: Truth or urban legend? Organizational Research Methods 9 221-232. Stone, A. A., Schwartz, J. E., Neale, J. M ., Shiffman, S., Marco, C. A., Hickcox, M., et al. (1998). A comparison of coping asse ssed by ecological momentary assessment and retrospective recall. Journal of Personality and Social Psychology, 74, 1670– 1680. Suls, J., & David, J. P. (1996). Coping and personality: Third time’s the charm? Journal of Personality, 64 (4), 993-1005, Thompson, C. A., Poelmans, S. A. Y., Allen, T. D., & Andreassi, J. K. (2007). On the importance of coping: A model and new directions for research on work and family. Research in Occupational Stress and Well Being, 6 73-113. Vollrath, M. (2001). Personality and stress. Scandanavian Journal of Psychology, 42 335-347. Vollrath, M., Banholzer, E., Caviezel, C., Fisc hli, C., & Jungo, D. (1994). Coping as a mediator or moderator of personality in me ntal health? In B. D. Raad, W. K. B. Hofstee, & G. L. M. Van Heck (Eds.), Personality psychology in Europe (Vol. 5, pp. 262-273). Tilburg: Tilburg University Press. Vollrath, M., Torgersen, S., & Alns, R. (1995) Personality as long-term predictor of coping. Personality and Individua l Differences, 18, 117-125.

PAGE 142

131 Watson, D., & Hubbard, B. (1996). Adaptational style and dispositional structure: Coping in the context of the five-factor model. Journal of Personality, 64, 737-774. Watson, D., & Tellegen, A. (1985). Toward a consensual structure of mood. Psychological Bulletin, 98, 219-235. Wayne, J. H., Musisca, N., & Fleeson, W. (2004) Considering the role of personality in the work-family experience: Relationships of the big five to work-family conflict and facilitation. Journal of Vocational Behavior, 64 108-130. Williams, K. J., & Alliger, G. M. (1994). Role stressors, mood spillover, and perceptions of work-family conflict in employed parents. The Academy of Management Journal, 37 (4), 837-868. Wright, J. C., & Mischel, W. (1987). A conditional approach to dispositional constructs: The local predictability of social behavior. Journal of Personality and Social Psychology, 53, 1159-1177.

PAGE 143

132 Appendices

PAGE 144

133 Appendix A: Hypothesized Relationships Figure 7 Hypothesized relationships between personality variables, coping styles, and work-family conflict. Problem Solving (direct action, planning, decision-making) Escape (avoidance, disengagement, denial) WIF/FIW Internal Locus of Control Extraversion Conscientiousness Neuroticism Support Seeking (comfort seeking, help seeking) Rumination (intrusive/negative thoughts, self-blame, worry) Cognitive Restructuring (positive thinking, self-encouragement) Coping Styles that Increase the Individual’s Resources Coping Styles that Do Not Increase the Individual’s Resources

PAGE 145

134 Appendix B: Big Five Personality Scale Items (Conscientiousness, Extraversion, an d Neuroticism; John et al., 1991) Directions : Here are a number of characteristics that may or may not apply to you [your spouse/significant other ]. For example, do you agree that you are someone [your spouse/significant other is someone] who likes to spend time with others? For each item below, indicate the extent to which you agree or disagree with that statement. Conscientiousness Items 1. Does a thorough job 2. Can be somewhat careless (R) 3. Is a reliable worker 4. Tends to be disorganized (R) 5. Tends to be lazy (R) 6. Perseveres until the task is finished 7. Does things efficiently 8. Makes plans and follows through 9. Is easily distracted (R) Extraversion Items 1. Is talkative 2. Is reserved (R) 3. Is full of energy

PAGE 146

135 Appendix B: (Continued) 4. Generates a lot of enthusiasm 5. Tends to be quiet (R) 6. Has an assertive personality 7. Is sometimes shy, inhibited (R) 8. Is outgoing, sociable Neuroticism Items 1. Is depressed, blue 2. Is relaxed, handles stress well (R) 3. Can be tense 4. Worries a lot 5. Is emotionally stable, not easily upset (R) 6. Can be moody 7. Remains calm in tense situations (R) 8. Gets nervous easily

PAGE 147

136 Appendix C: Locus of Control Scale Items (Paulhus & Van Selst, 1990) Directions : For each statement belo w, indicate how accurately or inaccurately each statement describes you. 1. I can usually achieve what I wa nt when I work hard for it. 2. Once I make plans I am almost certain to make them work. 3. I prefer games involving some lu ck over games of pure skill. (R) 4. I can learn almost anything if I set my mind to it. 5. My major accomplishments are entirely due to my hard work and ability. 6. I usually do not set goals because I have a hard time following through on them. (R) 7. Bad luck has sometimes prevente d me from achieving things. (R) 8. Almost anything is possible for me if I really want it. 9. Most of what will happen in my career is beyond my control. (R) 10. I find it pointless to keep wo rking on something that is too difficult for me. (R)

PAGE 148

137 Appendix D: Coping Scale Items (Adapted from Carver et al., 1989; Connor-Sm ith et al., 2000; Folkman & Lazarus, 1980; Folkman et al, 1986; Havlovic & Keen an, 1991; and Saffrey & Ehrenberg, 2007) Directions : In this section, think of stressful situations you have faced at work [home ]. Using the scale below, indicate how often you react in each of the following ways in response to such situations. Problem Solving 1. I come up with a couple of diffe rent solutions to the problem. 2. I make a plan of action. 3. I concentrate my efforts on doing something about it. 4. I try to work harder and more efficiently.1 5. I do something to try to fix the problem or take action to change things. Support Seeking 1. I talk to someone to find out more about the situation. 2. I talk to someone who could do some thing concrete about the problem. 3. I ask people who have had simila r experiences what they did. 4. I talk to someone about how I feel. 5. I get sympathy, understanding, or support from someone.

PAGE 149

138 Appendix D: (Continued) Positive Cognitive Restructuring 1. I try to see it in a different lig ht, to make it seem more positive. 2. I think of ways to use this si tuation to show what I can do. 3. I remind myself that other people have been in this situation and that I can probably do as well as they did. 4. I think about the challenges I can find in this situation. 5. I tell myself that everything will be alright. Rumination 1. I get so caught up with thinking about the stressful event that it’s hard to focus on anything else. 2. I dwell on my feelings following the stressful event. 3. I spend too much time focusi ng on the stressful situation. 4. I think about every single detail of the event over and over again. 5. Thoughts about the stressful situ ation just pop in to my head. Escape 1. I hope a miracle will happen.1 2. I go on as if nothing happened. 3. I try to forget the whole thing. 4. I pretend that it hasn't really happened. 1Item not used in analysis.

PAGE 150

139 Appendix E: Work-Family Conflict Scale Items (Netemeyer et al., 1996) Directions : This section asks questions about balancing work and family demands Please read the following items and indicate how often you experience each statement. WIF Items 1. The demands of my work interfere with my home and family life. 2. The amount of time my job takes up ma kes it difficult to fulfill family responsibilities. 3. Things I want to do at home do not get done because of demands my job puts on me. 4. My job produces strain that makes it difficult to fulfill family duties. 5. Due to work-related duties, I have to make changes to my plans for family activities. FIW Items 1. The demands of my family or spouse/partne r interfere with work -related activities. 2. I have to put off doing things at work because of demands on my time at home. 3. Things I want to do at work don't get done because of the demands of my family or spouse/partner. 4. My home life interferes with my responsibilities at work such as getting to work on time, accomplishing daily tasks, and working overtime. 5. Family-related strain interferes with my ability to perform job-related duties.

PAGE 151

140 Appendix F: Hypothesis Testing with Self-Reported Personalit y and with the Average of Selfand Significant Other-Reported Personality Table 32. Regression of WIF and FIW on Conscientiousness (Self-Report) Dependent Variable WIF FIW Independent Variable Conscientiousness -.13 -.09 Control Variables Work Time .38** SO1 Employment .23** RFD2 .37** in R2 .15 .16 Overall R2 .16 .17 Adjusted R2 .15 .15 Overall F 18.41** 13.19** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index Table 33. Regression of WIF and FIW on Cons cientiousness (Average of Selfand Significant Other-Report) Dependent Variable WIF FIW Independent Variable Conscientiousness -.10 -.04 Control Variables Work Time .38** SO1 Employment .23** RFD2 .37** in R2 .15 .16 Overall R2 .15 .16 Adjusted R2 .14 .15 Overall F 17.51** 12.66** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index

PAGE 152

141 Appendix F: (Continued) Table 34. Regression of WIF and FIW on Extraversion (Self-Report) Dependent Variable WIF FIW Independent Variable Extraversion -.07 -.04 Control Variables Work Time .37** SO1 Employment .23** RFD2 .37** in R2 .13 .16 Overall R2 .14 .16 Adjusted R2 .14 .15 Overall F 16.86** 12.66** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index Table 35. Regression of WIF and FIW on Extrav ersion (Average of Se lfand Significant Other-Report) Dependent Variable WIF FIW Independent Variable Extraversion -.06 -.07 Control Variables Work Time .37** SO1 Employment .22** RFD2 .37** in R2 .13 .16 Overall R2 .14 .16 Adjusted R2 .13 .15 Overall F 16.73** 12.95** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index

PAGE 153

142 Appendix F: (Continued) Table 36. Regression of WIF and FIW on Neuroticism (Self-Report) Dependent Variable WIF FIW Independent Variable Neuroticism .22** .16* Control Variables Work Time .37** SO1 Employment .20** RFD2 .37** in R2 .14 .15 Overall R2 .19 .18 Adjusted R2 .18 .17 Overall F 23.36** 14.86** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index Table 37. Regression of WIF and FIW on Neuro ticism (Average of Selfand Significant Other-Report) Dependent Variable WIF FIW Independent Variable Neuroticism .27** .19** Control Variables Work Time .38** SO1 Employment .19** RFD2 .36** in R2 .14 .14 Overall R2 .21 .19 Adjusted R2 .20 .18 Overall F 26.70** 15.72** p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Significant Other; 2Responsibility for Dependents Index

PAGE 154

143 Appendix F: (Continued) Table 38. Regression of Coping Styl e on Conscientiousness (Self-Report) Dependent Variable Rumination (work) Rumination (family) Independent Variable Conscientiousness -.29** -.23** Control Variable Gender -.18** -.20** in R2 .03 .04 Overall R2 .10 .08 Adjusted R2 .09 .07 Overall F 11.36** 8.91** p <.05; ** p <.01; s are standardized regression we ights from the final equation Table 39. Regression of Coping Style on C onscientiousness (Average of Selfand Significant Other-Report) Dependent Variable Rumination (work) Rumination (family) Independent Variable Conscientiousness -.27** -.24** Control Variable Gender -.17* -.19** in R2 .03 .04 Overall R2 .09 .09 Adjusted R2 .08 .08 Overall F 10.10** 9.34** p <.05; ** p <.01; s are standardized regression we ights from the final equation

PAGE 155

144 Appendix F: (Continued) Table 40. Regression of Coping Styl e on Extraversion (Self-Report) p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Instrumental Support Seeking; 2Emotional Support Seeking Table 41. Regression of Coping Style on Extrav ersion (Average of Selfand Significant Other-Report) p <.05; ** p <.01; s are standardized regression we ights from the final equation 1Instrumental Support Seeking; 2Emotional Support Seeking Dependent Variable Support Seeking – I1 (w) Support Seeking – E2 (w) Support Seeking – I1 (f) Support Seeking – E2 (f) Rumination (w) Rumination (f) Independent Variable Extraversion .21** .10 .19** .16* -.15* -.09 Control Variable Gender -.18* -.44** -.24** -.37** -.17* -.18* in R2 .03 .19 .06 .13 .03 .03 Overall R2 .09 .22 .11 .18 .05 .04 Adjusted R2 .08 .21 .10 .17 .04 .03 Overall F 9.59** 27.41** 11.92** 21.55** 4.66* 3.79* Dependent Variable Support Seeking – I1 (w) Support Seeking – E2 (w) Support Seeking – I1 (f) Support Seeking – E2 (f) Rumination (w) Rumination (f) Independent Variable Extraversion .20** .09 .16* .15* -.14* -.11 Control Variable Gender -.18** -.44** -.25** -.37** -.17* -.19** in R2 .03 .19 .06 .13 .03 .03 Overall R2 .08 .21 .10 .18 .04 .04 Adjusted R2 .08 .21 .09 .17 .03 .03 Overall F 9.18** 27.07** 10.61** 21.39** 4.41* 4.17*

PAGE 156

145 Appendix F: (Continued) Table 42. Regression of Coping St yle on Neuroticism (Self-Report) Dependent Variable Rumination (work) Rumination (family) Independent Variable Neuroticism .57** .48** Control Variable Gender -.06 -.10 in R2 .00 .01 Overall R2 .34 .25 Adjusted R2 .34 .24 Overall F 51.90** 33.54** p <.05; ** p <.01; s are standardized regression we ights from the final equation Table 43. Regression of Coping Style on Neurot icism (Average of Selfand Significant Other-Report) Dependent Variable Rumination (work) Rumination (family) Independent Variable Neuroticism .53** .43** Control Variable Gender -.03 -.08 in R2 .00 .01 Overall R2 .29 .20 Adjusted R2 .29 .20 Overall F 41.35** 25.65** p <.05; ** p <.01; s are standardized regression we ights from the final equation

PAGE 157

146 Appendix F: (Continued) Table 44. Mediated Regression of WIF on Rumination for Work Stressors and Neuroticism (Self-Report) Dependent Variable: WIF Step 1 Step 2 Step 3 Control Variable Work Time .37** .37** .37** Independent Variable Neuroticism .22** .15 Mediator Rumination (work) .12 in R2 -.05 .01 Overall R2 .14 .19 .20 Adjusted R2 .14 .18 .19 Overall F 32.60** 23.36** 16.49** p <.05; ** p <.01 Table 45. Mediated Regression of WIF on Rumination for Work Stressors and Neuroticism (Average of Self and Significant Other-Report) Dependent Variable: WIF Step 1 Step 2 Step 3 Control Variable Work Time .37** .38** .38** Independent Variable Neuroticism .27** .22** Mediator Rumination (work) .09 in R2 -.07 .01 Overall R2 .14 .21 .22 Adjusted R2 .14 .20 .20 Overall F 32.60** 26.70** 18.37** p <.05; ** p <.01

PAGE 158

About the Author Rebecca H. Bryant received a Bachelor’s Degree in Psychology from Emory University in 2002, with a minor in Sociology. She entered the Ph.D. program at the University of South Florida in 2004 and rece ived her Master’s De gree in Industrial and Organizational Psychology in 2006. While enrolled at the University of South Florida, Ms. Bryant conducted research in various topics, includi ng managing work and family roles, organizational citizenship behavior, a nd work team effectiveness. She also held positions at both Personnel Decisions Research Institutes, Inc. and Bank of America, where she gained experience in such areas as project management, developing and validating selection tools, and implementi ng and evaluating interventions to reduce attrition. Ms. Bryant has co-a uthored two book chapters and numerous technical reports. She has also presented at several profe ssional conferences, including the Society for Industrial and Organizationa l Psychology and American Psychological Society.