USF Libraries
USF Digital Collections

The provision of spousal support

MISSING IMAGE

Material Information

Title:
The provision of spousal support antecedents, consequences, and crossover effects
Physical Description:
Book
Language:
English
Creator:
Dorio, Jay M
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Work-family conflict
Work interference with family
Social support
Spouse
Receipt of social support
Dissertations, Academic -- Psychology -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: The present study had four main objectives. First, the relationship between the provision of spousal support and its theoretical antecedents and consequences was assessed as informed by the conservation of resources theory (Hobfoll, 1989). Second, the crossover of physicians' work interference with family conflict on their spouses' family demands (perceived family demands and family hours) was investigated. Next, the mediating role of social support as an explanation for the crossover process was examined using two distinct pathways. Lastly, the fourth objective of the present study was to investigate the relationships described above across multiple time points and using dual-source data (from physicians and their spouses). The final sample included matched responses from 126 couples across two time points. Results were generally supportive of the relationship between the provision of spousal support and the receipt of spousal support, perceived family demands, family hours, and work interference with family conflict (WIF) and were consistent with expected relationships according to COR theory. Results also provided support for the synchronous crossover of WIF on perceived family demands; however, results were generally unsupportive of the mediating role of the provision of spousal support in the crossover process. The present study makes several important contributions to the social support, work-family conflict, and crossover literatures by adding to the knowledge of the antecedents and consequences of the provision of spousal support, the growing body of research examining the crossover of WIF, and the understanding of the mediating role of the provision of spousal support in the crossover process. Major findings and areas of opportunity for future research 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 Jay M. Dorio.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 145 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 - 002028740
oclc - 436298510
usfldc doi - E14-SFE0002809
usfldc handle - e14.2809
System ID:
SFS0027126:00001


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 002028740
005 20090911155757.0
007 cr mnu|||uuuuu
008 090911s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002809
035
(OCoLC)436298510
040
FHM
c FHM
049
FHMM
090
BF121 (Online)
1 100
Dorio, Jay M.
4 245
The provision of spousal support :
b antecedents, consequences, and crossover effects
h [electronic resource] /
by Jay M. Dorio.
260
[Tampa, Fla] :
University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains 145 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: The present study had four main objectives. First, the relationship between the provision of spousal support and its theoretical antecedents and consequences was assessed as informed by the conservation of resources theory (Hobfoll, 1989). Second, the crossover of physicians' work interference with family conflict on their spouses' family demands (perceived family demands and family hours) was investigated. Next, the mediating role of social support as an explanation for the crossover process was examined using two distinct pathways. Lastly, the fourth objective of the present study was to investigate the relationships described above across multiple time points and using dual-source data (from physicians and their spouses). The final sample included matched responses from 126 couples across two time points. Results were generally supportive of the relationship between the provision of spousal support and the receipt of spousal support, perceived family demands, family hours, and work interference with family conflict (WIF) and were consistent with expected relationships according to COR theory. Results also provided support for the synchronous crossover of WIF on perceived family demands; however, results were generally unsupportive of the mediating role of the provision of spousal support in the crossover process. The present study makes several important contributions to the social support, work-family conflict, and crossover literatures by adding to the knowledge of the antecedents and consequences of the provision of spousal support, the growing body of research examining the crossover of WIF, and the understanding of the mediating role of the provision of spousal support in the crossover process. Major findings and areas of opportunity for future research 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
Work-family conflict
Work interference with family
Social support
Spouse
Receipt of social support
0 690
Dissertations, Academic
z USF
x Psychology
Doctoral.
773
t USF Electronic Theses and Dissertations.
856
u http://digital.lib.usf.edu/?e14.2809



PAGE 1

The Provision of Spousal Support: Antecedents, Consequences, and Crossover Effects by Jay M. Dorio A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology College of A rts and Sciences University of South Florida Major Professor: Tammy D. Allen, Ph.D. Walter C. Borman Ph.D. Peter J. Fabri, Ph.D. Vicky Phares Ph.D. Paul E. Spector Ph.D. Date of Approval: December 1 2008 Keywords: work family conflict, work i nterference with family, social support, spouse, receipt of social support, emotional concern, instrumental support, high stress occupation Copyright 200 9 Jay M. Dorio

PAGE 2

Dedication To Erika and Lily : I can never say thank you enough I lov e you.

PAGE 3

Acknowledgements On this journey there have been many who have helped me by either standing on the sidelines and cheering enthusiastically (especially on the uphill stretches) by running behind me and kicking me in the butt when needed or b y running along with me on the way Without you, this process would have likely been harder, longer, and nowhere near as much fun. Thank you to all of you. Thank you to my trainer, my advisor, my mentor, my role model, Tammy Allen. You are truly an outstan ding everything. I am honored to be on the list of your graduate students Thank you to my doctoral committee members : Wally Borman, Paul Spector, Vicky Phares, and Peter J. Fabri Your insightful comments and suggestions were critical to the success of th is project. Thank you to my fellow runners both at the University of South Florida and at Kenexa. I am truly honored to have share d this journey with you y our continuing ingenuity, drive, and passion for our field is truly inspirational. A special thank you to Dr. Adam Bandelli. You are truly an inspiration. Your persistence, especially on the steep uphill portion of this journey, and your unrelenting focus is remarkable. I am honored to call you a friend. Thank you. Finally, t hank you t o my family, both near and far. Thank you for your constant cheering and occasional butt kicking. I love you all.

PAGE 4

i Table of Contents List of Tables ................................ ................................ ................................ ..................... iv List of Figures ................................ ................................ ................................ .................... vi A bstract ................................ ................................ ................................ ............................. vii Chapter One : Introduction ................................ ................................ ................................ ... 1 Social Support ................................ ................................ ................................ .......... 3 Dimensions of Social Support ................................ ................................ ...... 3 Sources of Social Support ................................ ................................ ............. 5 Receiving Spousal Support ................................ ................................ ........... 6 Receiving Spousal Support versus Providing Spousal Support ................. 10 Providing Social Support ................................ ................................ ............ 10 Family Demands ................................ ................................ ................................ .... 14 Providing Spousal Support and Family Demands ................................ ...... 15 Work Family Conflict ................................ ................................ ............................ 16 WFC and the Receipt of Spousal Support ................................ .................. 17 WIF and the Provi sion of Spousal Support ................................ ................ 18 Crossover Effects ................................ ................................ ................................ ... 20 Research Examining Crossover Effects ................................ ..................... 21 Crossover and Work Family Conflict ................................ ........................ 22 Mechanisms of Crossover ................................ ................................ .......... 24 Summary ................................ ................................ ................................ ..... 28 Chapter Two : M ethod ................................ ................................ ................................ ........ 29 Participants ................................ ................................ ................................ ........ 29 Response rate ................................ ................................ .............................. 33 Study Design ................................ ................................ ................................ ........ 37 Measures ................................ ................................ ................................ ........ 38 Provision of Spousal Support ................................ ................................ ..... 38 Receipt of Spousal Support ................................ ................................ ........ 41 Perceived Family Demands ................................ ................................ ........ 43 Amount o f Time Engaged in Family Related Activities ............................ 43

PAGE 5

ii Work Interference with Family Conflict ................................ .................... 44 Work Interference with Family Conflict: Spouse Report .......................... 44 Perceived Work Demands ................................ ................................ .......... 45 Amount of Time Engaged in Work Related Activities .............................. 45 Demographic Variables ................................ ................................ .............. 45 Procedure ................................ ................................ ................................ ........ 46 Chapter Three : Analys es and Results ................................ ................................ ................ 49 Preliminary Data Steps ................................ ................................ .......................... 49 Descriptive Statistics for Study Variables ................................ ............................. 52 Inter Correlations Among Variables ................................ ................................ ...... 53 Hypotheses 1, 2, and 3 ................................ ................................ ........................... 58 Hypotheses 4a and 4b ................................ ................................ ............................ 59 Hypotheses 5 and 6 ................................ ................................ ................................ 61 Chapter Four : S upplemental Analyses and Results ................................ ........................... 78 Providing and Receiving Spousal Support ................................ ............................ 79 Self Report and Spouse ................................ ................ 80 Suppl emental Analyses: Hypotheses 1, 2, and 3 ................................ ................... 81 Supplemental Analyses: Hypotheses 4a and 4b ................................ ..................... 83 Supplemental Analyses: Hypotheses 5 a nd 6 ................................ ........................ 84 Chapter Five : D iscussion ................................ ................................ ................................ ... 94 Major Findings: Antecedents and Consequences of Spousal S upport ................... 95 Major Findings: Crossover Relationships ................................ ............................ 100 ....................... 102 ........................... 104 Implications and Future Research ................................ ................................ ........ 107 Study Strengths ................................ ................................ ................................ .... 109 Study Limitations ................................ ................................ ................................ 111 Conclusion ................................ ................................ ................................ ...... 113 References ................................ ................................ ................................ ........................ 114 Appendices ................................ ................................ ................................ ....................... 131 Appendix A: Provision of Spousal Support Items ................................ ............... 132 Appendix B: Receipt of Spousal Support Items ................................ .................. 133 Appendix C: Family Demands Items ................................ ................................ ... 134 A ppendix D: Work Interference with Family Conflict Items .............................. 135

PAGE 6

iii Appendix E: Spouse Report Work Interference with Family Conflict Items ...... 136 Appendix F: Work Demands Items ................................ ................................ ..... 137 Appendix G: Demographic Variables ................................ ................................ .. 138 Appendix G: (Continued) ................................ ................................ .................... 139 Appendix H: Medical Resident Invitation Email (Time 1) ................................ 140 Appendix I: Physician Invitation Email (Time 1) ................................ ................ 141 Appendix J: Spouse/Partner Invitation Email (Time 1) ................................ ....... 142 Appendix K: Spouse/Partner Reminder Email (Time 1) ................................ ..... 143 Appendix L: Physician and Spouse/Partner Invitation Email (Time 2) .............. 144 Appendix M: Physician and Spouse/Partner Reminder Email (Time 2) ............. 145 About the Author ................................ ................................ ................................ ... End Page

PAGE 7

iv List of Tables Table 1 Demographic Characteristics of Participants (Time 1) .............................. 31 Table 2 Estimated Number of Eligible Physicians Invited to Participate ............... 34 Table 3 Physician and Spouse Response Rates (Times 1 and 2) ............................ 36 Table 4 Physician Response Rates by Recruitment Source (Time 1) ..................... 36 Table 5 Fit Statistics from the Confirmatory Factor Analyses for the Provision of Support Scale ................................ ............................. 41 Table 6 Factor Loadings from the Exploratory Factor Analysis on the Receipt of Support Scale ................................ ................................ 42 Table 7 Fit Statistics from the Confirmatory Factor Analyses for the Receipt of Support Scale ................................ ................................ 43 Table 8 Descriptive Statistics for Study Variables ................................ ................. 52 Table 9 Inter Correlations among Physician Variables ................................ .......... 54 Table 10 I nter Correlations among Spouse Variables ................................ .............. 55 Table 11 Inter Correlations among Physician and Spouse Variables ....................... 56 Table 12 with Physician Report WIF ................................ ................................ ....... 68 Table 13 sal Support with Spouse report WIF ................................ ................................ ............. 69 Table 14 with Physician Report WIF (Time 1) ................................ ........................ 70 Table 15 with Physician Report WIF (Time 2) ................................ ........................ 71

PAGE 8

v Table 16 with Spouse Report WIF (Time 1) ................................ ............................ 72 Table 17 with Spouse Report WIF (Time 2) ................................ ............................ 73 Table 18 Regression Analyses for Mediation: Hypothesis 5 (Physician Report WIF) ................................ ....................... 74 Table 19 Regression Analyses for Mediation: Hypoth esis 5 (Spouse report WIF) ................................ ............................ 75 Table 20 Regression Analyses for Mediation: Hypothesis 6 (Physician Report WIF) ................................ ....................... 76 Table 21 Regression Analyses for Mediation: Hypothesis 6 (Spouse Report WIF) ................................ ........................... 77 Table 22 Means, Standard Deviations, and Correlations: Types of Spousal Support (Time 1) ................................ ........................... 87 Table 23 Means, Standard Deviations, and Correlations: ................................ ........................... 87 Table 24 ................. 88 Table 25 wi th Types of Physician Report WIF (Time 1) ................................ ......... 89 Table 26 with Types of Physician Report WIF (Time 2) ................................ ......... 90 Table 27 with Types of Spouse Report WIF (Time 1) ................................ ............. 91 Table 28 Predicting Spous with Types of Spouse Report WIF (Time 2) ................................ ............. 92 Table 29 Regression Analyses for Mediation: Supplemental Analyses (Hypothesis 6) ................................ ..................... 93

PAGE 9

vi List of Figures Figure 1 Models of Social Support ................................ ................................ ............ 9 Figure 2 Model Depicting the Relationships for the Proposed Hypotheses ............ 27

PAGE 10

vii The Provision of Spousal Support: Antecedents Consequences, and Crossover Effects Jay M. Dorio ABSTRACT T he present study had four main objectives. First, the relationship between the provision of spousal support and its theoretical antecedents and consequences was assessed as informed by the conservation of resources theory (Hobfoll, 1989). Second, the cross family demands ( perceived family demands and family hours) was investigated. Next, the mediating role of social support as an explanation for the crossover process was examined us ing two distinct pathways. Lastly, the fourth objective of the present study was to investigate the relationships described above across multiple time points and using dual source data (from physicians and their spouses). The final sample included matched responses from 126 couples across two time points. Results were generally supportive of the relationship between the provision of spousal support and the receipt of spousal support, perceived family demands, family hours, and work interference with family conflict (WIF) and were consistent with expected relationships according to COR theory Results also provided support for the synchronous crossover of WIF on perceived family demands; however, results were

PAGE 11

viii generally unsupportive of the mediating role of t he provision of spousal support in the crossover process. The present study makes several important contributions to the social support, work family conflict, and crossover literatures by adding to the knowledge of the antecedents and consequences of the provision of spousal support, the growing body of research examining the crossover of WIF, and the understanding of the mediating role of the provision of spousal support in the crossover process. Major findings and areas of opportunity for future research are discussed.

PAGE 12

1 Chapter One Introduction (Gerber, 1983, p. 10 4) Physicians, who must deal with the ever increasing complexity of medical care, administrative time demands, and pressure inherent in their jobs, are faced with what many consider excessive amounts of stress on a daily basis (Fabri, McDaniel, Gaskill, Ga rrison, Hanks, Maier, et al., 1989). According to the resource drain model, that posits that individuals have a limited amount of resources to devote to various roles, if one domain (e.g., work) requires a large expenditure of resources, performance in an alternate domain (e.g., home) is likely to suffer (Rothbard, 2001; Staines, 1980). Therefore, after dealing with a high level of stressors inherent in his/her work role, a physician may be es as he changes hats Furthermore, although the physician may directly bear the brunt of the stressors he/she encounters, empirical evidence and theoretical arguments suggest that stressors he/she experiences can crossover and impact stressors and strains experienced by his/her partner (Westman & Etzion, 1995; Westman & Vinokur, 1998). Westman (2001) suggests that crossover effects are more likely to occur when a member of the dyad

PAGE 13

2 occupies a high stress occupation (e.g., a physician) and encourages the investigation of such effects within these populations. When additional demands (i.e., requests for support from his/her partner, assistance with household tasks, or help with c hild care activities) are placed upon the already highly stressed physician, he/she may withdrawal from the situation (Gerber, 1983), may limit the amount of support he/she provides (Westman & Vinokur, 1998), or may become so inundated by the demands that he/she may be unable to contribute to the relationship in an effective manner (Luk & Shaffer, 2005). Therefore, physicians, who often see their primary responsibility as caring for their patients (Gerber, 1983), may expend less energy caring for themselves their partners, or their families, and may find balancing work and family roles increasingly difficult. Within the work family literature few studies have investigated the provision of support. Instead, previous studies have focused primarily on the rec ipient of support, eschewing the provider of support and the relationship between the two (Hobfoll, Dunahoo, Ben Porat, & Monnier, 1994; Pearlin & McCall, 1990). Thus, the present study had four main objectives. The first objective was to investigate the p rovision of spousal support over time, as well as its theoretical antecedents and consequences (e.g., family demands and work interference with family conflict). The second objective was to determine if stressors crossover from a physician to his/her partn er, thereby increasing of social support as an explanation for the crossover process (Westman, Etzion, & Danon, 2001; Westman, Vinokur, Hamilton, & Roziner, 2004), t his proposition had yet to be directly tested. Therefore, the third objective of the present study was to test this

PAGE 14

3 hypothesized mechanism and examine the mediating role spousal support plays in the crossover process. Finally, the fourth objective of the p resent study was to investigate the relationships described above across multiple time points and using dual source data ( from physicians and their partners). Therefore, by examining the provision of spousal support, its antecedents and consequences, cross over effects, and the mediating potential of spousal support across multiple time points and employing dual source data, the present study contributes to a largely neglected area within the social support and crossover literatures. Social Support Social su provide individuals with actual assistance or with a feeling of attachment to a person or support is considered an interpersonal resource, in that individuals who use support systems are likely to receive assistance meeting demands, whether directly or through the reduction in demands placed on them from other roles (House, 1981). Dimensions of Social Su pport Early conceptualizations of social support included work by Pinneau (1975) who described tangible, appraisal/informational, and emotional support, and Cobb (1976) who described esteem, network, and emotional support. A more recent and widely employe d conceptualization of social support was proposed by House (1981) and involved the dimensions of emotional concern, instrumental support, informational support, and appraisal support. Similar to the description proposed by Cobb (1976), House proposed that

PAGE 15

4 reflects behaviors that are p rovided that directly assist the person in need (e.g., helping information to anot her individual, such as suggestions or advice about a problem to a spouse. House further distinguished appraisal support as information provided to an individual that is useful in social comparisons (i.e. self evaluation), somewhat similar to tualization of esteem support. However, while esteem support emphasizes that the individual is valued by others, appraisal support does not connote only positive evaluations. Appraisal support would be evident when a spouse thanks his/her partner for doing a good job at home, but also when a spouse criticizes his/her partner for doing something wrong. Although social support has been conceptualized using a diverse range of dimensions (e.g., esteem, network, tangible, affectionate, positive social interactio n), researchers have most often examined emotional and instrumental support (Beehr, Jex, Stacy, & Murray, 2000). Furthermore, researchers have typically focused on one type of support at a time (Erdwins, Buffardi, & Casper, 2001) with greater attention foc used on emotional support (e.g., Bernas & Major, 2000; Grzywacz & Marks, 2000) to the exclusion of instrumental support (Lapierre & Allen, 2006). Additionally, while some researchers have included multiple dimensions within their studies, most combine item s across dimensions to create a composite measure of support (e.g., Aycan & Eskin, 2005; Erdwins et al., 2001; Parasuraman, Greenhaus, & Granrose, 1992). With a composite approach, valuable information on the differences across dimensions may be lost. Desp ite

PAGE 16

5 this loss of information, few researchers have examined the individual contributions of each dimension of social support within the same study (see Lapierre & Allen, 2006). port and investigated emotional concern, instrumental, informational, and appraisal support. Although hypotheses were proposed using a composite of the four types of social support described above, exploratory analyses were conducted that investigated each facet of support individually. Sources of Social Support Researchers have investigated a wide range of sources of social support, including work related (e.g., supervisor, co worker, organizational) and non work related sources (e.g., spouse, family, fr iend ) The domain specificity hypothesis suggests that support originating in one domain is more likely to be related to other variables within the same domain than to variables in an alternate domain ( Frone, 2003; Frone, Russell, & Cooper, 1992 ) Consiste nt with this hypothesis, research supports that work related social support has been negatively associated with work related variables such as work role conflict, role ambiguity (Beehr & Drexler, 1986; Beehr, King, & King, 1990 ) and turnover intentions (Ba tt & Valcour, 2003), and positively associated with job satisfaction (Beehr & Drexler, 1986; Frye & Breaugh, 2004 ) and work related support has been negatively associated with non work related variables such as family role conflict ( Carlson & Perrew 1999), family stressors (Bernas & Major, 2000), depression (Beatty, 1996; Rosenbaum & Cohen, 1999; Schwarzer & Gutirrez Doa, 2005), and anxiety ( Rosenbaum & Cohen, 1999; Westman, Etzion, & Horovitz, 2004),

PAGE 17

6 and positively associated with family satisfaction ( Carlson & Perrew 1999; Hill, 2005) and marital satisfaction (Beatty, 1996; Hill, 2005). With the recognition that the marital partner is the principal source of social support for mo st adults, and that spousal support is one of the most reliable sources of social support available (Beach, Martin, Blum, & Roman, 1993), the role that a supportive spouse plays has been recognized as critical to understand. Accordingly, spousal support ha s received the most empirical investigation of all non work related sources of social support (Aycan & Eskin, 2005; Cinamon & Rich, 2005 ) and was the focus of the present study. Receiving Spousal Support In addition to the various dimensions and sources o f social support studied, several mechanisms have been proposed and investigated to explain the relationship between receiving social support, stressors (e.g., job demands), and strains (e.g., depression). Figure 1 illustrates the three mechanisms discusse d below. The direct effects model of social support (Figure 1A) has received the most support in the current literature and impacted by the support received, regardless of the current level of stressor s (Cohen & Wills, 1985). In this model, social support serves as an antecedent to the strain variable and can serve a protective function for the individual (Brown & Bifulco, 1985; Brown & Harris, 1978). Empirical evidence supports a negative relationship between spousal support and a variety of strains including depression ( Beatty, 1996; Schwarzer & Gutirrez Doa, 2005) and anxiety (Ross, Mirowsky, & Huber, 1983). Viswesvaran, analysis supports these findings and reported a sample size weighted average corrected correlation of .21 between strains and all categories of

PAGE 18

7 support. Further supporting the direct effects model, the receipt of spousal support has been positively related to a variety of constructs as well, includi ng marital satisfaction (Beatty, 1996; Hill, 2005; Purdom, Lucas, & Miller, 2006), family satisfaction (Burke & Greenglass, 1999; Hill, 2005; Parasuraman et al., 1992) and job satisfaction (Beatty, 1996; Parasuraman et al., 1992; Rosin, 1990; Rudd & McKenr y, 1986). Figure 1B illustrates the moderating or buffering hypothesis, that has been referred to as the dominant hypothesis employed in occupational stress research (Ganster, Fusilier, & Mayes, 1986). It hypothesizes that the relationship between stressor s and strains is stronger for individuals receiving lower levels of social support, as compared to individuals receiving higher levels of social support (Beehr et al., 1990; Carlson & Perrew, 1999). Evidence in support of the buffering hypothesis was foun d in analysis, although limited research has tested the buffering capabilities of spousal support (e.g., Suchet & Barling, 1986). Conversely, a reverse buffering hypothesis has also been supported in the social support lite rature. In this model, the relationship between stressors and strains is stronger for individuals receiving greater social support than those receiving less (Glaser, Tatum, Nebeker, Sorenson, & Aiello, 1999; Viswesvaran et al., 1999). Finally, much less r esearch has investigated social support as an intervening or mediating variable (Figure 1C), in which social support is hypothesized to mediate the stressor strain relationship (Wheaton, 1985). In this model, social support is mobilized in response to a st ressor, thereby reducing the effects of the stressor (Carlson & Perrew, 1999). Although Viswesvaran et al. (1999) did not find support for the mediating effect of social support in their meta analysis; some limited empirical evidence does exist.

PAGE 19

8 Carlson a nd Perrew (1999) reported that a structural equation model depicting social support as a mediator of the relationship between stressors (role conflict, time demands, and role ambiguity) and work family conflict fit the data moderately well. Additionally, Burley (1995) found evidence that emotional spousal support mediated the relationship between work family conflict and marital adjustment. Although the direct effects model of social support has received the majority of research attention and support, sev eral authors have suggested that researchers should move beyond the study of direct effects and examine variables that may influence or explain the relationship between variables of interest (Allen, Herst, Bruck, & Sutton, 2000; Dorio, Bryant, & Allen, 200 8). Despite this recommendation, few studies have examined the mediating potential of social support. Therefore, although the beneficial effects associated with social support are well known, little is known about the processes that explain these effects. This lack of research attention may be attributable to the bias in the current literature of using only self report studies (Greenhaus, Allen, & Spector, 2006; Hammer Cullen, Neal, Sinclair, & Shafiro 2005) that are cross sectional in nature (Kahn & Byos iere, 1991; Zedeck & Mosier, 1990).

PAGE 20

9 Figure 1. Models of S ocial S upport Model A: Direct Effect Social Support Stressor Strain Social Support Stressor Strain Social Support Stressor Strain Mode l B: Moderating Effect Model C: Mediating Effect

PAGE 21

10 Receiving Spousal Support versus Providing Spousal Support It is important to note that while the receipt of support is not synonymous with the provi sion of support, a relationship exists between the two constructs providing useful insight. In the only study found that assessed both the receipt and provision of spousal support, Bruck (2002) found y related ( r =.43) to his/her have employed measures of the provision of spousal support as a proxy for the receipt of spousal support (e.g., Repetti, 1989); further supporting that the two construct s should be reasonably related. For example, it is likely that if one spouse provides support (e.g., helping with housework, listening empathically to his/her spouse), his/her partner will report receiving support. However, it is also possible that while one spouse may report he/she is providing as much support as possible (e.g., picking up the children from school, doing work around the house), his/her partner may report receiving little support. This disparity may be exacerbated if the type of support co nsidered differs between individuals (e.g., if one spouse focuses on providing instrumental support, while his/her partner focuses on the receipt of emotional support). Thus, while useful insight can be gained from the examination of previous research inve stigating the receipt of social support, differences in the constructs warrant the consideration of the provision of social support. Providing Social Support Clearly, a wealth of research exists demonstrating a beneficial relationship between the receipt of social support and a variety of outcomes. However, the impact of providing social support is virtually unknown, as nearly all previous research has focused solely on the recipient of social support, neglecting the provider of support and the relationsh ip between the two (Hobfoll et al., 1994; Pearlin &

PAGE 22

11 McCall, 1990). Despite this lack of empirical investigation, several authors have offered theories regarding the provision of social support. Dunkel Schetter and Skokan (1990) described stress, relations hip, recipient, and provider factors that are hypothesized to influence the provision of support. According to these authors, support is likely to be provided when the situation is stressful (stress factors), the recipient is distressed (recipient factors) there is intimacy or satisfaction with the relationship (relationship factors), or the provider of support is high in empathy (provider factors; Caldwell & Reinhart, 1988; Hobfoll & Lerman, 1988). Consistent with this theory, several researchers have inv estigated the characteristics of individuals who provide support (i.e., provider factors), and have found that women (Fischer, 1982; Griffith, 1985; Kessler, McLeod, & Wethington, 1985) and individuals high in empathy (Trobst, Collins, & Embree, 1994) are slightly more likely to provide social support than are men or those low in empathy. Similarly, Allen (2003) found that individuals high in other oriented empathy were more willing to mentor others (a form of social support). Haines, Hurlbert, and Beggs (1 996) found that age and income also impacted the provision of social support. The occasions when individuals are likely to provide support has also been investigated. Tyler (2006) found that individuals who had received support following a natural disaster were likely to provide support at a later time. Additionally, Bozionelos (2004) found that individuals who had received support in the form of mentoring were more likely to provide mentoring later in their careers than those who had not. Other authors ha ve theorized about the process of providing social support (e.g., Hobfoll, 1989; Hobfoll, Freedy, Lane, & Geller, 1990; Westman & Vinokur, 1998).

PAGE 23

12 maintain and accumulat e resources and that the loss or potential loss of these resources produces stress. R esources, such as objects (e.g., a home), personal characteristics (e.g., stress resistance), conditions (e.g., marriage), energies (e.g., knowledge), or social support ca n be useful in buffering an individual from stress; however, it is important to note that these resources are not inexhaustible commodities (Hobfoll, 1989). According to COR theory, the receipt of social support is classified as a resource, however, Hobfo ll (1989) discusses the provision of social support as well. Although seemingly contrary to the basic tenet of his theory, Hobfoll points out that individuals take a longitudinal approach to conserving resources and that providing social support can be see n as making an investment in future resources. According to Hobfoll, an individual in a supportive relationship would provide social support when needed, and would receive social support at a later time. However, the provision of social support can also be social support may become overwhelmed by the demands inherent in providing support and may have insufficient resources to continue in the same manner (Luk & Shaffer, 2005). In an attempt to regain some valued resources, this individual may limit or cease his/her provision of social support. Similarly, the resource drain model posits that individuals have a limited amount of resources to devote to various roles; thus, if one dom ain requires a large amount of resources, performance in the other domain will suffer (Rothbard, 2001; Staines, 1980). Resources can be physical, psychological, or social features that aid in goal completion and/or demand reduction, including dispositions or internal cognitions (Demerouti,

PAGE 24

13 Bakker, Nachreiner, & Schaufeli, 2001). Frone (2003) suggests that certain dispositional variables (e.g., positive affectivity, mastery, hardiness, extraversion) may function as individual resources that allow individuals to cope with stressors. Family friendly benefits, a supportive culture, and sources of social support would be considered external resources (Voydanoff, 2005). Despite these compelling theoretical arguments, few studies have investigated effects associate d with providing social support for the providers themselves, with the majority of these studies focusing on older adults. Results of these investigations have been equivocal, with some studies finding positive effects associated with providing support, so me finding negative effects, and some finding no effects at all. Brown, Nesse, Vinokur, and Smith (2003) found that mortality was lower for individuals who provided emotional support to their spouse among a sample of older adults. Krause and Shaw (2000) fo und that providing social support was related to increased self esteem using a similar sample. Conversely, Liang, Krause, and Bennett (2001) found that providing social support was related to increased negative interactions among a sample of older adults. impacted by providing instrumental support (assistance with household chores) to his wife. Thus, although little research has examined the provision of spousal support, limited evidence and several theoretical rationales exist suggesting that the provision of support may have both beneficial and detrimental outcomes (Kessler & McLeod, 1984). Therefore, to add to the body of research examining the provision of social support, the present study investigated the consistency of the provision of spousal support over time.

PAGE 25

14 A lthough no previous research had examined the provision of spousal support over time several investigations have assessed the receipt of social support longitudinal ly. Westman, Etzion, et al. (2004) assessed the receipt of social support using a two month interval and found a moderate correlation ( r= .56). Thompson, Jahn, Kopelman, and Prottas, (2004) found a similar relationship between perceived organizational famil y support assessed at two times, 18 months apart. Building on these examples, it is expected that the provision of support will follow the pattern of results found regarding the receipt of support. Thus the following hypothesis was proposed: Hypothesis 1 : There will be a positive relationship between the provision of spousal support at Time 1 and Time 2. Family Demands Family demands have been defined as e therefore associated with certain Family demands have been conceptualized and measured in a variety of ways including: time spent doing chores or the number of children living at ho me (Keene & Reynolds, 2005; Roehling, Moen, & Batt, 2003). Other researchers simply use measures of time spent in the family domain when conceptualizing family demands. The relationship between demands originating in the family domain and work family conf lict has received a great deal of empirical investigation (Frone, 2003). For example, Keene and Reynolds (2005) found a positive relationship between family demands and family interfering with work conflict. Gutek, Searle, and Klepa (1991) found a positive relationship between hours at home and family interfering with work

PAGE 26

15 conflict. Consistent with the rational model of work family conflict that states that conflict can be expected to increase as the amount of time spent in a domain increases (Staines, Plec analysis found a positive relationship between hours engaged in non work activities and family interfering with work conflict ( r =.21). Providing Spousal Support and Family Demands Despite a well documente d relationship between family demands and work family conflict, no research could be found investigating the relationship between providing spousal support and family demands. As mentioned previously, spousal support is considered an interpersonal resource in that individuals who use support systems are likely to receive assistance meeting demands, whether directly or through a reduction in demands placed on them (House, 1981). By definition, an individual who provides spousal support demonstrates interper sonal skills such as listening and understanding, as well as supportive behaviors thereby allowing the individual more time to devote to other tasks (Bruck & Allen, 2005 ). However, consistent with COR theory described above (Hobfoll, 1989), a physician who reports a high level of family demands initially, may attempt to conserve his/her resources and may be less likely or able to provide spousal support at a later time. T hus the following hypothesis was pr oposed: Hypothesis 2: There will be a negative relationship between physicians family demands at Time 1 and t he i r provision of spousal support at Time 2.

PAGE 27

16 Work Family Conflict Greenhaus and Beutell (1985) described work family conflict (WFC) as a form of inter role conflict that occurs when expectations and demands of work and family roles are incompatible. WFC is considered bi directional in that conflict can arise in the work domain and interfere with family (WIF) or c an arise in the family domain and interfere with work (FIW). Greenhaus and Beutell (1985) identified three types of work family conflict: time based, strain based, and behavior based conflict. Time based conflict occurs when the time required by one role interferes with the effective functioning of the other role, or prevents participation in the other role (e.g., a physician has to miss dinner with his/her spouse due to a late meeting at work). Strain based conflict occurs when stressors associated with o ne role are carried over and negatively affect the other role (e.g., a physician who loses a patient at work is too upset to have a conversation with his/her spouse at home). Behavior based conflict occurs when the behaviors exhibited in one role are incom patible with the other role (e.g., a physician who is authoritative at work attempts to use the same behaviors at home). More recently, Carlson and Frone (2003) described internally generated work family conflict. Internally generated conflict occurs when an individual is psychologically preoccupied with one role while physically present in the other role (e.g., a physician who cannot stop worrying about his/her patient while at home).

PAGE 28

17 A limitation of previous work family conflict research is that WFC has been operationalized and measured in a variety of ways, making comparisons across studies difficult (Allen et al., 2000). For example, WFC has been assessed using single item measures or composite measures that confound the directionality of conflict (WIF/ FIW; Carlson, Kacmar, & Williams, 2000). Other studies have employed measures that exclude behavior based conflict, or have different foci (e.g., work interference with leisure, work interference with home life). Thus, the present study employed Greenhaus conflict (WIF) and all three types of conflict (time based, strain based, and behavior ension of internally generated conflict would be especially pertinent for the sample of interest (physicians), who may often be preoccupied with their jobs. Therefore, internally generated conflict was also included in the present study. Although hypothese s were proposed using a composite of the four types of WIF described above, exploratory analyses were conducted that investigated each facet of conflict individually. WFC and the Receipt of Spousal Support The receipt of social support has garnered a weal th of investigation and has consistently been negatively associated with WFC. That is, the receipt of social support has been associated with lower levels of WFC. Consistent with the domain specificity hypothesis ( Frone, 2003; Frone et al., 1992 ), the majo rity of e mpirical research has found support for a relationship between spousal support and FIW conflict (Aycan & Eskin, 2005; Burke & Greenglass, 1999; Frone,

PAGE 29

18 (2005) meta ana lysis, a weighted average corrected correlation of .17 was reported between family support (f amily and spouse ) and FIW. A relationship has also been found between receiving spousal support and WIF (Cinamon & Rich, 2005; Matsui, Ohsawa, & Onglatco, 1995), as well as composite measures of WFC (Aryee, 1992; Erdwins et al., 2001; Kim & Ling, 2001; Parasuraman et al., 1992). Byron (2005) found a weighted average corrected correlation of .11 between family support (f amily and spouse ) and WIF. Furthermore, rese archers have found support for the buffering effect of spousal support on the relationship between WFC and a variety of stressors, including parental role overload (Aryee, Luk, Leung, and Lo 1999), parental demands (Matsui et al., 1995), family demands, a nd job stressors such as role overload, role ambiguity, and a lack of autonomy (Westman & Etzion, 2005). While the relationship between WFC and the receipt of spousal support has been consistently demonstrated, no research could be found investigating the relationship between WFC and the provision of spousal support. Despite the lack of empirical studies directly examining this relationship, a wealth of research exists linking WFC to a variety of constructs that provide the rationale for the proposed hypoth eses. WIF and the Provision of Spousal Support A large body of research has investigated the relationship between WIF and its consequences. Consistent with the domain specificity hypothesis (Frone, 2003; Frone et al., 1992 ) high levels of WIF have been as meta analysis, WIF was positively associated with family related distress (e.g., Frone et al., 1997; Williams & Alliger, 1994) and burnout (e.g., Bacharach, Bamberger, & Con ley, 1991; Kinnunen & Mauno, 1998). Additionally, Frone et al. (1997) found a

PAGE 30

19 strong negative relationship between WIF and family performance, indicating that higher levels of WIF were associated with decreased family performance. Allen et al. (2000) also found that WIF was negatively related to satisfaction in the family domain. A negative relationship was found between WIF and marital satisfaction and between WIF and life satisfaction, suggesting that individuals reporting high levels of WIF were less sat isfied with their marriages (e.g., Greenglass, Pantony, & Burke, 1988; Suchet & Barling, 1986) and their lives (e.g., Judge, Boudreau, & Bretz, 1994; Rice, Frone, & McFarlin, 1992) Therefore, consistent with previous research that illustrates that WIF ha s been related to decreased functioning in the family domain (e.g., decreased performance and satisfaction, and increased distress and burnout) it is expected that a physician reporting a high level of WIF will provide less spousal support. For example, a physician who experiences a high level of WIF may be unable to participate in family related activities due to experiencing a high level of work related stressors (e.g., work demands, role overload), or may be too anxious, burned out, or preoccupied with w ork related issues to perform effectively in the family domain. As a result of this conflict, it is expected that the physician will be less likely to provide spousal support. Thus the following hypothesis was proposed: Hypothesis 3: There will be a negat ive relationship between physicians WIF at Time 1 and t he i r provision of spousal support at Time 2.

PAGE 31

20 Crossover Effects Although a physician may directly bear the brunt of the stressors he/she encounters, empirical evidence and theoretical arguments sugges t that the stressors he/she experiences can crossover and impact the stressors and strains experienced by his/her partner (Westman & Etzion, 1995; Westman & Vinokur, 1998) Crossover effects are described as the effect of stressors/strains experienced by o ne member of a dyad impacting the stressors/strains of the other member of a dyad (Westman, 2001; Westman & Etzion, 1995; Westman & Vinokur, 1998). Crossover effects are considered inter individual stressors, in that the source of stress originates from th e interaction between According to Westman (2001), crossover effects can occur uni directionally (transferred from only one member of the dyad to the other) or bi directionally (transferred simultaneously between b oth members of the dyad). For example, burnout. On the other hand, bi directional cro directional crossover (e.g., Jones & Fletcher, 1993; Westman et al., 2001), the majority of research has focused on bi directional crossover (e.g., Hammer, Allen, & Grigsby, 1997; Hammer, Bauer, & Grandey, 2003; Westman & Etzion, 1995).

PAGE 32

21 Research Examining Crossover Effects Westman (2001) suggested four ways that crossove r can occur. First, crossover can occur when stressors transmitted from one member of the dyad impact the strain of the other member. The majority of crossover research has focused on this pathway, with the preponderance focusing on job stressors. Specific increased anxiety (Jones & Fletcher, 1993), psychological distress (Rook, Dooley, & Catala no, 1991), psychosomatic symptoms (Burke, Weir, & DuWors, 1980), and burnout (Pavett, 1986); as well as decreased mental health (Morrison & Clements, 1997), and general well being (Long & Voges, 1987). Despite the abundance of research examining the crosso depressi on. Second, crossover can occur when strain from one member of the dyad impacts the strain of the other member. Several studies have demonstrated bi directional crossover of strains, including anxiety (Westman, Etzion, et al., 2004) and burnout (Westman & Etzion, 1995). Other researchers have demonstrated uni directional crossover of strains, with some finding crossover from husbands to wives (Westman et al., 2001; Westman & Vinokur, 1998; Westman, Vinokur, et al., 2004), and others finding crossover from w ives to husbands ( Chan & Margolin, 1994; Demerouti, Bakker, & Schaufeli, 2005).

PAGE 33

22 Third, Westman (2001) suggested that crossover can occur when the strain of one member of the dyad impacts the stressors of the other member. Although no studies could be found assessing this combination of variables, this pathway would be evident if the anxiety of one member of the dyad crossed over and impacted the work overload of the other member. Finally, crossover can occur when stressors transmitted from one member of th e dyad impact the stressors of the other member. Although limited research has been conducted examining stressor to stressor crossover, several studies have found support for this pathway. Bolger, Delongis, Kessler, and Wethington (1989) found that husband Crossover and Work Family Conflict Despite the g rowing body of research examining crossover effects, few researchers have examined crossover effects associated with work family conflict. Using a composite measure of WFC, Hammer et al. (1997) found bi perceived flexibility, and family involvement. Westman and Etzion (2005) found that WIF beyond their individual variables.

PAGE 34

23 Consistent wi th the domain specificity hypothesis ( Frone, 2003; Frone et al., 1992 ), crossover research has demonstrated that the WIF of one partner can crossover and has been associated with family related consequences of the other partner, and that the FIW of one par tner can crossover and has been associated with work related consequences of the other partner. For example, Hammer et al. (2005) found support for Acitelli, and Barnes Farrel l (2006) found support for the crossover of WIF on increased associated with decreased relation ship tension for their wives. Finally, Parasuraman et al. (1992) failed to find support for the crossover of WFC on family satisfaction; although a composite measure of conflict was used obfuscating the direction of conflict. On the other hand, Hammer et al. (2003) found support for the crossover of FIW rk and absences. Greenhaus, Parasuraman, Granrose, Rabinowitz, and Beutell (1989) investigated and found support for the crossover of job involvement and career priority on time based and strain based WFC. Therefore, despite the paucity of research examin ing crossover effects associated with WFC, a trend is evident in that domain specific crossover effects occur. While the logical (because these effects occur within the s

PAGE 35

24 environments in which these events occur (although a common stressors explanation is plausible and is discussed below). Thus, t he present study adds to the growing body of research examining the crossover of WFC, and focused specifically on WIF. Consistent with domain specific predictive of the famil y demands of his/her partner. For example, a physician who experiences a high level of WIF may be frequently absent from family related activities due to work responsibilities, or may be too anxious, burned out, or preoccupied with work related issues to p erform effectively in the family domain. As a result of this conflict, the partner of the physician is likely to be faced with additional responsibilities, pressures, and demands. Thus the following two hypotheses were proposed: Hypothesis 4a: There will be a positive relationship between physicians WIF at Time 1 and t he i r partners family demands at Time 2. Hypothesis 4b: There will be a positive relationship between physicians WIF at Time 2 and t he i r partners family demands at Time 2. Mechanisms of Cr ossover Three mechanisms have been posited to explain the crossover process: direct crossover, common stressors, and indirect crossover (Westman & Vinokur, 1998). The direct crossover of stressors/strains is typically explained via empathy (Westman, Vinok ur, et al., 2004). Crossover is hypothesized to occur when stressors/strains experienced by one partner cause an empathic reaction in the other Westman & Vinokur, 1998). A lthough the mechanism of direct empathic crossover is frequently cited in the crossover literature, support for this pathway is generally assumed

PAGE 36

25 rather than directly tested (Westman, 2001; Westman et al., 2001; Westman, Etzion, et al., 2004). The second m echanism, the common stressors explanation of crossover, hypothesizes that crossover effects are spurious, and that increased stressors/strains are caused by common stressors simultaneously affecting both members of the dyad (Westman & Vinokur, 1998). West man and Etzion (1995) suggested that aspects of a shared environment could lead to increased stressors/strains for both partners giving the appearance of crossover effects. Although researchers have investigated a variety of stressors (e.g., unemployment, financial hardship), mixed results have been found (Westman, Etzion, et al., 2004; Westman, Vinokur, et al., 2004). Finally, the third mechanism, an indirect process, hypothesizes that interaction style plays a key role in the crossover process (Westman & Vinokur, 1998). responsible for crossover effects including social undermining and social support. According to the indirect crossover hypothesis, stressors/str ains experienced by one partner are hypothesized to impact the interaction he/she has with his/her partner. This negative interaction is hypothesized to result in increased stressors/strains for the d that social support could serve in this mediating capacity, research has focused exclusively on social undermining. For example, Westman, Vinokur, et al. (2004) found that social undermining mediated the relationship between distress and marital dissatis faction. However, Westman and Vinokur (1998) and Westman et al. (2001) failed to find mediating effects of social

PAGE 37

26 undermining in their investigations of the crossover of burnout and depression respectively. Although researchers frequently discuss the media ting role of social support in the crossover process (Westman et al., 2001; Westman & Vinokur, 1998; Westman, Vinokur, et al., 2004), no research could be found testing this hypothesis. Thus, by examining the mediating potential of spousal support, the pre sent study focuses on a largely neglected area within the social support and crossover literatures. First, it was proposed that a physician who experiences a high level of WIF may be less likely to provide spousal support for his/her partner. In turn, this decreased provision of support may be associated with increased family demands for his/her partner. Thus the following hypothesis was proposed: Hypothesis 5: The relationship between physicians WIF at Time 1 and t he i r partners family demands at Time 2 will be partially mediated by the physicians provision of spousal support at Time 2. Second, it was expected that a high level of WIF reported by a physician may be associated with a high level of support provided by his/her partner. In turn, this increas ed provision of spousal support may be associated with an increased level of family demands for the partner. Thus the following hypothesis was proposed: Hypothesis 6: The relationship between physicians WIF at Time 1 and t he i r partners family demands at Time 2 will be partially mediated by the partners provision of spousal support at Time 2. The relationships described above for the proposed hypotheses are depicted in Figure 2.

PAGE 38

27 Figure 2. Model Depicting the Relationships fo r the Proposed Hypotheses Note. The numbers inset on the paths represent the hypotheses discussed above. Spouse Time 1 Physician Time 2 Family Demands WIF Provision of Spousal Support WIF Provision of Spousal Support Provision of Spousal S upport Family Demands H2 H1 H3 H4a H4b H5 H5 H6 H6

PAGE 39

28 Summary While the beneficial effects associated with the receipt of spousal support are well known, little is known about the impact of providing s pousal support. Therefore, the present study investigated the provision of spousal support, its theoretical antecedents, and consequences using COR theory as an organizing structure. Next, d espite the growing body of research examining crossover effects, f ew researchers have examined crossover effects associated with WFC. Heeding that crossover effects are more likely to occur when a member of the dyad occupies a high stress occupation, the present study investigated the crossove r of WIF from physicians to their partners. Next, the present study tested a frequently cited yet untested mechanism of the crossover process, and investigated the mediating role spousal support plays in the crossover process. Finally, the present study in vestigated the relationships described above across multiple time points and using dual source data.

PAGE 40

29 Chapter Two Method Participants To establish the desired sample size, two a priori power analyses were conducted. First, for hypotheses employing mult iple hierarchical regression, an a priori power analysis was conducted following procedures outlined by Cohen (1992). Based on an analysis involving a candidate set of five independent variables, a desired alpha of .05, and a medium effect size (f 2 =.15), a minimum of 91 couples were required to achieve a power of .80. Next, for hypotheses involving mediation, an a priori power analysis was conducted following procedures outlined by Fritz and MacKinnon (2007). For this analysis, three effect sizes were requi red; the relationship between the independent variable and the dependent variable (tau), the relationship between the independent variable and the mediator (alpha), and the relationship between the dependent variable and the mediator (beta). Based on an an alysis using the Sobel test (Sobel, 1982) a lar ge The final sample included matched responses from 126 couples across two time points. At Time 1, matched responses were obtained from 148 couples (180 physicians and 148 spouses) and at Time 2, matched responses were obtained from 137 couples (156

PAGE 41

30 used in analyses investigating Hypoth eses 4a and 4b, most analyses were conducted using 137 pairs of participants. Demographic characteristics for the total sample, physician sample, and spouse sample (collected at Time 1) are provided in Table 1. The majority of participants (78.7%) were mar ried and had been married an average of 13.0 years ( SD =10.5). Two thirds (66.5%) of participants had children ( M =1.66, SD =1.36), with 22.0% having at least one child two years of age or younger. Participants were primarily born between 1966 and 1981 (47.9% ) and between 1954 and 1965 (22.6%). The majority of physicians were male (59.4%) and identified themselves as Caucasian (74.4%). Physicians were employed in p rivate practice (43.3%), as m edical residents (26.7%), and as faculty physicians (22.2%) ; and re ported a variety of areas of practice including: Pediatrics (8.9%), Internal Medicine (6.7%), Surgery (6.7%), Obstetrics and Gynecology (5.6%), Psychiatry (5.0%), and Family Medicine (4.4%) among others. The average length of practice for physicians was 12 .8 years ( SD =10.8). The majority of spouses were female (48.6%) and identified themselves as Caucasian (65.5%). Nearly one third (31.8%) of spouses were employed in a medical profession, with more than half of those employed as physicians (59.6%). Thirty nine holding an advanced degree.

PAGE 42

31 Table 1 Demographic Characteristics of Participants (Time 1) Total sample Physician sample Spouse sample N=328 N=180 N=148 Variable N % N % N % Gender Male 154 47.0% 107 59.4% 47 31.8% Female 134 40.9% 62 34.4% 72 48.6% Ethnicity Caucasian, Non Hispanic 231 70.4% 134 74.4% 97 65.5% Hispanic 20 6.1% 10 5.6% 10 6.8% Asian 16 4.9% 9 5.0% 7 4.7% Black or African Am erican 9 2.7% 6 3.3% 3 2.0% Other 9 2.7% 6 3.3% 3 2.0% American Indian 1 0.3% 1 0.6% 0 0.0% Pacific Islander 1 0.3% 1 0.6% 0 0.0% Year of Birth 1982 1995 9 2.7% 5 2.8% 4 2.7% 1966 1981 157 47.9% 91 50.6% 66 44.6% 1954 1965 74 22.6% 40 22.2% 34 23.0% 1942 1953 41 12.5% 27 15.0% 14 9.5% 1925 1941 8 2.4% 6 3.3% 2 1.4% Marital Status Married 258 78.7% 151 83.9% 107 72.3% Living with partner 31 9.5% 18 10.0% 13 8.8% Single 0 0.0% 0 0.0% 0 0.0% Participants with C hildren No children 77 23.5% 51 28.3% 26 17.6% Children (any age) 218 66.5% 129 71.7% 89 60.1% Children ( 2 years or younger) 72 22.0% 42 23.3% 30 20.3% Note. Numbers/percentages may not sum to total sample size due to missing data

PAGE 43

32 Table 1 (Con tinued) Total sample Physician sample Spouse sample N=328 N=180 N=148 Variable N % N % N % Job type Private practice 78 43.3% % Medical resident 48 26.7% % Faculty physician 40 22.2% % Other 10 5.6% % In a medical profession % 47 31.8% Employed as a physician % 28 18.9% Education High school diploma/GED % 4 2.7% Vocational school % 1 0.7% Bachelor degree % 58 39.2% Master's degree % 32 21.6% P rofessional degree % 38 25.7% Ph.D. % 2 1.4% Other % 3 2.0% Mean SD Mean SD Mean SD Length of marriage (in years) 13.0 10.5 13.0 10.6 % 13.0 10.3 % Job tenure (in years) 12.8 10.8 % Note. Numbers/percentages may not sum t o total sample size due to missing data

PAGE 44

33 Response rate Consistent with previous work family research (e.g., Bedeian, Burke, & Moffett, 1988; Beutell & Greenhaus, 1982), physicians, faculty physicians, and medical residents were required to be employed a t least 20 hours per week and married or living permanently with their partner. Because approximately half (50.7%) of the individuals residing in Florida are married ( U.S. Bureau of the Census 2006), and with the recognition that the divorce rate among ph ysicians is 10% to 20% higher than that of the general population (Sotile & Sotile, 1996), it is likely that a portion of the physicians invited to participate in the present study were ineligible. Across the recruitment sources described below, approxima tely 2,200 physicians were invited to participate in the present study. Additionally, individuals were asked to forward the email invitation to other physicians. Due to this sampling strategy, calculating an exact response rate is difficult. First, physici ans may have received an email invitation who were not eligible to participate in the study (i.e., not married or living permanently with their partner). Second, physicians may have received the email invitation from multiple sources, artificially inflatin g the number of invited physicians. Third, individuals may have forwarded the email invitation to many physicians or to none at all. In an attempt to address these issues, a survey link was created that individuals could use to opt out of the study. Howeve r, at Time 1 only 79 physicians accessed the opt out page, with 62.0% indicating that they did not have the time or did not wish to participate. Thirty eight percent of these physicians indicated they were not eligible to Therefore, while an exact number of eligible physicians invited to participate is not known, a reasonable estimate was determined and is provided in Table 2.

PAGE 45

34 Table 2 Estimated Number of Eligible Physicians Invited to Particip ate Recruitment source Invited Self reported ineligible Invalid emails Likely ineligible Approx. # eligible USF medical residents 607 0 0 *238 369 USF faculty 275 0 0 110 165 HCMA **700 0 0 280 420 North collier hospital 394 16 25 117 236 Onl ine 187 11 1 63 112 Snowball sample 40 40 Total 2203 27 26 808 1342 Opted out / ineligible 79 / 30 Grand total 1312 Note. HCMA=Hillsborough County Medical Association. Estimates of likely ineligible physicians were calculated assumin g approximately 40% of invited participants were ineligible to participate (unless noted). According to records obtained from the University of South Florida, 52.5% of medical residents invited to participate in the present study were not married and the refore not likely to be eligible to participate. If approximately 25% of these individuals were permanently living with their partners, an additional 80 medical residents were likely eligible. **Approximately 1,300 members of the Hillsborough County Medic al Association (HCMA) were invited to participate. However, because the HCMA is composed of physicians and medical residents located within Hillsborough County (Florida), it is likely that a significant overlap occurred between the individuals invited to p articipate from HCMA and other recruitment sources (i.e., USF faculty and USF medical residents). If a large percentage (e.g., 70%) of USF medical residents and faculty were members of HCMA (and thus already invited to participate), the total number of HCM A members who had not previously received an email invitation could be approximately 700 individuals.

PAGE 46

35 Thus, the number of eligible physicians invited to participate was estimated at 1,312. Based on this estimate, the response rate for physicians at Time 1 was 13.7% (180 out of 1,312); while at Time 2, 86.7% of physicians participated. For spouses, 83.1% participated at Time 1 and 81.5% participated at Time 2. Response rates and the number of physicians and spouses participating at Times 1 and 2 are provid ed in Table 3. The number of physicians participating from each recruitment source is provided in Table 4. Although a response rate of approximately 38% was anticipated for physicians at Time 1 (Beebe, Locke, Barnes, Davern, & Anderson, 2007), several fac tors likely led to the decreased participation rate obtained. Contrary to the investig ation conducted by Beebe et al. that employed a cross sectional, self report methodology, the present study required physicians to participate at multiple time points. A dditionally, physicians were asked to recruit the participation of their spouses/partners at multiple time points as well. Thus, while a higher response rate may have been obtained if the current study was cross sectional, or self report only; the design o f the present study may have played a role in detracting participants. However, while the initial response rate at Time 1 may have been lower than expected, response rates for physicians at Time 2 and for spouses at Times 1 and 2 were quite high. For examp le, while 83.1% of spouses participated at Time 1, nearly the same percentage ( 81.5% ) participated at Time 2. These response rates suggest that although physicians may have initially been reluctant to participate at Time 1, those that did participate were willing to do so at multiple time points, as were their spouses/partners.

PAGE 47

36 Table 3 Physician and Spouse Response Rates (Times 1 and 2) Participant type Time Invited Opted out / ineligible Participants Response rate Physician 1 1342 79/ 30 180 13.7% Phys ician 2 180 2 / 0 156 86.7% Spouse 1 178 0 148 83.1% Spouse 2 178 3 145 81.5% Table 4 Physician Response Rates by Recruitment Source (Time 1) Recruitment source Approx. # eligible Participants Response rate USF medical residents 369 35 9.5% USF facu lty 165 28 17.0% HCMA 420 67 16.0% North Collier hospital 236 41 17.4% Online 112 3 2.7% Snowball sample 40 6 15.0% Total 1342 Opted out / ineligible 79 / 30 Grand total 1312 180 13.7% Note. HCMA=Hillsborough County Medical Association The recruitment source for 69 participants could not be verified. Therefore, these individuals were assigned to HCMA and snowball sampling in proportion to the total response rate.

PAGE 48

37 Study Design Data were collected at two time points separated by approximately three months. A three mo nth time lag, although relatively short, was utilized to attempt to examine specific demands (family demands and work demands) that may have been present in the lives of the participants during the course of the study. B ecause a large portion of the partic ipants were expected to have school aged children, a three month time interval over time, a measure of work demands was included in the present study as a statistical control, and was assessed at both time points. Previous crossover studies have empl oyed time delays that have ranged from 2 to 18 months, although limited justification for the length of time chosen has been provided. For example, while Hammer et al. (2005) examined depressive symptoms, WFC, and work family enrichment using a time lag of 12 months, they provided no justification for the length of time selected. On the other hand, Westman, Etzion, et al. (2004) employed a two month time lag in their investigation of the anxiety of unemployed individuals that was necessitated by the frequen Westman, Vinokur, et al. (2004) employed a time lag of 18 months in their investigation of marital dissatisfaction in the Russian Army, and suggested that the time lag was appropriate due to a pendi ng downsizing of the Army.

PAGE 49

38 Measures Provision of Spousal Support The provision of spousal support was measured using 20 items designed to assess emotional concern, instrumental support, informational support, and appraisal support. F ew, if any, instrument s exist that assess all four instrument developed for use in the present study included a combination of pre existing items assessing each dimension of support adapted from Gr eenhaus, Parasuraman, and Wormley (1990), Hill (2005), and King, Mattimore, King, and Adams (1995), and items Six items for each dimension were employed to assess emotional concern an d instrumental support that were adapted from the Family Support Inventory for Workers ( FSIW, King et al., 1995). Four items were employed to assess informational support including two items adapted from the FSIW (King et al., 1995) that originally assesse d emotional concern but were modified to assess informational support, one item adapted informational support. Similarly, four items were employed to assess appraisal s upport including two items adapted from Greenhaus et al. (1990), one item adapted from Hill support. All items were rated using a five social support. To assess the provision of spousal support, items were reworded to focus on the act of providing support as opposed to receiving support. Provision of s pousal support items are provided in Appendix A

PAGE 50

39 Because the provision of spousal support measure was developed for the present study using a combination of pre existing and newly developed items, the psychometric soundness of the measure was assessed. First, the internal consistency reliability of the oefficient alphas for the overall provision of spousal support scale were acceptable across both samples (Physicians Time However, the coefficient alphas for two To investigate if the exc lusion of individual items would increase the reliability of these subscales, i tem analyses were conducted. For example, while the exclusion of item increased the reliability of the appraisal support subscale from .23 to .48, the reliability was still well below acceptable levels. Additionally, when item analyses were conducted to investigate the impact of removing the entire informational and appraisal support subscales from the provision of spousal support scale, the reliability of the total scale actually decreased from .78 to .74. Thus, based on these results item analyses did not suggest that removing any items or subscales would substantially increase the reliability of the overall provision of spousal support scale. T o test the factor structure of the provision of spousal support scale, physician and spouse responses were subjected to exploratory principle factor analyses using promax rotation. For both samples, six factor s were extracted with eigenvalues greater than one. These factors accounted for approximately 60% of the total variance for both physician and spouse responses. Although six factors were identified, an examination of the factor

PAGE 51

40 loadings greater than .40, a s suggested by Ford, MacCallum, and Tait (1986), did not reveal an interpretable factor structure. Additionally, in an attempt to replicate the four factor structure as discussed by House (1981) confirmatory factor analyses were conducted using maximum l ikelihood estimation Specifically, physician and spouse responses were entered into MPlus (Muthen & Muthen, 1998) and a four factor model was specified with each item serving as an indicator of its intended latent construct. Across both samples, when a four factor solution was specified the model estimation did not converge, indicating severe problems with the specified model. Despite repeated attempts to remove individual items that did not fit the intended model; these data did not fit a four factor solution. To examine if the data supported a unitary construct as opposed to the four factor model, a one factor model was specified. Fit statistics indicated the fit of the data to a one factor solution within acceptable parameter s. For example, across both samples, RMSEA values as well as standard root mean square residual values (SRMSR) were approximately .10 indicating a level of fit. Fit statistics for the one factor solution for each sample are provided in Table 5. Thus, results of the confirmatory factor analyses supported keeping the scale as a unitary construct. Therefore, a composite measure of provision of spousal support that included all four subscales was used in the main analyses.

PAGE 52

41 Table 5 Fit Statistics f rom the Confirmatory Factor Analyses for the Provision of Support Scale Sample (Time) N 2 df p< CFI TLI RMSEA SRMSR Physician (1) 175 428.29 170 .001 .65 .61 .09 .09 Physician (2) 150 459.49 170 .001 .62 .58 .11 .10 Spouse (2) 139 404.84 170 .001 .60 .55 .10 .10 Receipt of Spousal Support As mentioned earlier, a limitation of previous social support research is that the receipt and provision of social support have seldom been simultaneously assessed; therefore, the strength of the relationship between these two constructs is relatively unknown. Thus, for exploratory purposes the receipt of spousal support was as sessed using a modified version of the provision of spousal support scale described above. Receipt of s pousal support items are provided in Appendix B Because the receipt of spousal support measure was developed for the present s tudy the psychometric soundness of th is measure was also assessed. Consistent with the results of the provision of spousal support scale, the coefficient alpha for the receipt of spousal support scale was high ( 92) while the reliability of the appraisa l support T o test the factor structure of the receipt of spousal support scale, spouse responses were subjected to an exploratory principle factor analys i s using promax rotation. In this analysis, four facto rs were extracted with eigenvalues greater than one, accounting for 65% of the variance in spouse responses. However, upon examination, factor loadings for the four factor solution identified did not match the four factors identified by House (1981). Facto r loadings for the four factors are provided in Table 6

PAGE 53

42 Table 6 Factor Loadings from the Exploratory Factor Analysis on the Receipt of Support Scale Item (Type) I II III IV My spouse/partner asks me regularly about my day. (E) 0.68 0.36 0.36 0.06 My sp ouse/partner occasionally doesn't want to listen to my problems. (E) *R 0.65 0.37 0.41 0.05 My spouse/partner makes time for me if I need to discuss something. (E) 0.76 0.39 0.48 0.12 When I talk about my day, my spouse/partner doesn't really listen. (E) *R 0.68 0.37 0.48 0.04 When something is bothering me, my spouse/partner shows that he/she understands how I am feeling. (E) 0.79 0.57 0.53 0.10 I have difficulty discussing things with my spouse/partner. (E) *R 0.71 0.37 0.64 0.03 My spouse/partner b urdens me with things that he/she could handle on his/her own. (Inst) *R 0.54 0.44 0.72 0.10 My spouse/partner cooperates with me to get things done around the house. (Inst) 0.80 0.41 0.64 0.41 It seems as if my spouse/partner is always asking me to do s omething for him/her. (Inst) *R 0.40 0.30 0.73 0.02 I can depend on my spouse/partner to help out when I am running late. (Inst) 0.68 0.28 0.58 0.31 If I had to go out of town, my spouse/partner would have a hard time managing household responsibilities (Inst) *R 0.63 0.04 0.65 0.28 When I am having a difficult week, my spouse/partner tries to do more of the work around the house. (Inst) 0.77 0.48 0.58 0.29 My spouse/partner often provides a different way at looking at problems for me. (Inf) 0.49 0.6 5 0.35 0.27 I regularly ask my spouse/partner for advice about a problem. (Inf) 0.37 0.66 0.33 0.09 My spouse/partner sometimes forgets to keep me informed of things I need to know. (Inf) *R 0.60 0.45 0.51 0.19 My spouse/partner keeps me informed about news or events that are occurring. (Inf) 0.56 0.56 0.39 0.28 My spouse/partner recognizes when I do a good job. (A) 0.75 0.46 0.40 0.14 My spouse/partner tends to criticize me when I do something wrong. (A) *R 0.44 0.20 0.52 0.43 My spouse/partner give s me helpful feedback. (A) 0.71 0.75 0.47 0.26 My spouse/partner gives me advice about improving my performance at home when needed. (A) 0.16 0.22 0.08 0.71 Eigenvalue 8.78 1.73 1.26 1.16 Percent of Total Variance 43.88 8.66 6.30 5.79 Note. Factor lo adings in bold represent the largest factor loading for each item. E = Emotional Support Inst = Instrumental Support Inf = Informational Support A = Appraisal Support *R indicates a reverse coded item.

PAGE 54

43 Consistent with the analysis described for the provision of support scale, confirmatory factor analys e s were also conducted on the receipt of spousal support scale. In these analyses spouse responses were entered into MPlus and a four factor and a one factor model were specified. Results of confirmat ory factor analyses suggested that although a four factor model converged, the data fit a unitary model slightly better than the four factor model. Fit statistics for the one factor and the four factor solutions are provided in Table 7. Thus, results of co nfirmatory factor analyses supported keeping the receipt of spousal support scale as a unitary construct. Table 7 Fit Statistics from the Confirmatory Factor Analyses for the Receipt of Support Scale Sample (Factor) N 2 df p< CFI TLI RMSEA SRMSR Spouse (1) 137 426.61 170 .001 .82 .80 .11 .08 Spouse (4) 137 492.85 166 .001 .78 .75 .12 .08 Perceived Family Demands Perceived family demands were assessed using three items based on Aryee et al. (1999). These items asse ss the extent that individuals perceive their family makes demands of them and were rated using a five point Likert scale from The coefficient alphas for this scale obtained in the present study were .84 (Time 1 and Time 2) for physicians and .90 (Time 1) and .88 (Time 2) for spouses. Perceived family demands items are provided in Appendix C Amount o f Time Engaged i n Family Related Activities The amount of time engaged in family related activities was assessed with one open en ded item developed for this study, which is provided in Appendix C

PAGE 55

44 Work Interference with Family Conflict WIF was assessed using the nine item Carlson et al. (2000) scale. Carlson et al. reported acceptable coefficient alphas for the subscales (from .78 to .87). Additionally, three items from the Carlson and Frone (2003) scale were employed to assess internally generated WIF. Carlson and Frone reported an acceptable coeffici point Likert scale from (1) of conflict. The coefficient alpha for the total 12 item scale obtained in the present study was .91 at Time 1 and .92 at Time 2. The time based, strain based, and behavior based conflict subscales demonstrated high internal consistency reliability (from .88 to .91, Time 1; from .87 to .92, Time 2). Similarly, the coefficient alpha for the intern ally generated conflict subscale obtained in the present study was .88 at Time 1 and .89 at Time 2. WIF items are provided in Appendix D Work Interference with Family Conflict: Spouse Report A limitation of previous work family conflict research is that typically only self report data have been collected (Greenhaus et al., 2006; Hammer et al., 2005). Therefore, to supplement the information were asked to provide ratings of the WIF they felt their partner (the physician) experienced. Spouse report WIF was assessed using a modified version of the nine item Carlson et al. (2000) scale and employed the same Likert rating scale. The three Carlson and Frone (2003) items were e xcluded from the spouse report scale due their internal hypotheses proposed above employed self report measures of WIF, analyses were also conducted using spouse report measures of WIF. The coefficient alpha for the spouse

PAGE 56

45 report WIF scale obtained in the present study was .91 at Time 1 and .88 at Time 2. The time based, strain based, and behavior based conflict subscales also demonstrated high internal consistency reliab ility (from .84 to .89, Time 1; from .81 to .89, Time 2). Spouse report WIF items are provided in Appendix E Perceived Work Demands Perceived work demands were assessed using three items based on Aryee et al. (1999). These items assess the extent that i ndividuals perceive their jobs make demands of them and were rated using a five point Likert scale from (1) The coefficient alpha for this scale obtained in the present study was .90 at Time 1 and .87 at Time 2. Perceived work dema nds items are provided in Appendix F Amount o f Time Engaged i n Work Related Activities The amount of time spent engaged in work related activities was assessed with one open ended item developed for this study which is provided in Appendix F Demographic Variables Demographic variables assessed included gender, ethnicity, generation, marital status, length of relationship/marriage, number/age of children; job type, area of practice, and length of practice in years (physicians only); highest level of educ ation obtained, employed in a medical profession, and employed as a physician (spouses/partners only). Demographic items are provided in Appendix G

PAGE 57

46 Procedure Target respondents (i.e. physicians, faculty physicians, and medical residents) were recruited f rom the University of South Florida medical system, Hillsborough County Medical Association (FL), and North Collier Hospital (FL) using a non probability convenience sampling strategy. Additionally, a non probability snowball sampling strategy was used to recruit physicians from various other sources. Due to the of the target respondents were recruited to participate in the study with the assistance of the target respon dents. A n email invitation was sent to physicians (Appendices H and I ) from all recruitment sources, inviting them to participate in the present study. The physician email invitation included (1) a brief description of the study, (2) eligibility criteria, (3) a statement explaining the importance of gathering information from their spouse/partner, (4) a link participants could use to access the survey, (5) a link individuals could use to decline participation in the study, (6) and contact information for t he author. The physician survey included measures of the amount of time engaged in work related activities, perceived work demands, work interference with family conflict, the amount of time engaged in family related activities, perceived family demands, provision of spousal support, and demographic information (Time 1 only). Physicians were asked to enter four pieces of information that allowed their responses. First, physi Second, physicians were asked to provide their own primary email address. Finally,

PAGE 58

47 physicians were asked to provide their own birthday (month and day only, e.g., 05/14) as well as their spou survey was designed so that individuals could not advance to subsequent sections of the survey without providing these four pieces of information. Upon completion of the survey, physicians submitted the survey electronically and an email invitation was sent to their spouse /partner inviting them to participate (Appendix J ). The spouse/partner email invitation included (1) a brief description of the study, (2) a statement explaining the importance of their participation, (3) a link participants could use to access the survey, (4) a link individuals could use to decline participation in the study, (5) and contact information for the author. Approximately one week after each spouse/partner email invitation wa s sent, reminder emails were sent (Appendix K ). The spouse/partner survey included measures of the amount of time engaged in family interference with family conflict, receipt of spousal support (Time 1 only), provision of spousal support (Time 2 only), and demographic information (Time 1 only). Spouses were asked to answer the same four linking questions as physicians (email addresses and birthdates). The spouse/partner survey was also designed so that individuals could not advance to subsequent sections of the survey without providing these four pieces of information. To promote participation at Time 2 of the present study, a method commonly employed in clinical psychology res earch was used, in which monetary donations are made on behalf of the participants to a charity of their choosing. A fter submitting their surveys at Time 1 of the study, participants selected a charity from a list provided (All

PAGE 59

48 Children's Hospital, St. Jud e Children's Research Hospital, or Moffitt Cancer Center). Participants were informed that the charity they chose would receive a donation, as a way Additionally, to promo te participation, individuals were offered a summary of the results. Approximately three months after participants completed the survey at Time 1, individual email invitations were sent inviting their participation at Time 2 (Appendix L ). Approximately one week after each email invitation was sent, reminder emails were sent (Appendix M ). Participation in the current study was voluntary, there were no negative consequences associated with failing to complete the study for participants, and all individual res ponses were kept confidential.

PAGE 60

49 Chapter Three Analyses and Results Preliminary Data Steps time points were linked using responses from the four matching questions describe d earlier (birthdates and email addresses). When inconsistencies in one of the matching questions occurred (e.g., an email address differed between physician and spouse reports) the remaining three matching questions were used. Once matched, d across both time points representing one case in the dataset. This final database was used for all subsequent analyses. Personally identifying information (e.g., emai l addresses and birthdates) was purged from the final database. First, outliers were investigated. For the variables that assessed the number of hours participants spent engaged in family related activities, responses ranged from 0 to 168 hours per week. H owever, because the intent of this item was to capture the amount of time individuals spent engaged in family related activities (e.g., chores, child care), a response of 168 hours per week (24 hours per day for 7 days) was unrealistic. Therefore, these re sponses (N=2) were modified to 130 hours per week to equal the maximum amount of time participants spent engaged in work related activities. No other outliers were detected in the dataset.

PAGE 61

50 After reverse scoring items where appropriate, scale scores were cr eated for each study variable by taking the average of all responses for each measure. Preliminary analyses were conducted to investigate assumptions required of the data when conducting multiple regression analyses. First, the relationship between each pr edictor and criterion normality of the data was investigated by examining the skew and kurtosis of the data. With the exception of family hours which demonstrated a positive skew, all other variables exhibited only minor skew and kurtosis. Finally, the homoscedasticity of errors was investigated by examining a scatter plot of standardiz ed residuals In general, residuals were evenly distributed throughout the d istribution ; thus, no modifications or transformations were made to the data. Next, to ensure that there were no differences across sampling sources or demographic characteristics investigate the homogeneity of covariance matrices. Specifically, there were no differences found in covariance matrices across physician job type ( =80.5, n.s.), gender ( =16.4, n.s.), or gen eration ( =50.2, n.s.), indicating no demonstrable differences across sampling sources. Although no differences were found, slight differences were observed on variables assessing hours spent engaged in family nder. Specifically, while male physicians and female physicians spent nearly equal numbers of hours engaged in paid employment (male: M =62.3, SD =18.8, female: M =61.4, SD =16.8), female physicians spent almost ten hours more per week engaged in family relate d activities ( M =30.1, SD =21.6) than their

PAGE 62

51 male colleagues ( M =20.8, SD 15.4). Therefore, to account for these differences, Although not the main focus of the present study, diff erences across demographic characteristics of the spouse sample were also investigated. As opposed to the physician sample, differences were observed in covariance matrices across M=71.2, p<.01) and if the spouse was employed as a ph However, upon further examination, hours spent engaged in family related activities was the only variable found to differ across categories. For example, while male spouses reported spending an average of 21.8 hours engaged i n family related activities ( SD =16.6); female spouses reported an average of 49.7 hours ( SD =29.3). Furthermore, spouses who were employed as physicians reported spending an average of 26.1 hours engaged in family related activities ( SD =17.3) as compared to 43.6 hours ( SD =31.8) for gender was used as a control variable in analyses where applicable. Descriptive statistics for study variables for physicians and spouses are provided in Table 8 As expected, spouses reported spending more hours engaged in family related activities (Time 1: M =39.6 SD =29.9) than physicians (Time 1: M =23.9 SD =18.1); while physicians reported spending more hours engaged in paid employment (Time 1 : M =61.8 SD =17.9; Time 2: M =61.2 SD =17.6) than spouses (Time 2: M =31.2, SD =25.7). D espite reports) of the WIF they experienced constituted an average level of WIF.

PAGE 63

52 Tab le 8 Descriptive Statistics for Study Variables Physicians Spouses Variable Items Min Max Mean SD Skew Kurt Items Min Max Mean SD Skew Kurt Time 1 Work hours 1 20.0 130.0 61.8 17.9 0.5 1.2 -Work demands 3 1.0 5.0 3.2 0.8 0.1 0.2 0.90 WIF 12 1.0 7.0 4.0 1.0 0.3 0.7 0.91 9 1.0 7.0 4.0 1.2 0.0 0.4 0.91 Family hours 1 0.0 118.0 23.9 18.1 1.7 4.5 -1 1.0 130.0 39.6 29.9 0.9 0.1 -Family demands 3 1.0 4.7 2.2 0.8 0.6 0.3 0.84 3 1.0 5.0 2.7 0.9 0.1 0 .1 0.90 Spousal support* 20 2.6 4.6 3.6 0.4 0.1 0.2 0.78 20 1.7 5.0 3.5 0.6 0.4 0.3 0.92 Time 2 Work hours 1 20.0 120.0 61.2 17.6 0.5 0.8 -1 0.0 105.0 31.2 25.7 0.3 0.7 -Work demands 3 1.3 5.0 3.2 0.8 0.3 0.5 0.87 WIF 12 2.0 7.0 4.0 1.0 0.3 0.2 0.92 9 1.0 7.0 3.9 1.0 0.2 0.2 0.88 Family hours 1 0.0 110.0 24.4 16.0 1.9 7.1 -1 3.0 130.0 41.9 31.7 0.8 0.2 -Family demands 3 1.0 4.7 2.2 0.7 0.4 0.4 0.84 3 1.0 5.0 2.7 0.8 0.1 0.1 0.88 Spousal support* 20 2.3 4. 5 3.5 0.4 0.4 0.1 0.80 20 2.6 4.7 3.9 0.4 0.4 0.4 0.74 Note. Physicians: Time 1 N =180, Time 2 N =156; Spouses: Time 1 N=148, Time 2 N=145. Physicians reported their own WIF at both time points. Spouses provided ratings of the WIF they felt physic ians experienced at both time points. Physicians reported their own provision of spousal support at both time points. Spouses reported their receipt of spousal s upport at Time1 and their own provision of spousal support at Time 2.

PAGE 64

53 Inter Correlations A mon g Variables Inter correlations among physician variables at Times 1 and 2 are provided in Table 9 Consistent with previous work family conflict research (Frone, 200 3 ), there was no relationship between gender and WIF (Time 1: r =.00, n.s.; Time 2: r = .02, n.s.). There were also no gender differences in the amount of support provided (Time 1: r =.08, n.s.; Time 2: r =.01, n.s.). P demands (Time 1: r = .04, n.s.; Time 2: r = .07, n.s.), suggesting th at perceived family demands and family hours may operate differentially and were therefore considered separately. Inter correlations among spouse variables at Times 1 and 2 are provided in Table 10 Consistent with previous research that found that indivi duals who received support were more likely to provide support at a later time ( Bozionelos, 2004; Tyler, 2006), spouses who received spousal support at Time 1 reported providing higher levels of spousal support at Time 2 ( r 1 and Time 2 were strongly related ( r =.74, p<.01), and did not differ by gender (Time 1: r =.05, n.s.; Time 2: r =.10, n.s.) or generation (Time 1: r = .05, n.s.; Time 2: r = .09, n.s.). Contrary to the relationship observed in the phy were moderately related to their perceived family demands (Time 1: r =.32, p<.01; Time 2: r =.38, p<.01). Inter correlations among physician and spouse variables at Times 1 and 2 are provided in Table 11 rts of the extent that their work interfered with their family life was only modestly related to the extent that their spouses reported it did (Time 1: r =.48, p<.01; Time 2: r =.57). These findings suggest that different results may be found when using phys ician or spouse responses.

PAGE 65

54 Table 9 Inter Correlations among Physician Variables Time 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 Gender ** 2 Generation .24** ** 3 # of children .36** .49** ** 4 Work hours .02 ** .19* .18* ** 5 Work demands .07 ** .15 ** .00 ** .43** ** 6 Family hours .25** .01 ** .13 ** .28** .19* ** 7 Family demands .08 ** .10 ** .03 ** .07 ** .32** .04 ** ** 8 WIF .00 ** .13 ** .07 ** .37** .71** .29** .44** ** 9 Support provided .08 ** .01 ** .08 ** .07 ** .38** .24** .34** .53** ** Time 2 10 Work hours .02 ** .17** .13 ** .67** .41** .20** .06 ** .39** .10 ** ** 11 Work demands .07 ** .11 ** .06 ** .39** .78** .22** .28** .69** .39** .56** ** 12 Family hours .15 ** .12 ** .10 ** .19** .20** .69** .11 ** .24** .19** .18** .23** ** 13 Family demands .05 ** .02 ** .07 ** .01 ** .36** .05 ** .66** .41** .35** .12 ** .40** .07 ** ** 14 WIF .02 ** .11 ** .01 ** .34** .59** .28** .37** .80** .56** .52** .74** .23** .47** ** 15 Support provided .01 ** .01 ** .13 ** .13 ** .31** .22** .35** .48** .82** .20** .40** .16 ** .38** .56** Note. N ranges from 145 to 180. Gender: 1=Male, 2=Female. p <.05; ** p<.01.

PAGE 66

55 Table 10 Inter Correlations among Spouse Variables Time 1 1 2 3 4 5 6 7 8 9 10 11 1 Gender ** 2 Generation .12 ** ** 3 # of children .36** .56** ** 4 Family hours .48** .15 ** .29** ** 5 Family de mands .12 ** .13 ** .02 ** .32** ** 6 Spouse report WIF .05 ** .05 ** .11 ** .11 ** .47** ** 7 Support received .01 ** .03 ** .04 ** .08 ** .48** .66** ** Time 2 8 Work hours .50** .10 ** .29** .59** .23** .10 ** .14 ** ** 9 Family hours .48** .24* .26** .75** .29** .15 ** .14 ** .65** ** 10 Family demands .17 ** .20* .10 ** .23** .76** .47** .49** .17* .38** ** 11 Spouse report WIF .10 ** .09 ** .05 ** .05 ** .40** .74** .63** .15 ** .17* .50** 12 Support provided .16 ** .04 ** .16 ** .10 ** .12 ** .19* .31** .14 ** .25** .14 ** .20* Note. N ranges from 100 to 148. Gender: 1=Male, 2=Female. p<.05; ** p<.01.

PAGE 67

56 Table 11 Inter Correlations among Physician and Spouse Variables Time 1 1 2 3 4 5 6 7 8 9 10 1 Work hours (P) ** 2 Work demands (P) .43** ** 3 Family hours (P) .28** .19* ** 4 Family hours (S) .01 ** .01 ** .20* ** 5 Family demands (P) .07 ** .32** .04 ** .10 ** ** 6 Family demands (S) .08 ** .13 ** .01 ** .32** .18* ** 7 WIF (P) .37** .71** .29** .01 ** .44** .18* ** 8 Spouse report WIF (S) .27** .41** .20* .11 ** .17* .47** .48** ** 9 Support provided (P) .07 ** .38** .24** .06 ** .34** .19* .53** 45** ** 10 Support received (S) .04 ** .26** .21* .08 ** .27** .48** .39** .66** .54** ** Time 2 11 Work hours (P) .67** .41** .20* .00 ** .06 ** .19* .39** .35** .10 ** .16 ** 12 Work demands (P) .39** .78** .22** .09 * .28** .15 ** .69** .43** .39** .33** 13 Work hours (S) .03 ** .06 ** .01 ** .59** .13 ** .23** .11 ** .10 ** .05 ** .14 ** 14 Family hours (P) .19* .20** .69** .16 ** .11 ** .05 ** .24** .16 ** .19* .12 ** 15 Family hours (S) .09 ** .09 ** .03 ** .75** .06 ** .29** .07 ** .15 ** .02 ** .14 ** 16 Family demands (P) .01 ** .36** .05 ** .02 ** .66** .14 ** .41** .25** .35** .27** 17 Family demands (S) .13 ** .24** .09 ** .23* .21* .76** .26** .47** .17* .49** 18 WIF (P) .34** .59** .28** .03 ** .37** .19* .80** .45** .56** .40** 19 Spouse report WIF (S) .30** .40** .24** .05 ** .23* .40** .52** .74** .43** .63** 20 Support provided (P) .13 ** .31** .22** .03 ** .35** .22* .48** .47** .82** .61** 21 Support provided (S) .01 ** .00 ** .12 ** .10 ** .09 ** .12 ** .09 ** .19* .17* .31** Note. N ranges from 126 to 180. (P)=Physician variable, (S)=Spouse variable. p<.05, ** p<.01.

PAGE 68

57 Table 11 (Continued) Time 2 11 12 13 14 15 16 17 18 19 20 11 Work h ours (P) ** 12 Work d emands (P) .56** ** 13 Work h ours (S) .11 ** .00 ** ** 14 Family h ours (P) .18* .23* .03 ** ** 15 Family h ours (S) .15 ** .07 ** .65** .07 ** ** 16 Family d emands (P) .1 2 ** .40** .04 ** .07 ** .01 ** ** 17 Family d emands (S) .21* .20* .17* .04 ** .38** .18* ** 18 WIF (P) .52** .74** .08 ** .23** .10 ** .47** .28** ** 19 Spouse r eport WIF (S) .38** .42** .15 ** .26** .17* .25* .50** 57** ** 20 Support p rovided (P) .20* .40** .10 ** .16* .06 ** .38** .18* .56** .55** 21 Support p rovided (S) .03 ** .04 ** .14 ** .04 ** .25** .08 ** .14 ** .08 ** .20* .18* Note. N ranges from 126 to 180. (P)=Physician variable, ( S)=Spouse variable. p<.05, ** p<.01.

PAGE 69

58 Hypotheses 1, 2, and 3 Hypotheses 1, 2, and 3 investigated four proposed theoretical antecedents of the provision of spousal support. These hypotheses were examined simultaneously using one regression analysis. In thi s analysis, physicians provision of spousal support at Time 2 served as the dependent variable. Physician variables assessed at Time 1 (provision of spousal support, perceived family demands, amount of time spent engaged in family related activities and W IF) were entered first. Physician variables assessed at Time 2 (perceived family demands, amount of time spent engaged in family related activities and variable. Results o f the multiple regression analysis conducted to test these hypotheses are provided in Table 1 2 Results of an additional analys i s conducted using spouse report data are provided in Table 1 3 First, Hypothesis 1 predicted that physicians who reported provid ing higher levels of spousal support at Time 1 would report providing higher levels of spousal support at predictive of their provision of spousal support three months la <.001). Thus, Hypothesis 1 was supported. predictive of their provision of spousal support at Time 2. Despite a significant relationship at the correlat ional level ( r = .35, p<.01), regression analyses suggested that .19, .14, p < .05). These findings suggested that

PAGE 70

59 physicians who reported higher levels of perceived family demands also reported providing lower levels of spousal support at the same point in time. Additionally, the relationship between ph n.s.) or synchrono .03, n.s.). In sum, Hypothesis 2 received partial support. spousal support at Time 2 was examined (Hypothesis 3). Although correlation analyses supported a negative relationship ( r = .48, p<.01), regression analyses failed to find .43, p < .001; Time .19, p < .05). These findings suggested that physicians who reported higher levels of WIF also reported providing less spousal support at that time. Thus, Hypothesis 3 received partial support. Results of additional analyses conducted using spouse r eport data were consistent with results using physician report data. Hypotheses 4a and 4b WIF w ould be predictive of higher levels of perceived family demands and family hours for their spouses. This relationship was expected to exist over time (Hypothesis 4a) as well as synchronously (Hypothesis 4b).

PAGE 71

60 These hypotheses were examined using two sets o f regression analyses. In the first amount of time spent engaged in family related activities was used as the dependent variable. Results of multiple regression analyses c onducted to test these hypotheses are provided in Tables 1 4 and 1 5 perceived family demands at Time 2 was examined (Hypothesis 4a). Results indicated (Hypothesis 4a). Results also suggested that there was no lagged relationship between .18, p <.01) in which spouses of male physicians reported spending more hours engaged in fam ily related activities than spouses of female physicians. In sum, Hypothesis 4a received no support. perceived family demands was examined (Hypothesis 4b). Findings indicated tha t 2

PAGE 72

61 sp .18, p <.05) in which spouses of male physicians reported s pending more hours engaged in family related activities than spouses of female physicians. Thus, Hypothesis 4b received partial support. Results of regression analyses conducted using spouse report WIF were generally consistent with analyses using physicia n report WIF and are provided in Tables 1 6 and 1 7 reported WIF and lt may be due to the lower number of spouses who reported their gender, making the analysis less powerful to detect significant effects. Hypotheses 5 and 6 Time 1 and their spouse of spousal support at Time 2. To examine these hypotheses, the Aroian version of the Sobel test of indirect effects was employed (Sobel, 1982). Contrary to the method described by Baron and Ken ny (1986) in which a series of regression analyses are conducted to determine evidence of mediation, the Sobel test provides a significance test of the indirect effect. Specifically, the Sobel test examines if the indirect effect of the independent variabl e on the dependent variable is significantly different from zero, once the mediator is considered (Preacher & Hayes, 2004).

PAGE 73

62 To calculate the Sobel test, four data elements are required: the regression coefficient for the relationship between the independe nt variable and the mediator, the regression coefficient for the relationship between the mediator and the dependent variable (with the IV included in the analysis), and the standard errors of both. The regression coefficients are multiplied together and d ivided by the standard error. The result is then compared to the unit normal distribution (Baron & Kenny, 1986; Preacher & Hayes, 2004; Preacher & Leonardelli, 2006). (1986) method in terms of power and usability (MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002), it is not without its limitations. First, the Sobel test is computed based on the assumption that the product of the two regression coefficients approximates a normal di stribution. However, according to Bollen and Stine (1990), this is not always the case, especially when small samples are employed. Furthermore, the Sobel test has been described as very conservative (MacKinnon, Warsi, & Dwyer, 1995), requiring a large sam ple (e.g., N>500) to detect small effect sizes (Cohen, 1992). To deal with these limitations a bootstrapping methodology as described by Preacher and Hayes (2004), that has become a popular alternative to the Sobel test (Shrout & Bolger, 2002) was also emp loyed. An SPSS macro supplied by Preacher and Hayes (2004) was used to conduct bootstrapping analyses to examine the relationships described in Hypotheses 5 and 6. In this approach, numerous random samples are taken (with replacement) from the original sa mple. The indirect effect for each sample is calculated forming a bootstrap distribution

PAGE 74

63 that is used to calculate a confidence interval. The confidence interval is examined and if it does not include 0, the indirect effect is significantly different from zero (p < .05). sets of regression analyses were conducted. Results of these analyses are provided in Tables 1 8 and 1 9 de .48, p<.001) indicating that physicians who reported higher levels of WIF at Time 1 reported providing lower levels w s Regression coefficients and standard errors were obtained from these two regression an alyses and were used in the Aroian version of the Sobel test of indirect effects (Preacher & Leonardelli, 2006). Results of the Sobel test suggested that the indirect effect of the independent variable on the dependent variable was not significantly differ ent from zero (z=.85, n.s.).

PAGE 75

64 To provide an additional examination of the relationship proposed in Hypothesis 5, the bootstrapping methodology described earlier was employed. Utilizing the SPSS macro supplied by Preacher and Hayes (2004), the relationship proposed in Hypothesis 5 was tested using 5000 bootstrapped samples. Consistent with results of the Sobel test, the 95% confidence interval that resulted from the bootstrapping analysis ( .03, .12) contained 0. Thus, findings of the bootstrapping method we re consistent with the results demands at Time 2. Hypothesis 5 also predicted th provision of spousal support at Time 2. Utilizing the process described above this relationship was tested. Results of the Sobe l test (z=.22, n.s.) and the bootstrapping method ( 2.64, 3.67) suggested that this indirect effect was not significantly different from zero. In sum, Hypothesis 5 received no support. tivities differed by gender, analyses were conducted to test if the relationships examined in by Preacher, Rucker, and Hayes (2007), moderated mediation was tested. Althou gh Preacher et al. describe moderated mediation using five distinct models, model three, that tests for moderation of the relationship between the mediator and the dependent variable was the focus of this analysis. Through the utilization of the SPSS macro the moderating

PAGE 76

65 estimate of the indirect effect specified as well as the conditional indirect effect at each level of the moderator. Because the moderator was dichotomous, results of the test for moderated mediation provided an estimate of the indirect effect for each gender. Findings indicated that gender did not moderate the relationship between the mediator and the dependent variable, as the conditional indirect effect was not significant for male (z=.01, n.s.) or female spouses (z= 1.43, n.s.). support at Time 2. To examine Hypothesis 6, two sets of regression analyses, like those serving as the mediator. Results of these analyses are provided in Tables 20 and 2 1. provision of spousal support at Time 2 was examined. This relationship was not .09, n.s.). .19, p<.05). Therefore, relationship between spous .09, n.s.).

PAGE 77

66 Regression coefficients and standard errors were obtained from these regression analyses and were used in the Sobel test and the bootstrapping method. Results of the Sobel test (z=.91, n.s.) and the bootstrapping method ( .01, .04) suggested that the indirect effect was not significantly different from zero. Hypothesis 6 also predicted that the relatio spousal support at Time 2. Utilizing the process described above this relationship was also tested. First, the relationship between spous and their own family hours at Time 2 was examined. This relationship was significant support at Time 2 reported higher family hours at the same point in time. As the provision of spousal support at Time 2 had already been examined, regression coefficients and standard errors were obtained from these t wo regression analyses and were used in the Sobel test and the bootstrapping method. Despite significant relationships obtained using regression analyses, results of the Sobel test suggested that the indirect effect of the independent variable on the depen dent variable was not significantly different from zero (z=1.64, p<.10) once the mediator was considered. Additionally, the confidence interval provided through the utilization of the bootstrapping method contained 0 ( 3.2, .01) suggesting there was not an indirect effect. Thus Hypothesis 6 was not supported.

PAGE 78

67 Utilizing the moderated mediation analysis described above, analyses were conducted to test if the relationships examined in Hypothesis 6 differed according to spousal support). Results did not support a conditional indirect effect for male (z= .89, n.s.) or female spouses (z= .85, n.s.) effects. Consistent with results described above, no moderating effects were found (male: z= .21, n.s; female: z= 1.05, n.s.).

PAGE 79

68 Table 1 2 Report WIF Provision of spousal support (P2) Step and variable Step 1 Family hours (P1) .04 *** Perceived family demands (P1) .10 *** WIF (P1) .13 *** Support provided (P1) .78*** 2 .68*** Step 2 Family hours (P2) .03 *** Perceived family demands (P2) .14* ** WIF (P2) .19* ** 2 .03** Step 3 Gender (P) .06 *** 2 .00 *** Total R 2 .71 *** Total F 80.25 *** N 147.00 *** Note. egression weights from the final step of the equation. Gender: 1=Male, 2=Female (P) = Physician variable, (1) = Time 1, (2) = Time 2. p < .05, ** p < .01, *** p < .001.

PAGE 80

69 Table 1 3 report WIF Provision of spousal support (P2) Step and variable Step 1 Family hours (P1) .09 *** Perceived family demands (P1) .13 *** Spouse report WIF (S1) .02 *** Support provided (P1) .73*** 2 .71*** Step 2 Family hours (P2) .07 *** Perceived family demands (P2) .18* ** Spouse report WIF (S2) .18* ** 2 .03* ** Step 3 Gender (P) .07 *** 2 .01 *** Total R 2 .74 *** Total F 74.31 *** N 116.00 *** Note. Gender: 1=Male, 2=Female (P) = Physician variable, (S ) = Spouse variable, (1) = Time 1, (2) = Time 2. p < .05, *** p < .001.

PAGE 81

70 Table 1 4 Report WIF (Time 1) Perceived family demands (S2) Family hours (S2) Step and variable Step 1 Family hours (S1) .06 *** .65*** Perceived family demands (S1) .74*** .03 *** 2 .58*** .55*** Step 2 WIF (P1) .10 *** .03 *** 2 .01 *** .00 *** Step 3 Gender (P) .09 *** .18* ** 2 .01 *** .03* ** Overall R 2 .60 *** .57 *** Overall F 90.81 ** 80.79 *** N 125.00 *** 125.00 *** Note. Gender: 1=Male, 2=Female (P) = Physician variable, (S) = Spouse variable, (1) = Time 1, (2) = Time 2. p < .05, *** p < .001.

PAGE 82

71 Table 1 5 Report WIF (Time 2) Perceived family demands (S2) Family hours (S2) Step and variable Step 1 Family hours (S1) .06 *** .66*** Perceived family demands (S1) .74*** .07 *** R 2 .58*** .58*** Step 2 WIF (P2) .13* ** .08 *** 2 .02* ** .01 *** Step 3 Gender (P) .10 *** .18** 2 .01 *** .03** Overall R 2 .61 *** .61 *** Overall F 88.20 *** 89.82 *** N 120.00 *** 120.00 *** Note. the final step of the equation. Gender: 1=Male, 2=Female (P) = Physician variable, (S) = Spouse variable, (1) = Time 1, (2) = Time 2. p < .05, ** p < .01, *** p < .001.

PAGE 83

72 Table 1 6 Report WIF (Time 1) Perc eived family demands (S2) Family hours (S2) Step and variable Step 1 Family hours (S1) .02 *** .67*** Perceived family demands (S1) .70*** .11 *** 2 .60*** .59*** Step 2 Spouse report WIF (S1) .13 *** .01 *** 2 .01 *** .00 *** Step 3 Gender (S) .06 *** .11 *** 2 .00 *** .01 *** Overall R 2 .60 *** .60 *** Overall F 80.15 *** 76.87 *** N 106.00 *** 106.00 *** Note. Gender: 1=Male, 2=Female (S) = Spouse variable, (1) = Time 1, (2) = Time 2. *** p < .001.

PAGE 84

73 Table 1 7 Predicting Sp Report WIF (Time 2) Perceived family demands (S2) Family hours (S2) Step and variable Step 1 Family hours (S1) .01 *** .67*** Perceived family demands (S1) .65*** .07 *** 2 .60*** .59*** Step 2 Spouse repo rt WIF (S2) .25*** .09 *** 2 .05*** .01 *** Step 3 Gender (S) .05 *** .11 *** 2 .00 *** .01 *** Overall R 2 .65 *** .57 *** Overall F 90.37 *** 81.51 *** N 105.00 *** 105.00 *** Note. tion. Gender: 1=Male, 2=Female. (S) = Spouse variable, (1) = Time 1, (2) = Time 2. *** p < .001.

PAGE 85

74 Table 18 Regression Analyses for Mediation: Hypothesis 5 (Physician Report WIF) Provision of spousal support (P2) Step and variable Step 1 WIF (P1 ) .48*** Overall R 2 .23*** Overall F 46.52 *** N 156.00 *** Perceived f amily d emands (S2) Family h ours (S2) Step and variable Step 1 WIF (P1) .23* ** .07 Provision of spousal support (P2) .08 *** .02 Overall R 2 .07** .01 Overall F 5.38 *** .50 N 137.00 *** 137.00 Note. (P) = Physician variable, (S) = Spouse variable, (1) = Time 1, (2) = Time 2. p < .05, ** p < .01, *** p < .001.

PAGE 86

75 Table 1 9 Regression Analyses for Mediation: Hypothesis 5 (Spouse report WIF) Provision of spousal support (P2) Step and variable Step 1 Spouse report WIF (S1) .47*** Overall R 2 .22*** Overall F 36.72 *** N 129.00 *** Perceived f amily d emands (S2) Family h ours (S2) Step and variable Step 1 Spouse report WIF (S1) .52*** .17 Provision of spousal support (P2) .09 *** .03 Overall R 2 .23*** .02 Overall F 18.33 *** 1.51 N 124.00 *** 124.00 Note. quation (P) = Physician variable, (S) = Spouse variable, (1) = Time 1, (2) = Time 2. *** p < .001.

PAGE 87

76 Table 20 Regression Analyses for Mediation: Hypothesis 6 (Physician Report WIF) Provision of spousal support (S2) Step and variable Step 1 WIF ( P1) .09 Overall R 2 .01 Overall F 1.24 N 143.00 Perceived f amily d emands (S2) Family h ours (S2) Step and variable Step 1 WIF (P1) .24** .09 ** Provision of spousal support (S2) .12 ** .26** Overall R 2 .08** .07** Overall F 6.05 ** 5.21 * N 143.00 ** 143.00 ** Note. (P) = Physician variable, (S) = Spouse variable, (1) = Time 1, (2) = Time 2. ** p < .01.

PAGE 88

77 Table 2 1 Regression Analyses for Mediation: Hypothesis 6 ( Spouse Report WIF) Provision of spousal support (S2) Step and variable Step 1 Spouse report WIF (S1) .19* ** Overall R 2 .04* ** Overall F 4.53 *** N 126.00 *** Perceive d f amily d emands (S2) Family h ours (S2) Step and variable Step 1 Spouse report WIF (S1) .45*** .19* ** Provision of spousal support (S2) .09 *** .25** Overall R 2 .23*** .08** Overall F 17.96 *** 5.41 *** N 126.00 *** 126.00 *** Note. (S) = Spouse variable, (1) = Time 1, (2) = Time 2. p < .05, ** p < .01, *** p < .001.

PAGE 89

78 Chapter Four Supplemental Analyses and Results Despite a growing body of research examining the receipt of social support and the crossover of work family conflict, p revious research has been hampered by several issues. First, although the receipt and provision of spousal support are presumed to be related, this relationship is rarely assessed (e.g., Bruck, 2002). Next, despite requests for researchers to supplement se lf report data with information provided from other sources, few researchers have heeded this advice and collected data from dual sources (Greenhaus et al., 2006; Hammer et al., 2005). Finally, although researchers often use measures of WFC and social supp ort that assess multiple dimensions of each construct, most combine items across dimensions to create composite measures (e.g., Aycan & Eskin, 2005; Erdwins et al., 2001; Parasuraman et al., 1992), thereby losing valuable information on the differences acr oss dimensions. Therefore, supplemental analyses were conducted to satisfy three main objectives. First, to augment the limited evidence describing the relationship between the receipt and the provision of spousal support, this relationship was assessed. Second, to determine the strength of the relationship between data provided by different sources, the relationship between physician report and spouse report WIF was investigated. Finally, to assess differences across dimensions of WIF and spousal support, the roles that individual types of WIF and spousal support play in the hypothesized relationships were examined.

PAGE 90

79 Providing and R eceiving S pousal S upport Although the receipt of support is not synonymous with the provision of support, the two constructs are presumed to be related. To investigate this relationship, the spousal support was examined. Consistent with previous research that found a modest relationship between the two variables ( r =.43; Bruck, 2002), support they received at Time 1 ( r =.54, p<.01). To investigate the provision and receipt of spousal support in greater detail, means, standard deviations, and correlations between emotional and instrumental support provided and received were examined and are provided in Table 22 Across both samples, mean levels of emotional support provided and receive d were high (physicians: M =3.79, SD =.59; spouses: M =3.75, SD =.70), while mean levels of instrumental support were lower (physicians: M =3.41, SD =.61; spouses: M =3.38, SD =.81). To investigate if upport provided and tests were conducted. Across both types of support and spousal support overall, there were no significant differences found; suggesting that at a mean level, physicians and their sp ouses agreed on the amount of support provided and received. Next, correlations between types of spousal support were examined. Consistent the emotional support th received were strong ( r =.61, p<.01). Reports of instrumental support provided and

PAGE 91

80 received were also moderately related ( r =.45, p<.01). Thus, these results illustrated a relatively strong relat ionship between the provision and receipt of spousal support at the overall level as well as when individual types of support were examined. These results suggest that future research utilizing different types of spousal support may be fruitful. Due to the unacceptable alpha reliabilities obtained for the informational and appraisal support subscales, supplemental analyses were not conducted using these subscales. The unreliability of these subscales may have been due, in part, to the nature of the construc ts assessed. For example, the items that composed these subscales tended to be somewhat ambiguous and subject to different interpretations (e.g., My spouse/partner often provides a different way of looking at problems for me.). Additionally, the behaviors that were identified in these items may have been difficult to identify for participants. Thus, despite the inclusion of items adapted from preexisting measures, the content of these subscales may have been a contributing factor in the unacceptable reliabi lities obtained. Self R eport and Spouse R As discussed earlier, the relationship between that their work interfered with their family and spouse reports of the same construct were moderately relat ed (Time 1: r =.48, p<.01; Time 2: r =.57). To investigate this relationship at a more fine grained level, differences across types of WIF reported were investigated. Consistent with the investigation described above, means, standard deviations, and correlat ions between of each type of WIF were examined and are provided in Table 2 3

PAGE 92

81 At a mean level, physicians reported experiencing moderate levels of internally generated WIF ( M =4.52, SD =1.13) and time based WIF ( M =4.25, SD =1.22). Although spouses did not repo based WIF ( M =4.52, SD reports, paired samp les t tests were conducted. Results indicated that there were no significant differences across types of WIF reported. At a correlational level, the time based WIF ( r =.56, based WIF ( r =.43, p<.01) was consistent with the relationship reported at the overall level. Finally, a relatively modest relationship was observed based WIF ( r =.25, p<.01). These findings illustrate that physicians and their spouses may have different perspectives on certain variables included in the present study, and suggest that differential results are possib le when using data from different sources. Supplemental A nalyses: Hypotheses 1, 2, and 3 Next, supplemental analyses were conducted to investigate the proposed antecedents of emotional and instrumental support. Two regression analyses were used with each p redicting a different type of spousal support. Physician variables assessed at Time 1 were entered first (individual type of spousal support, perceived family demands, amount of time spent engaged in family related activities and types of WIF). Physician v ariables assessed at Time 2 were entered in the next step (perceived family demands, amount of time spent engaged in family related activities and types of WIF). Physician

PAGE 93

82 gender was entered last as a control variable. Results of the multiple regression a n alyses are provided in Table 24 Consistent with results of Hypothesis 1, across both analyses, the individual type of spousal support examined at Time 1 was a significant predictor of that type of spousal support at Time 2. For example, the provision of e motional support at Time 1 was words, physicians who provided higher levels of emotional support at Time 1 reported providing higher levels of emotional support at Time 2 Next, similar to the results obtained at the overall level for Hypothesis 2, there Time 1 and emotional or instrumental support provided at Time 2. However, a synchronous perceived family demands and their provision of emotional support at Time 2 was .20, p < .05 at Time 2). These findings suggest that physicians who reported higher levels of perceived family demands reported providing lower levels of emotional spousal support at the same point in time. Lastly, supplemental analyses were conducted to examine Hypothesis 3, which support. Contrary to results obtained at the overall level, lagged and synchronous relationship s were found between strain based WIF and emotional support. However, while a negative synchronous relationship was expected, in which higher levels of WIF were expected to be associated with lower provision of spousal support, a positive lagged relationsh ip was also found. Strain based WIF at Time 1was a predictor of emotional

PAGE 94

83 of strain based conflict at Time 1 reported providing higher levels of emotional support at based WIF at Time 2 was predictive of the provision of emotional support at the same .29, p < .05). C ontrary to the lagged results described above, these results suggested that physicians reporting higher levels of strain based conflict at Time 2 reported providing lower levels of emotional spousal support at Time 2. These findings suggest that the relationship between WIF and the provision of spousal support may operate differentially over time and depending on the type of support considered. Thus, results supported both lagged and synchronous relationships between strain based WIF and the provision of emotional spousal support. Supplemental A nalyses : Hypotheses 4a and 4b Supplemental analyses were also conducted to investigate crossover effects amily demands as a control variable. Results of the multiple regression analyses are provided in Tables 2 5 and 2 6 Con trary to results at the overall level, there were no lagged or synchronous family demands. However, consistent with results at the overall level, a main effect for gender

PAGE 95

84 .20, p <.01) in which spouses of male physicians reported spending more hours engaged in family related activities than spouses of female physicians. Next, to investigate if the source of information (physician or spouse) impacted the crossover relationship, supplemental analyses were conducted to investigate the Consistent with the analyses described above, two sets of regression analyses were conducted. Results of these analyses are presented in Tables 2 7 and 2 8 Consistent with results of Hypothesis 4a, there were no lagged relationships found between any of the ouse based WIF (spouse behavior based WIF (spouse report) and sp strain based and behavior based WIF also reported high levels of perceived family demands at the same point in time. Supplemental A nalyses: Hypotheses 5 and 6 Despite results of Hypotheses 5 and 6 that did not support a mediated relationship analyses were conducted investigating these relationship s using different types of WIF and different types of spousal support. Utilizing the methodology described previously, several relationships were tested using the Sobel test and the bootstrapping method.

PAGE 96

85 Although a thorough investigation involving the dif ferent types of spousal support and WIF was conducted, supplemental analyses examining Hypothesis 5 were consistent support did not mediate the relationship between the demands. For example, while there was a strong relationsh s (Time 2) was not significant .04,n.s.). Results of the Sobel test and the bootstrapping method confirmed these null based WIF (Time 1) of spousal support, supplemental analyses provided additional evidence of the direct Finally, despite results of Hy pothesis 6 at the overall level that failed to find port, supplemental analyses were conducted. Contrary to results at the overall level, several mediated relationships were found. Results of regression analyses used to calculate the Sobel test are provided in Table 2 9

PAGE 97

86 First, the relationship between phys report) at Time 1 and spousal support at Time 2. Results of the Sobel test suggested that the indirect effect of the independent variable on the dependent v ariable was significantly different from zero (z=2.06, p<.05). Additionally, the 99% confidence interval provided through the utilization of the bootstrapping methodology did not contain 0 ( 4.92, .03) suggesting there is an indirect effect. Similarly, re sults of the Sobel test (z= 2.15, p<.05) and the 99% confidence interval provided by the bootstrapping method ( 4.22, .09) suggest ed that based WIF (spouse report) at Time 1 based WIF (spouse provision of emotional spousal support at Time 2. Results of the Sobel test (Z= 2.34, p<.05) and the bootstrapping method ( 4.68, .03) support this mediated relationship. with lower levels of support pro vided by their spouses at Time 2. This decreased provision of support was associated with decreased family hours for spouses. In summary, through the utilization of the Aroian version of the Sobel test of indirect effects as well as a bootstrapping methodo logy, evidence of indirect effects w as WIF at Time 1 (spouse report) and their own fa mily hours at Time 2.

PAGE 98

87 Table 22 Means, Standard Deviations, and Correlations: Types of Spousal Support (Time 1) Physician support provided Spouse support received Mean SD Correlatio n Mean SD Spousal support 3.56 0.41 .54** 3.51 0.62 Emotional support 3.79 0.59 .61** 3.75 0.70 Instrumental support 3.41 0.61 .45** 3.38 0.81 Note. Physician: N=180, Spouse: N=141. ** p < .01. Table 2 3 Physician report WIF Spouse report WIF Mean SD Correlation Mean SD WIF 3.99 0.96 .48** 3.96 1.18 Time based WIF 4.25 1.22 .56** 4.52 1.35 Strain based WIF 3.59 1.22 .43** 3.80 1.34 Behavior based WIF 3.62 1.32 .25** 3.57 1.47 Internally generated WIF 4.52 1.13 ---Note. Physici an: N=180, Spouse: N=141. ** p < .01.

PAGE 99

88 Table 2 4 Emotional s upport (P2) Instrumental s upport (P2) Step and variable Step 1 Family h ours (P1) .08 *** .03 *** Perceived f amily d emands (P1 ) .07 *** .00 *** Time based WIF (P1) .02 *** .02 *** Strain based WIF (P1) .23* ** .03 *** Behavior based WIF (P1) .15 *** .16 *** Internally generated WIF (P1) .05 *** .02 *** Emotional s upport (P1) .69*** -*** Instrumental s upport (P1) -*** .64*** R 2 .58*** .60*** Step 2 Family h ours (P2) .10 *** .08 *** Perceived f amily d emands (P2) .20* ** .01 *** Time based WIF (P2) .03 *** .19 *** Strain based WIF (P2) .29* ** .06 *** Behavior based WIF (P2) .08 *** .09 *** Internally generated WIF (P2) .12 *** .03 *** 2 .06** .03 *** Step 3 Gender (P) .08 *** .03 *** 2 .01 *** .00 *** R 2 Total .64 *** .63 *** Overall F 31.96 *** 31.01 *** N 144.00 *** 144.00 *** Note. e final step of the equation. Gender: 1=Male, 2=Fema le (P) = Physician variable, (1) = Time 1, (2) = Time 2. p < .05, ** p < .01, *** p < .001.

PAGE 100

89 Table 2 5 Report WIF (Time 1) Perceived f amily d emands (S2) Family h ours (S2 ) Step and variable Step 1 Family h ours (S1) .06 *** .64*** Perceived f amily d emands (S1) .73*** .01 *** 2 .58*** .55*** Step 2 Time based WIF (P1) .03 *** .07 *** Strain based WIF (P1) .12 *** .06 *** Behavior based WIF (P1) .09 *** .09 *** Internally generated WIF (P1) .07 *** .01 *** 2 .02 *** .01 *** Step 2 Gender (P1) .12 *** .20** 2 .01 *** .03** R 2 Total .61 *** .58 *** Overall F 85.33 *** 81.61 *** N 125.00 *** 125.00 *** Note. he final step of the equat ion. Gender: 1=Male, 2=Female (P) = Physician variable, (S) = Spouse variab le, (1) = Time 1, (2) = Time 2. ** p < .01, *** p < .001.

PAGE 101

90 Table 2 6 Report WIF (Time 2) Perceived f amily d emands (S2) F amily h ours (S2) Step and variable Step 1 Family h ours (S1) .07 *** .67*** Perceived f amily d emands (S1) .73*** .01 *** 2 .57*** .58*** Step 2 Time based WIF (P2) .11 *** .14 *** Strain based WIF (P2) .03 *** .03 *** Behavior based WIF (P2) .04 *** .14 *** Internally generate d WIF (P2) .03 *** .05 *** 2 .02 *** .03 *** Step 2 Gender (P1) .09 *** .20** 2 .01 *** .03** R 2 Total .59 *** .64 *** Overall F 77.88 *** 90.01 *** N 118.00 *** 118.00 *** Note. the final step of the equa tion. Gender: 1=Male, 2=Female (P) = Physician variable, (S) = Spouse variable, (1) = Time 1, (2) = Time 2. ** p < .01, *** p < .001.

PAGE 102

91 Table 2 7 Report WIF (Time 1) Perceived f amily d emands (S2) Fam ily h ours (S2) Step and variable Step 1 Family h ours (S1) .01 *** .67*** Perceived Family Demands (S1) .70*** .11 *** 2 .60*** .59*** Step 2 Time based WIF (S1) .01 *** .01 *** Strain based WIF (S1) .17 *** .04 *** Behavior based WIF (S1) .04 *** .04 *** 2 .02 *** .00 *** S tep 2 Gender (S1) .07 *** .11 *** 2 .00 *** .01 *** R 2 Total .62 *** .60 *** Overall F 78.53 *** 76.80 *** N 106.00 *** 106.00 *** Note. Gender: 1=Male, 2=Female (S) = Spouse var iab le, (1) = Time 1, (2) = Time 2. *** p < .001.

PAGE 103

92 Table 28 Report WIF (Time 2) Perceived f amily d emands (S2) Family h ours (S2) Step and variable Step 1 Family h ours (S 1) .03 *** .69*** Perceived f amily d emands (S1) .62*** .06 *** 2 .60*** .59*** Step 2 Time based WIF (S2) .13 *** .04 *** Strain based WIF (S2) .23** .12 *** Behavior based WIF (S2) .24*** .03 *** 2 .09*** .01 *** Step 2 Gender (S1) .08 *** .13 *** 2 .01 *** .01 *** R 2 Total .69 *** .61 *** Overall F 85.27 *** 78.12 *** N 105.00 *** 105.00 *** Note. the final step of the equation. Gender: 1=Male, 2=Female (S) = Spouse variable, (1) = Time 1, (2) = Time 2. ** p < .01, *** p < .001.

PAGE 104

93 Table 29 Regression Analyses for Mediation: Supplemental Analyses (Hypothesis 6) Step and variable Provision of emotional support (S2) Perceived f amily d emands (S2) Family h ours (S2) Step 1 WIF (S1) .26** 44*** .21* ** Provision of emotional support (S2) -*** .12 *** .26** Overall R 2 .07** .23*** .09** Overall F 9.24 *** 18.58 *** 5.72 *** Step and variable Provision of spousal support (S2) Perceived f amily d emands (S2) Family h ours (S2) Ste p 1 Behavior based WIF (S1) .30** .31** .16 *** Provision of support (S2) -*** .08 *** .26** Overall R 2 .09** .12** .07* ** Overall F 11.82 *** 8.19 *** 4.59 *** Step and variable Provision of emotional support (S2) Perceived f amily d emands (S2) Family h ours (S2) Step 1 Behavior based WIF (S1) .35*** .29** .18 *** Provision of emotional support (S2) -*** .14 *** .27** Overall R 2 .12*** .13*** .07* ** Overall F 17.31 *** 8.99 *** 4.67 *** Note. p of the equation. (S) = Spouse variable, (1) = Time 1, (2) = Time 2. N=126. p < .05, ** p < .01, *** p < .001.

PAGE 105

94 Chapter Five Discussion The present study had four main objectives. First, the relationship between the provision of spousal support as sessed at two time points as well as its theoretical antecedents and consequences was assessed as informed by the conservation of resources theory (Hobfoll, 1989). Results provided support for the relationship between the provision of spousal support acros s two time points separated by approximately three months, as well as the relationship between the provision and receipt of spousal support. Additionally, synchronous and lagged relationships were found between the provision of spousal support, perceived f amily demands, family hours, and WIF. perceived family demands and family hours was investigated. Findings supported on Next, the mediating role of social support as an explanation for the crossover process was examined. Specifically, the role that spousal support plays in mediating the investigated. Although hypothesis testing was unsupportive of the mediating role the provision of spousal support pla ys in the crossover process at the overall level, results did

PAGE 106

95 provide limited support for the mediating role of individual types of spousal support. Additionally, results provided further evidence of the direct effects associated with e crossover process. Lastly, the fourth objective of the present study was to investigate the relationships described above across multiple time points and using dual source data (from physicians and their spouses). Therefore, in addition to main analyses described above, additional analyses were conducted to investigate if relationships differed according to the source of data. Results of supplemental analyses were supportive of results of the main analyses but also illustrated differential relationships depending on the source of the data. Major findings across all four objectives are discussed in detail below. Major Findings: Antecedents and Consequences of Spousal Support According to COR theory (Hobfoll, 1989), an individual in a supportive relationshi p is expected to take a longitudinal approach to conserving resources, and would likely provide social support as a way to make an investment in future resources. Although no previous research had examined the provision of spousal support across two differ ent time points, consistent with COR theory, it was predicted that physicians who provided higher levels of spousal support initially would continue to do so over time. As expected, physicians who reported providing higher levels of spousal support at Time 1 also reported providing higher levels of spousal support three months later at Time 2. These findings were consistent with previous research that found a similar relationship between the receipt of social support assessed over time (Thompson et al., 200 4; Westman, Etzion, et al., 2004); and suggested that the provision of social support is likely to be relatively stable across time. Supplemental analyses examining emotional and

PAGE 107

96 instrumental support were consistent with results of the main analysis. In th ese analyses, type of support at Time 2. For example, physicians who reported providing higher levels of instrumental support at Time 1 (e.g., helping their spouse wit h chores around the house) reported that they continued to provide that type of support at Time 2. Thus, supplemental analyses provided additional support for the stability of the provision of spousal support over time. Next, consistent with COR theory (Ho bfoll, 1989), it was expected that physicians who reported higher levels of perceived family demands at Time 1 would attempt to conserve their resources and would be less likely to provide spousal support at a later time. Contrary to expectations, there wa perceived family demands at Time 1 and their subsequent provision of spousal support. However, synchronous relationships were found. In this case, physicians who reported higher levels of perceived family demands reported providing lower levels of spousal support. Supplemental analyses examining each type of spousal support provided additional support for synchronous relationships. Specifically, higher levels of r levels of their provision of emotional support at the same point in time. Thus, physicians who experienced higher levels of perceived family demands reported providing less support, and specifically emotional support, to their spouses. This relationship may intuitively make sense, in that a physician who is mentally preoccupied with family demands, may have less cognitive resources available to provide emotional support to his/her spouse.

PAGE 108

97 Although a lagged relationship was expected, the timeframe (three m onths) employed in the present study may be a possible explanation for the lack of significant lagged findings. Although the timeframe was prudent in terms of examining demands in the lives of the participants, a three month window between the occurrence o f family demands and the provision of spousal support may have simply been too long. It seems more likely that increased family demands may have a much more immediate relationship with the provision of spousal support, as evidenced by the significant synch ronous relationships. It may be that physicians, who experience a high level of family demands, respond to those demands relatively quickly and attempt to conserve some of their own resources by reducing the amount of support they provide their spouses for a brief period of time. As opposed to the reduction of spousal support three months later, an increase in family demands may momentarily be associated with a drop proce ss a different approach to the measurement of the provision of support might be prudent (e.g., a diary study, or several additional observations). Next, it was expected that physicians who reported high levels of family hours at Time 1, would attempt to c onserve their resources and would report providing lower levels of spousal support at a later time. Contrary to findings reported above, there were no lagged or synchronous relatio nships found between the numbers spent engaged in famil y related activities and the amount of support they provided. This finding may partially be due to the limited amount of time physicians spent engaged in family related activities. Due to the very high number of hours physicians reported spending engaged i n work related activities, the range of responses concerning family

PAGE 109

98 family hours and their provision of spousal support. These results may also be explained due to the type of support considered. Specifically, as opposed to instrumental support (e.g., assistance with household chores) that would clearly involve additional family hours; emotional support may be provided with little change in the hours spent engaged in fa mily related activities. Finally, building on research that has demonstrated that increased WIF has been related to decreased functioning in the family domain (e.g., decreased performance and satisfaction, and increased distress and burnout; e.g., Bachara ch et al., 1991; Frone et al., 1997; Kinnunen & Mauno, 1998; Williams & Alliger, 1994), it was expected that a physician reporting greater WIF would provide less spousal support over time. Contrary to expectations, there was no lagged relationship found be their provision of spousal support at the overall level. However, when supplemental analyses were conducted examining each type of WIF individually, a lagged relationship w as found. Specifically, physicians who reported higher lev els of strain based conflict at Time 1 reported providing higher levels of emotional spousal support at Time 2, as opposed to lower levels as predicted. Although these results were not expected, it is possible that physicians who reported higher levels of strain based conflict, a more emotionally focused element of WIF, may have been more emotionally connected with their spouses. Therefore, these physicians may have been more likely to react to higher levels of WIF at Time 1 by providing more emotional sup port (that is obviously more emotionally focused than other forms of social support). Thus, although these results are contrary to initial expectations,

PAGE 110

99 if physicians experiencing high levels of emotionally draining conflict recognized the impact of that emotional drain on their spouses, they may have responded over time by providing increased emotionally focused support to their spouses. spousal support were found at the overall lev el as well as in supplemental analyses examining each type of WIF individually. Consistent with expectations, and contrary to their provision of spousal support at the s ame point in time. Additionally, higher levels of strain based WIF were associated with lower levels of emotional support provided. As compared to the findings discussed above in which physicians provided higher levels of support after experiencing higher levels of WIF three months earlier, physicians faced with higher levels of WIF reported providing lower levels of spousal support at the same point in time. Thus it appears that that the relationship between WIF and the provision of spousal support may fun ction differentially over time. In summary, several important findings were illustrated by examining Hypotheses 1 3. Specifically, results support that the provision of spousal support is likely to be relatively stable across time; however, the provision o f support was also related to various antecedents with differential results found across time.

PAGE 111

100 Major Findings: Crossover Relationships Adding to the limited research that has examined crossover effects associated with work family conflict, it was expected that related to the perceived family demands and family hours of their spouses. Additionally, to investigate if the source of information (physician or spouse) impacted the crossover relationship, additional analyses were conducted that investigated the crossover of on their own family demands. Contrary to previous research ( Westman & Etzion, 2005) no lagged relationships report or spou se perceived family demands. However, synchronous crossover effects were found. Consistent with expectations at Time 2 explained unique variance in family at 2, regardless of the level of perceived family demands or family hours re ported by ( Frone, 2003; Frone et al., 1992 ) that suggests that conflict impacting one domain is more likely to be related to other variables within the same domain tha n to variables in an alternate domain. Additionally, these findings add support to previous research that has found the crossover of the WIF of one spouse on the family related consequences of the other spouse (Hammer et al., 2005; Matthews et al., 2006). Supplemental analyses that investigated each type of WIF individually, were also supportive of synchronous crossover relationships.

PAGE 112

101 based WIF (spouse based WIF (spouse report) at Time perceived family demands and family hours at Time 1. Thus, it appears that a relatively fam ily demands was found that was consistent across source of information (physician or spouse). as sociated with higher levels of perceived family demands for their spouses, there was no relationship with the number of hours spouses spent engaged in family related activities. perceptions of their family demands as opposed to the time they spend dealing with those demands. However, results did demonstrate gender differences in family hours, in which spouses of male physicians reported spending more hours engaged in family related activities than spouses of female physicians. Despite this difference, results were not and their spouses family hours based on gender. In summary, these findings added support to t he growing body of research related outcomes, perceived family demands. These findings are consistent with previous rese arch that has demonstrated domain specific crossover effects, and illustrated the importance of utilizing information provided from more than one source.

PAGE 113

102 Although researchers have frequently hy pothesized about the mediating role social support plays in the crossover process (Westman et al., 2001; Westman & Vinokur, 1998; Westman, Vinokur, et al., 2004), no research could be found testing this hypothesis. Thus, the current study investigated this mechanism of the crossover process using two distinct pathways. First it was expected that Contrary to expectations, there was no support for the mediating role of spousal support in this relationship; however, results of this analysis do lend additional support to earlier hypotheses. To establish the existence of the relationships indicative of mediation, the relation While results of previous hypotheses provided partial support for this pathway that differed in direction according to time, results of this analysis were supportive of a lagged nega tive relationship. In other words, physicians at Time 1 who reported high levels of WIF reported providing less spousal support at Time 2. Contrary to previous analyses, prov ision of spousal support assessed at Time 1, these analyses examined the relationship Thus it appears that when examined separately, physicians experiencing high level s of WIF at Time 1 were less likely to provide spousal support at Time 2. However, when other factors are included in the analysis, results were different.

PAGE 114

103 y demands at Time 2 was also examined. Contrary to expectations, demonstrated a relati they provide d d it appears that spouses were receiving support. However, it appears that this support wa s not related to the family demands they experience d This result, although unexpected, may be due to the nature of the relationships between physicians and their spouses. As physicians are typically faced with excessive amounts of stressors on a daily basis (Fabri et al., 1989) and reporte d moderate levels of work interference with family conflict and high work hours, they may be left with little time or energy to provide spousal support on a regular basis. As a result, their spouses may be asked to regularly deal with family demands with l ittle assistance. Thus, spouses may become accustomed to dealing with family demands regardless of the level of spousal support they receive. In other words, spouses may accumulate the needed resources to effectively deal with the level of family demands t hey experience regardless of the level of support they receive. Thus, whether physicians provide additional resources (in the form of spousal support) or not, spouses may simply deal with family demands as they occur without relying on spousal support to a lleviate some of these demands.

PAGE 115

104 Hypothesis 4 and add additional support to the cro on the family related variables of their spouses. demands. Ho wever, results do lend additional support to the direct crossover of Next, it was expected that other words, would be associated with higher levels of support provided by their spouses. In turn, this increased provision of spousal support would likely be associated with increased levels of family demands for their spouses. Contrary to expectations, there was no support for the mediating role of spousal support in this relationship at the overall level ; however, direct effects were found. Result (spouse reported providing lower levels of spousal support at Time 2 (as opposed to higher). However, this relationship may be explained when the results of the next analysis are considered.

PAGE 116

105 WIF at Tim their spouses at Time 2. Therefore, it appears that instead of providing more support to phy sicians who experience higher levels of WIF, spouses take on, or at least perceive that they take on additional family demands. provision of spousal support and their own family ho urs. As expected, spouses who provided higher levels of support at Time 2 also reported spending more hours engaged in family related activities. Interestingly, there was no relationship found between ed family demands. In this case, it appears that spouses who provided higher levels of support spent more time engaged in family related activities but did not perceive this time to be an increased demand. In summary, these results illustrated differentia l findings when perceived family demands and family hours were used in the analyses. While a direct relationship was family demands was considered, the relationship was reversed. Specifically, a t and their own family hours, but not their perceived family demands. Thus although no support was found for the mediating potential of spousal support at the overall level, several noteworthy direct effects were found.

PAGE 117

106 As compared to other methods commo nly employed to test for evidence of procedure), the methods employed in the present study are recommended as they provide a direct significance test of the indirect effect s (MacKinnon et al., 2002; Preacher & Hayes, 2004). Therefore, although mediation was not supported at the overall level, the use of these methods provides a high level of confidence in the results obtained. Contrary to results at the overall level, severa l mediated relationships were found when supplemental analyses were conducted examining different types of WIF and (spouse provision of emotional spousal support at Time 2. Additionally, the relationship between based WIF (spouse provision of emotional spousal support at Time 2. Thus, across all three relationships, provided by their spouses at Time 2. This decrea sed provision of support was associated with decreased family hours for spouses. In summary, these results provide support for the role that the provision of spousal support plays in the crossover process and suggests that the provision of support can help hours. These findings make an important contribution to the crossover literature by investigating this hypothesized mechanism of the crossover process.

PAGE 118

107 Implications and Future Rese arch Findings of the present study suggest several important implications and areas of opportunity for research involving social support, work family conflict, and crossover. First, results of the present study provided initial evidence for several antece dents and consequences of the provision of spousal support. F indings suggested that individuals who experienced higher levels of work interference with family conflict and perceived family demands reported providing lower levels of spousal support. Additio nally, results suggested that individuals who reported providing higher levels of spousal support reported spending more time engaged in family related activities. Thus, while a wealth of evidence exists demonstrating the relationship between the receipt o f spousal support and its outcomes ( Beatty, 1996; Burke & Greenglass, 1999; Hill, 2005; Purdom et al., 2006 ), the present study is one of the first to address the provision of spousal support directly. Therefore, rather than focusing on the recipient of so cial support as previous research has (Hobfoll et al., 1994; Pearlin & McCall, 1990) future research should be conducted focusing on the provision of support, its antecedents and consequences. Similar to the work family conflict literature, in which multi ple studies have been conducted in an attempt to build a nomological network surrounding the construct, the same approach could be used to further investigate the provision of social support. Building on the preliminary evidence offered in the present stud y, a wealth of other variables should be examined (e.g., negative affectivity, emotional competence, guilt). Personal characteristics (e.g., dispositional characteristics), which are considered resources according to COR theory (Hobfoll, 1989) and can be h elpful in buffering an individual from stressors, would be useful to examine. Additionally, future research should examine

PAGE 119

108 positive outcomes associated with the provision of support, such as self esteem (Krause & Staw, 2000) or positive social interactions Although a moderate relationship was found between the provision and receipt of spousal support, differential relationships were observed based on the type of spousal support investigated. Thus, these results suggest that future research should be condu cted to examine the relationship between the provision and receipt of social support and other variables of interest at the overall level as well as at the facet level. This research could further help illustrate if certain types of social support have adv antages over others. Next, results of the present study have important implications for work family conflict researchers. While findings of the present study were generally consistent with previous work family conflict research, findings also illustrated that differential results are likely when different types of work family conflict are examined. Therefore, as previously suggested, future research should consider the type of conflict in relationship to the research question. Additionally, organizations m ay be prudent to focus on issues raised in the present study. For example, while the direct relationship between high levels of WIF and detrimental outcomes is well known (e.g., general psychological strain, somatic symptoms, depression, substance abuse, b urnout, work related stress, family related stress; Allen et al., 2000); results of the present study suggested that additional consequences may be likely. Specifically, results of the present study demonstrated that lated affect the physician directly, but it also affects the amount of support the physician likely receives from his/her spouse.

PAGE 120

109 This decreased provision of su pport may have additional detrimental consequences for the physician. Therefore, organizations are encouraged to implement programs designed to alleviate WIF Family supportive policies such as telecommuting (e.g., reading patient x rays from a home computer, or working on administrative paperwork at home), child care, elder care, or information and referral services (Thomas & Ganster, 1995) may be viable options for organizations to he lp reduce the WIF of their employees. Results of the present study also have important implications for crossover research. Although researchers have frequently discussed the mediating role of social support in the crossover process (Westman et al., 2001; Westman & Vinokur, 1998; Westman, Vinokur, et al., 2004), the present study is one of the first to directly test this mechanism. Despite a very thorough investigation of the mediating role of spousal support, only limited evidence was found. These findings are consistent with previous crossover research that has failed to find evidence of the mediating effects of social undermining (Westman & Vinokur, 1998; Westman et al., 2001), and suggest that the direct crossover of stressors and strains may be a more f ruitful avenue of investigation as compared to the indirect pathway. However, it is recommended that future research investigate the mediating role of spousal support using other variables of interest. Study Strengths The present study had several strength s that differentiate it from typical studies conducted in the work family conflict, social support, and crossover literatures. First, contrary to most studies that collect data at a single time point, in the present study data were collected at two time po ints separated by approximately three months. Although

PAGE 121

110 examining variables of interest at multiple time points is beneficial in allowing an investigation of lagged and synchronous relationships, such designs do not automatically permit the discussion of ca usality (Zapf, Dormann, & Frese, 1996). Because all plausible explanations were not controlled for in the current study, it is possible that other factors may be found to explain these results. Next, the present study utilized data collected from both mem bers of a couple across both points in time. Additionally, as opposed to simply collecting self report data from both members of the couple, spouses in the present study were asked to provide reports of certain physician variables. Thus, not only were phys ician and spouse self variables were employed. Furthermore, the present study assessed all of the main variables of interest at both time points, and utilized the sam e measurement method for each variable as suggested by Zapf et al. (1996). Third, contrary to previous social support research that has seldom examined the individual contributions of different dimensions of social support within the same study; the pres investigated emotional concern and instrumental support as well as spousal support at the composite level. Next, the present study investigated a previously untested mechanism of th e crossover process, in which social support was hypothesized to mediate the crossover process (Westman & Vinokur, 1998). Additionally, as compared to other methods commonly employed to test for evidence of mediation that are subject to high levels of Type

PAGE 122

111 study w ere recommended as they provide a direct significance test of the indirect effects (MacKinnon et al., 2002; Preacher & Hayes, 2004). Finally, the sample used in the present study is noteworthy. Consistent with recommendations to investigate crossover effects occurring within couples in which one member of the dyad occupies a high stress occupation (Westman, 2001), the present study investigated crossover effects occur ring between physicians and their spouses. Additionally, the physicians who participated in the present study represented a wide range of positions (faculty physicians, private practice, medical residents, hospital based physicians, etc.) and were employed in a variety of areas of practice (e.g., Pediatrics, Internal Medicine, Surgery, Obstetrics and Gynecology, Psychiatry). Therefore, while the sampling methodology employed did not ensure a representative sample, a fairly diverse group of physicians partic ipated. Additionally, although the majority of physicians identified themselves as Caucasian, a fairly equal number of male and female physicians participated increasing the generalizability of the results. Study Limitations The present study also had sev eral limitations that should be considered. First, physicians who were asked to participate in the current study were recruited from a limited number of organizations. Because a relatively large percentage of participants were recruited from within the Uni versity of South Florida medical system, idiosyncratic characteristics of this organization may limit the generalizability of study results. Next, the response rate obtained in the present study at Time 1 for physicians was well below the expected respons e rate. As mentioned earlier, the unique requirements of the present study may be a plausible explanation. However, it should be mentioned that

PAGE 123

112 the characteristics of those physicians who were willing to participate in a study involving two data collection s, that also involved the recruitment of their spouses, may be different in some way from those who did not. For example, physicians who did not respond may have simply been too busy and therefore did not have the time required to complete the surveys. Alt hough this explanation is plausible, participating physicians reported spending between 20 and 130 hours engaged in work related activities; therefore this explanation does not seem likely. Third, although the sample of physicians who participated was rela tively diverse in terms of age, gender, tenure, position, and area of practice, these individuals were obviously all highly educated and were employed in high income but also highly stressful jobs. Therefore, the generalizability of these results to any ot her group is severely limited. Next, as mentioned previously, problems with the reliability of certain social support subscales limited their effectiveness in the present study. Specifically, although the composite social support scales demonstrated accep table internal consistency, the appraisal and informational subscales did not, and were therefore not included in supplemental analyses. Finally, although several significant relationships were found, the consistency of the variables assessed in the presen t study may have served as a limitation. According to Hammer et al. (2005) highly stable outcome measures may limit the ability to detect significant lagged relationships.

PAGE 124

113 Conclusion The present study examined several theoretical antecedents and consequen ces of the provision of spousal support, as well as the mediating role spousal support plays in the crossover of WIF, across multiple time points and using dual source data. Prior to the present study, limited research had examined the provision of support and no previous research had examined the mediating role spousal support plays in the crossover process. Results provided preliminary evidence for the relationship between the provision of spousal support and several antecedents and consequences; as well as evidence of the crossover of WIF on perceived family demands. However, results were not generally supportive of the mediating role of spousal support in the crossover process. Future research should be conducted to examine additional antecedents and co nsequences of the provision of spousal support as well as the mediating potential of spousal support in the crossover of other stressors and strains. In summary, the present study makes several important contributions to the social support, work family con flict, and crossover literatures by adding to the knowledge of the antecedents and consequences of the provision of spousal support, the growing body of research examining the crossover of WIF, and the understanding of the mediating role of the provision o f spousal support in the crossover process.

PAGE 125

11 4 References Allen, T. D. (2003). Mentoring others: A dispositional and motivational approach. Journal of Vocational Behavior, 62 134 154. Allen, T. D., Herst, D., Bruck, C. S., & Sutton, M. (2000). Consequen ces associated with work to family conflict: A review and agenda for future research. Journal of Occupational Health and Psychology, 5 278 308. Aryee, S. (1992). Antecedents and outcomes of work family conflict among married professional women: Evidence from Singapore. Human Relations, 45 813 837. Aryee, S., Luk, V., Leung, A., & Lo, S. (1999). Role stressors, interrole 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. Aycan, Z., & Eskin, M. (2005). Relative contributions of childcare, spousal support, and organizational support in reducing work family conflict for men and women: The case of Turkey. Sex Roles, 53 453 471. Bacharach, S. B., Bamberger, P., & Conley, S. (1991). Work home conflict among nurses and engineers: Mediating the impact of role stress on burnout and satisfaction at work. Journal of Organizational Behavior, 12 39 53.

PAGE 126

115 Baron, R. M., & Kenny, D. A. (1986). The mode rator mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51 1173 1182. Batt, R., & Valcour, P. M. (2003). Human resources practices as predic tors of work family outcomes and employee turnover. Industrial Relations: A Journal of Economy & Society, 42 189 220. Beach, S. R. H., Martin, J., Blum, T., & Roman, P. (1993). Effects of marital and co worker relationships on negative affect: Testing th e central role of marriage. American Journal of Family Therapy, 21 213 322. Beatty, C. A. (1996). The stress of managerial and professional women: Is the price too high? Journal of Organizational Behavior, 17 233 251. Bedeian, A. G., Burke, B. G., & Moff ett, R. G. (1988). Outcomes of work family conflict among married male and female professionals. Journal of Management, 14 475 491. Beebe, T. J., Locke, R., Barnes, S. A., Davern, M. E., & Anderson, K. J. (2007). Mixing web and mail methods in a survey of physicians. Health Services Research, 42 1219 1234. Beehr, T. A., & Drexler, J. A. (1986). Social support, autonomy, and hierarchical level as moderators of the role characteristics outcome relationship. Journal of Occupational Behaviour, 7 207 214. Bee hr, T. A., Jex, S. M., Stacy, B. A., & Murray, M. A. (2000). Work stressors and coworker support as predictors of individual strain and job performance. Journal of Organizational Behavior, 21 391 405.

PAGE 127

116 Beehr, T. A., King, L. A., & King, D. W. (1990). Socia l support and occupational stress: Talking to supervisors. Journal of Vocational Behavior, 36 61 81. Bernas, K. H., & Major, D. A. (2000). Contributors to stress resistance: Testing a model family conflict. Psychology of Women Quarterly, 2 4 170 178. Beutell, N. J., & Greenhaus, J. H. (1982). Interrole conflict among married women: The influence of husband and wife characteristics on conflict and coping behavior. Journal of Vocational Behavior, 21 99 110. Bolger, N., Delongis, A., Kessler R. C., & Wethington, E. (1989). The contagion of stress across multiple roles. Journal of Marriage and the Family, 51 175 183 Bollen, K. A., & Stine, R. (1990). Direct and indirect effects: Classical and bootstrap estimates of variability. Sociological Methodology, 20 115 140. personality, and mentoring received. Journal of Vocational Behavior, 64 24 46. Brown, G., & Bifulco, A. (1985). Social support, life events and depre ssion. In I. Sarason & B. Sarason (Eds.), Social support: Theory, research and applications (pp. 349 370). The Hague, Netherlands: Martinus Nijhoff. Brown, G., & Harris, T. 1978. The social origins of depression New York: The Free Press. Brown, S. L., Ne sse, R. M., Vinokur, A. D., & Smith, D. M. (2003). Providing social support may be more beneficial than receiving it: Results from a prospective study of mortality. Psychological Science, 14 320 327. Bruck, C. S. (2002). Theoretical antecedents to work fa mily conflict: A comprehensive approach. Unpublished doctoral dissertation, University of South Florida.

PAGE 128

117 Bruck, C. S., & Allen, T. D. (2005). Theoretical antecedents to work family conflict: A comprehensive investigation. Manuscript in preparation. Burke, R. J., & Greenglass, E. R. (1999). Work family conflict, spouse support, and nursing staff well being during organizational restructuring. Journal of Occupational Health Psychology, 4 327 336. Burke, R. J., Weir, T., & DuWors, R. E. (1980). Work demands o n administrators and spouse well being. Human Relations, 33 253 278. Burley, K. A. (1995). Family variables as mediators of the relationship between work family conflict and marital adjustment among dual career men and women. Journal of Social Psychology, 135 483 497. Byron, K. (2005). A meta analytic review of work family conflict and its antecedents. Journal of Vocational Behavior, 67 169 198. Caldwell, R. A., & Reinhart, M. A. (1988). The relationship of personality to individual differences in the us e of type and source of social support. Journal of Social and Clinical Psychology, 6 140 146. Carlson, D.S., & Frone, M. R. (2003). Relation of behavioral and psychological involvement to a new four factor conceptualization of work family interference. Jo urnal of Business and Psychology, 17 515 535. 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.

PAGE 129

118 Carlson, D.S., & Perrew, P. L. (1999). The role of social support in the stressor strain relationship: An examination of work family conflict. Journal of Management, 25 513 540. Chan, C. J., & Margolin, G. (1994). The relationship between dual work mood and home affect. Journal of Social and Personal Relationships, 11 573 586. Cinamon, R. G., & Rich, Y. (2005). Work family conflict among female teachers. Teaching and Teacher Education, 21 365 378. Cobb, S. (1976). Social support as a moderator of life stress. Psychosomatic Medicine, 38 300 314. Cohen, J. (1992). A power primer. Psychological Bulletin, 112 155 159. Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin 98 310 357. Demerout i, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B. (2001). The job demands resources model of burnout. Journal of Applied Psychology, 86 499 512. Demerouti, E., Bakker, A. B., & Schaufeli, W. B. (2005). Spillover and crossover of exhaustion and life satisfaction among dual earner parents. Journal of Vocational Behavior, 67 266 289. Dorio, J. M., Bryant, R. H., & Allen, T. D. (2008). Work related outcomes of the work family interface: Why organizations should care. In K. Korabik, D. S. Lero, & D. L. Whitehead (Eds.), Handbook of work family integration: Research, theory, and best practices (pp. 157 176 ). London: Elsevier.

PAGE 130

119 Dunkel Schetter, C., & Skokan, L. A. (1990). Determinants of social support provision in personal relationships. Journal of Social and Personal Relationships, 7 437 450. Erdwins, C. J., Buffardi, L. C., Casper, W. J. & O'Brien, A. S. (2001). The relationship of efficacy. Family Relations, 50 230 238. Fabri, P. J., Mc Daniel, M. D., Gaskill, H. V., Garrison, R. N., Hanks, J. B., Maier, R. V., et al. (1989). Great expectations: Stress and the medical family. Journal of Surgical Research, 47 379 382. Fischer, C. S. (1982). To dwell among friends: Personal networks in tow n and city Chicago, IL: University of Chicago Press. Ford, J. K., MacCallum, R. C., & Tait, M. (1986). The application of exploratory factor analysis in applied psychology: A critical review and analysis. Personnel Psychology, 39 291 314. Fritz, M. S., & MacKinnon, D. P. (2007). Required sample size to detect the mediated effect. Psychological Science, 18 233 239. Frone, M. R. (2003). Work family balance. In J. C. Quick & L. E. Tetrick (Eds.), Handbook of Occupational Health Psychology (pp. 143 162) Wa shington, DC: American Psychological Association. Frone, M. R., Russell, M., & Cooper, M. L. (1992). Antecedents and outcomes of work family conflict: Testing a model of the work family interface. Journal of Applied Psychology, 77 65 78.

PAGE 131

120 Frone, M. R., Ya rdley, J. K., & Markel, K. S. (1997). Developing and testing an integrative model of the work family interface. Journal of Vocational Behavior 50 145 167. Frye N. K., & Breaugh, J. A. ( 2004 ). Family friendly policies, supervisor support, work family conf lict, family work conflict, and satisfaction: A test of a conceptual model. Journal of Business and Psychology, 19 197 220. Fu, C., & Shaffer, M. A. (2000). The tug of work and family: Direct and indirect domain specific determinants of work family confli ct. Personnel Review, 30 502 522. Ganster, D. C., Fusilier, M. R., & Mayes, B. T. (1986). Role of social support in the experience of stress at work. Journal of Applied Psychology 71 102 110. Gerber, L. A. (1983). Married to their Careers New York, NY: Tavistock Publications. Glaser, D. N., Tatum, C. B., Nebeker, D. M., Sorenson, R. C., & Aiello, J. R. (1999). Workload and social support: Effects on performance and stress. Human Performance, 12 155 176. Greenglass, E. R., Pantony, K. L., & Burke, R.J. ( 1988 ). A gender role perspective on role conflict, work stress, and social support. Journal of Social Behavior and Personality, 3 317 328. 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). Sources of conflict between work and family roles. Academy of Management Review, 10 76 88.

PAGE 132

121 Greenhaus, J. H., Parasuraman, S ., Granrose, C. S., Rabinowitz, S., & Beutell, N. J. (1989). Sources of work family conflict among two career couples. Journal of Vocational Behavior, 34 133 153. Greenhaus, J., Parasuraman, S., & Wormley, W. (1990). Effects of race on organizational expe rience, job performance evaluation and career outcomes. Academy of Management Journal, 33 64 86. Griffith, J. (1985). Social support providers: Who are they? Where are they met? And the relationship of network characteristics to psychological distress. Ba sic and Applied Social Psychology, 6 41 60. Grzywacz, J. G., & Marks, N. F. (2000). Reconceptualizing the work family interface: An ecological perspective on the correlates of positive and negative spillover between work and family. Journal of Occupationa l Health Psychology 5 111 126. Gupta, N., & Jenkins, G. D. (1985). Dual career couples: Stress, stressors, strain, and strategies. In T. A. Beehr & R. S. Bhagat (Eds.), Human stress and cognition in organizations: An integrated perspective (pp. 141 175). New York: Wiley Interscience. Gutek, B., Searle, S., & Klepa, L. (1991). Rational versus gender role explanations for work family conflict. Journal of Applied Psychology, 76 560 568. Haines, V. A., Hurlbert, J. S., & Beggs, J. J. (1996). Exploring the de terminants of support provision: Provider characteristics, personal networks, community contexts, and support following life events. Journal of Health and Social Behavior, 37 252 264.

PAGE 133

122 Hammer, L. B., Allen, E., & Grigsby, T. D. (1997). Work family conflic t in dual earner couples: Within individual and crossover effects of work and family. Jou rnal of Vocational Behavior, 50 185 203. Hammer, L. B., Bauer, T. A., & Grandey, A. A. (2003). Work family conflict and work related withdrawal behaviors. Journal of Business and Psychology, 17 419 436. Hammer, L. B., Cullen, J. C., Neal, M. B., Sinclair, R. R., & Shafiro, M. V. (2005). The longitudinal effects of work family conflict and positive spillover. Journal of Occupational Health Psychology, 10 138 154. Hill E. J. (2005). Work family facilitation and conflict, working fathers and mothers, work family stressors and support. Journal of Family Issues, 26 793 819. Hobfoll, S. E. (1989). Conservation of resources: A new attempt at conceptualizing stress. America n Psychologist, 44 513 524. Hobfoll, S. E., Dunahoo, C. L., Ben Porat, Y. & Monnier, J. (1994). Gender and coping: The dualaxis model of coping. American Journal of Community Psychology, 22 49 82. Hobfoll, S. E., Freedy, J., Lane, C., & Geller, P. (1990) Conservation of social resources: Social support resource theory. Journal of Social and Personal Relationships, 7 465 478. Hobfoll, S. E., & Lerman, M. (1988). Personal relationships, personal attributes, and stress resistance: Mothers' reactions to the ir child's illness. A merican Journal of Community Psychology, 16 565 589.

PAGE 134

123 Hobfoll, S. E., & Stokes, J. P. (1988). The process and mechanics of social support. In S. Duck, D. F. Hay, S. E. Hobfoll, W. Ickes, & B. M. Montgomery, (Eds.), Handbook of personal relationships: Theory, research and interventions (pp. 497 517). Oxford, England: John Wiley & Sons. House, J. S. (1981). Work stress and social support Reading, MA: Addison Wesley. Jones, F. & Fletcher, B. (1993). An empirical study of occupational stre ss transmission in working couples. Human Relations, 46 881 902. Judge, T. A., Boudreau, J. W., & Bretz, Jr., R. D. (1994). Job and life attitudes of male executives. Journal of Applied Psychology, 79 767 782. Kahn, R. L., & Byosiere, P. (1991). Stress i n organizations. In M. D. Dunnette, & L. M. Hough (Eds.), Handbook of Industrial and Organizational Psychology, Vol. 3 (2 nd Ed.), (pp. 571 650). Palo Alto, CA: Consulting Psychologists Press, Inc. Karambayya, R., & Reilly, A. H. (1992). Dual earner couples : Attitudes and actions in restructuring work for family. Journal of Organizational Behavior, 13 585 601. Keene, J. R., & Reynolds, J. R. (2005). The job costs of family demands. Journal of Family Issues, 26 275 299. Kessler, R. C., & McLeod, J. D. (1984 ). Sex differences in vulnerability to undesirable life events. American Sociological Review, 49 620 631. Kessler, R. C., McLeod, J. D., & Wethington, E. (1985). The costs of caring: A perspective on the relationship between sex and psychological distress In I. G. Sarason & B. R. Sarason (Eds.), Social support: Theory, research and applications (pp. 491 506). The Hague: Martinus Nijhoff.

PAGE 135

124 Kim, J. L. S., & Ling, C. S. (2001). Work family conflict of women entrepreneurs in Singapore. Women in Management Revi ew, 16 204 221. King, L. A., Mattimore, L. K., King, D. W., & Adams, G. A. (1995). Family support inventory for workers: A new measure of perceived social support from family members. Journal of Organizational Behavior, 16 235 258. Kinnunen, U., & Mauno, S. (1998). Antecedents and outcomes of work family conflict among employed women and men in Finland. Human Relations, 51 157 177. Krause, N., & Shaw, B. A. (2000). Giving social support to others, socioeconomic status, and changes in self esteem in late life. Jo urnals of Gerontology: Series B: Psychological Sciences and Social Sciences, 55B S323 S333. Lapierre, L. M., & Allen, T. D. (2006) Work supportive family, family supportive supervision, use of organizational benefits, and problem focused coping: Implications for work family conflict and employee well being. Journal of Occupational Health Psychology, 11 169 181. Liang, J., Krause, N. M., & Bennett, J. M. (2001). Social exchange and well being: Is giving better than receiving? Psychology and Aging, 16 511 523. occupational stress? Journal of Occupational Psychology, 60 235 42. Luk, D. M., & Shaffer, M. A. (2005). Work and family domain stressors and support: Withi n and cross domain influences on work family conflict. Journal of Occupational and Organizational Psychology, 78 489 508.

PAGE 136

125 MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002). A comparison of methods to test mediation and o ther intervening variable effects. Psychological Methods, 7 83 104. MacKinnon, D. P., Warsi, G., & Dwyer, J. H. (1995). A simulation study of mediated effect measures. Multivariate Behavioral Research, 30 41 62. Matsui, T., Ohsawa, T., & Onglatco, M. L. (1995). Work family conflict and the stress buffering effects of husband support and coping behavior among Japanese married working women. Journal of Vocational Behavior, 47 178 192. Matthews, R. A., Del Priore, R. E., Acitelli, L. K., & Barnes Farrell, J L., (2006). Work to relationship conflict: Crossover effects in dual earner couples. Journal of Occupational Health Psychology, 11 228 240. ess: A serendipitous, but naturalistic, experiment. Journal of Occupational and Organizational Psychology 70 307 24. Muthen, L. K. and Muthen, B. O. (1998). Mplus User's Guide Los Angeles: Muthen & Muthen. Parasuraman, S., Greenhaus, J. H. & Granrose, C. S. (1992). Role stressors, social support and well being among two career couples. Journal of Organizational Behavior, 13 339 56. Pavett, C. M. (1986). High stress professions: Satisfaction, stress, and well being of spouses of professionals. Human Rel ations, 39 1141 1145.

PAGE 137

126 Pearlin, L. I., & McCall, M. E. (1990). Occupational stress and marital support: A description of micro processes. In J. Eckenrode and S. Gore (Eds.), Stress between Work and Family (pp. 39 60). New York: Plenum. Pinneau, S. R. (1975 ). Effects of social support on psychological and physiological stress. Unpublished doctoral dissertation, University of Michigan. Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. B ehavior Research Methods, Instruments, & Computers, 36 717 731. Preacher, K. J., & Leonardelli, G. J. (2006). Calculation for the Sobel test: An interactive calculation tool for mediation tests. Retrieved August 1, 2007, from University of Kansas Web site : http://people.ku.edu/~preacher/sobel/sobel.htm Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariat e Behavioral Research, 42 185 227. Purdom, C. L., Lucas, J. L., & Miller, K. S., (2006). Couple type, parental status, and the mediating impact of social support. North American Journal of Psychology, 8 1 12. Repetti, R. L. (1989). Effects of daily workl oad on subsequent behavior during marital interaction: The roles of social withdrawal and spouse support. Journal of Personality and Social Psychology, 57 651 659. Rice, R. W., Frone, M. R., & McFarlin, D. B. (1992). Work nonwork conflict and the perceive d quality of life. Journal of Organizational Behavior, 13 155 168.

PAGE 138

127 Roehling, P., Moen, P., & Batt, R. (2003). Spillover. In P. Moen (Ed.), (pp. 101 121). Ithaca, NY: Cornell University Press. Rook, S. K., Dooley, D., & Catalano, R. (1991). Stress transmission: The effects of Journal of Marriage and the Family, 53 165 177. Rosenbaum, M., & Cohen, E. (1999). Equalitarian marriages, spousal support, resourcefulness, and psychologica l distress among Israeli working women. Journal of Vocational Behavior, 54 102 113. Rosin, H. M. (1990). The effects of dual career participation on men: Some determinants of variation in career and personal satisfaction. Human Relations 43 169 182. Ros s, C. E., Mirowsky, J., & Huber, J. (1983). Dividing work, sharing work, and in between: Marriage patterns and depression. American Sociological Review, 48 809 823. Rothbard, N. P. (2001). Enriching or depleting? The dynamics of engagement in work and fam ily roles. Administrative Science Quarterly, 46 655 684. Rudd, N. M., & McKenry, P. C. (1986). Family influences on the job satisfaction of employed mothers. Psychology of Women Quarterly, 10 363 371. Schwarzer, R., & Gutirrez Doa, B. (2005). More spou sal support for men than for women: A comparison of sources and types of support. Sex Roles, 52 523 532. Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods, 7 422 445.

PAGE 139

128 Sobel, M. E. (1982). Asymptotic intervals for indirect effects in structural equations models. In S. Leinhart (Ed.), Sociological methodology (pp. 290 312). San Francisco: Jossey Bass. Sotile, W. M., & Sotile, M. O. (1996). The medical marriage : A couple's survival guide New York: Carol Publishing. Staines, G. L. (1980). Spillover versus compensation: A review on the literature between work and non work. Human Relations, 33 111 129. and marital adjustment: Yet another look. Psychology of Women Quarterly, 3 90 120. Suchet, M., & Barling, J. (1986). Employed mothers: Interrole conflict, spouse support, and marital functioning. Journal of Occupational Behavior 7 167 178. Taubman, R. (1974). Medical marriages. In D. Ashe, E. Nash, and L. Londen (Eds.), Marital and Sexual Counseling in Medical Practice Hagerstown, Maryland: Harper and Row. Thomas, L. T., & Ganster, D. C. (1995). Impact of family supportive wor k variables on work family conflict and strain: A control perspective. Journal of Applied Psychology, 80 6 15. Thompson, C. A., Jahn, E. W., Kopelman, R. E., & Prottas, D. J. (2004). Perceived organizational family support: A longitudinal and multilevel a nalysis. Journal of Managerial Issues, 16 545 565.

PAGE 140

129 Trobst, K. K., Collins, R. L., & Embree, J. M. (1994). The role of emotion in social support provision: Gender, empathy and expressions of distress. Journal of Social and Personal Relationships, 11 45 62 Tyler, K. A., (2006). The impact of support received and support provision on changes in perceived social support among older adults. International Journal of Aging & Human Development, 62 21 38. U.S. Bureau of the Census. (2006). American Community Sur vey Washington, DC: U.S. Government Printing Office. Viswesvaran, C., Sanchez, J. I., & Fisher, J. (1999). The role of social support in the process of work stress: A meta analysis. Journal of Vocational Behavior, 54 314 334. Voydanoff, P. (2005). Conseq uences of boundary spanning demands and resources for work to family conflict and perceived stress. Journal of Occupational Health Psychology, 10 491 503. Westman, M. (2001) Stress and strain crossover. Human Relations, 54 717 751. Westman, M., & Etzion, D. (1995). Crossover of stress, strain and resources from one spouse to another. Journal of Organizational Behavior, 16 169 181. Westman, M., & Etzion, D. (2005) The crossover of work family conflict from one spouse to the other. Journal of Applied Socia l Psychology, 35 1936 1957. Westman, M., Etzion, D., & Danon, E. (2001). Job insecurity and crossover of burnout in married couples. Journal of Organizational Behavior, 22 467 481. Westman, M., Etzion, D., & Horovitz, S. (2004). The toll of unemployment does not stop with the unemployed. Human Relations, 57, 823 844.

PAGE 141

130 Westman, M., & Vinokur, A. D. (1998). Unraveling the relationship of distress levels within couples: Common stressors, empathic reactions, or crossover via social interaction? Human Relations 51 137 156. Westman, M., Vinokur A. D., Hamilton V. L., & Roziner I. (2004). Crossover of marital dissatisfaction during military downsizing among Russian army officers and their spouses. Journal of Applied Psychology, 89 769 779. Wheaton, B. (1985). M odels for the stress buffering functions of coping resources. Journal of Health and Social Behavior, 26 252 364. Williams, K. J., & Alliger, G. M. (1994). Role stressors, mood spillover, and perceptions of work family conflict in employed parents. Academ y of Management Journal, 37 837 868. Zedeck, S., & Mosier, K. L. (1990). Work in the family and employing organization. American Psychologist, 45 240 251. Zapf, D., Dormann, C., & Frese, M. (1996). Longitudinal studies in organizational stress research: A review of the literature with reference to methodological issues. Journal of Occupational Health Psychology, 1 145 169.

PAGE 142

131 Appendices

PAGE 143

132 Appendix A : Provision of Spousal Support Items All items will be rated using a five point Likert scale items are adapted from the Family Support Inventory for Workers (King et al., 1995) unless otherwise noted. Emotional Concern Items 1. I ask my spouse/partner regularly about his/her day. 2. 3. I make time for my spouse/partner if he/she needs to discuss something. 4. lly listen. (R) 5. When something is bothering my spouse/partner, I show that I understand how he/she is feeling. 6. My spouse/partner has difficulty discussing things with me. (R) Instrumental Assistance Items 7. I ask my spouse/partner to help me with things tha t I could handle on my own. (R) 8. I cooperate with my spouse/partner to get things done around the house. 9. It seems as if I am always asking my spouse/partner to do something for me. (R) 10. My spouse/partner can depend on me to help out when he/she is running l ate. 11. If my spouse/partner had to go out of town, I would have a hard time managing household responsibilities. (R) 12. When my spouse/partner is having a difficult week, I try to do more of the work around the house. Informational Support Items 13. I often prov ide a different way of looking at problems for my spouse/partner. 14. My spouse/partner regularly asks me for advice about a problem. 15. I sometimes forget to keep my spouse/partner informed of things he/she needs to know. (R)*H 16. I keep my spouse/partner informed about news or events that are occurring.*D Appraisal Support Items 17. I recognize when my spouse/partner does a good job. *H 18. I tend to criticize my spouse/partner when he/she does something wrong. (R)*D 19. I give my spouse/partner helpful feedback. *G 20. I give my spouse/partner advice about improving his/her performance at home when needed.*G Note. *H item adapted from Hill (2005); *G item adapted from Greenhaus et al. ( 1990); *D

PAGE 144

133 Appendix B : Receipt of Spousal Support Items All items will be rated using a five items are adapted from the Family Support In ventory for Workers (King et al., 1995) unless otherwise noted. Emotional Concern Items 1. My spouse/partner asks me regularly about my day. 2. 3. My spouse/partner makes time for me if I n eed to discuss something. 4. 5. When something is bothering me, my spouse/partner shows that he/she understands how I am feeling. 6. I have difficulty discussing things with my spouse/partner. ( R) Instrumental Assistance Items 7. My spouse/partner burdens me with things that he/she could handle on his/her own. (R) 8. My spouse/partner cooperates with me to get things done around the house. 9. It seems as if my spouse/partner is always asking me to do som ething for him/her. (R) 10. 11. If I had to go out of town, my spouse/partner would have a hard time managing household responsibilities. (R) 12. r tries to do more of the work around the house. Informational Support Items 13. My spouse/partner often provides a different way of looking at problems for me. 14. I regularly ask my spouse/partner for advice about a problem. 15. My spouse/partner sometimes forgets to keep me informed of things I need to know. (R)*H 16. My spouse/partner keeps me informed about news or events that are occurring.*D Appraisal Support Items 17. My spouse/partner recognizes when I do a good job.*H 18. My spouse/partner tends to criticize me when I do something wrong. (R)*D 19. My spouse/partner gives me helpful feedback.*G 20. My spouse/partner gives me advice about improving my performance at home when needed.*G Note. *H item adapted from Hill (2005); *G item adapted from Greenhaus et al. (1990); *D

PAGE 145

134 Appendix C : Family Demands Items Perceived Family Demands Based on Aryee et al., 1999. These items will be rated using a five 1. How often do you feel your family makes too many demands of you? 2. How often do you feel you have too much family related work to do? 3. In general, how often do you feel overwhelmed by the demands of your family? Am ount of Time Engaged in Family Related Activities Items 1. In an average week, how many hours do you spend engaged in family related activities (e.g., chores, child care)?

PAGE 146

135 Appendix D : Work Interference with Family Conflict Items Items 1 9 are taken from Car lson et al., 2000; items 10 12 are modified from Carlson and Frone (2003). All items will be rated using a seven Time based work interference with family 1. My work keeps me from my family activities more than I would like. 2. The time I must devote to my job keeps me from participating equally in household responsibilities and activities. 3. I have to miss family activities due to the amount of time I must spend on work responsibilities. Strain based work interference with family 4. When I get home from work I am too frazzled to participate in family activities / responsibilities. 5. I am so emotionally drained when I get home from wo rk that it prevents me from contributing to my family. 6. Due to all the pressures at work, when I come home I am too stressed to do the things I enjoy. Behavior based work interference with family 7. The problem solving behaviors I use in my job are not effect ive in resolving problems at home. 8. Behavior that is effective and necessary for me at work would be counterproductive at home 9. The behaviors I perform that make me effective at work do not help me to be a better spouse and/or parent. Internally generated w ork interference with family 10. When I am at home, I think about work related problems. 11. When I am at home, I think about things I need to accomplish at work. 12. When I am at home, I try to arrange, schedule, or perform job related activities outside of my normal work hours.

PAGE 147

136 Appendix E : Spouse Report Work Interference with Family Conflict Items Items are modified from Carlson et al., 2000, and will be rated using a seven point Likert Time based work interference with family 1. than he/she would like. 2. The time my spouse/partner must devote to his/her job keeps him/he r from participating equally in household responsibilities and activities. 3. My spouse/partner has to miss family activities due to the amount of time he/she must spend on work responsibilities. Strain based work interference with family 4. When my spouse/part ner gets home from work he/she is too frazzled to participate in family activities/responsibilities. 5. My spouse/partner is so emotionally drained when he/she gets home from work that it prevents him/her from contributing to the family. 6. Due to all the pressu res at work, when my spouse/partner comes home he/she is too stressed to do the things he/she enjoys. Behavior based work interference with family 7. The problem solving behaviors my spouse/partner uses in his/her job are not effective in resolving problems at home. 8. Behavior that is effective and necessary for my spouse/partner at work would be counterproductive at home 9. The behaviors my spouse/partner performs that make him/her effective at work do not help him/her to be a better spouse and/or parent.

PAGE 148

137 Append ix F : Work Demands Items Perceived Work Demands Based on Aryee et al., 1999. These items will be rated using a five 1. How often do you feel your work makes too many demands on you? 2. How often do you feel you have too much job related work to do? 3. In general, how often do you feel overwhelmed by the demands of your work? Amount of Time Engaged in Work Related Activities Items 1. In an average week, how many hours do you spend engaged in paid job or job related work? Please include time spent engaged in paid job or job related work at home.

PAGE 149

138 Appendix G : Demographic Variables What is you gender? Male Female What is your ethnicity? Alaskan Native American Indian A sian Black or African American Caucasian, Non Hispanic Hispanic, Latino, or of Spanish Ancestry Native Hawaiian or Other Pacific Islander Other __________________ What year were you born in? 1982 1995 1966 1981 1954 1965 1942 1953 1925 1941 1924 or earlie r What is your marital status? M arried Living with partner Single How long have you been married/in your current relationship (in years)? Are you employed in a medical profession (spouse/partners only)? Yes No Are you a Physician (spouse/partners only )? Yes No

PAGE 150

139 Appendix G : (Continued) If you have children, please enter the number of children you have in each age group (e.g., if you have a 7 year old child, enter 1 in the box next to the 6 9 years old category. Under 1 year of age 1 2 years old 3 5 yea rs old 6 9 years old 10 14 years old 15 18 years old Over 18 years old What is your highest level of education obtained (spouse/partners only)? Did not graduate high school High school diploma/GED Vocational school Bachelor degree Pro fessional degree (e.g., MD, JD) Ph.D. What is your job type (medical personnel only)? Private practice Faculty physician Medical resident Other: ____________________ What is your area of practice (e.g., Interventional Radiology) (medical personnel only ? How long have you been in practice (in years)? (medical personnel only):

PAGE 151

140 Appendix H : Medical Resident Invitation Email (Time 1) Hello, With the recognition that the divorce rate among physicians is 10% to 20% higher than that of the general populati on, and considering the significant stress experienced by physicians and their families, a collaborative investigation between the University of This investigation involve s an online research study focusing on the role that stress and social support play in the lives of physicians and their spouses/partners. Your participation in this study would be greatly appreciated! To be eligible to participate you must be married or living permanently with your partner. You and your spouse/partner will be asked to complete two very brief surveys (approximately 7 minutes each) one now and one in several months. By participating, you will help make an extremely valuable contribution to the understanding of these issues. If you are willing to participate, please click on the link below: LINK If you are NOT willing to participate, please click on the link below: LINK Please note that you are fully free to not participate in this stud y and may withdraw from this study at any time. No personally identifying information will be used in this study, and only aggregated information will be used in publications. Medical residents who do not participate in this study will not be adversely aff ected in their training program. To thank you for participating in this study, we would like to make a monetary donation to a charity of your choosing, and we would like to offer you a summary of the results from this study (You will be able to select a charity and indicate that you are interested in receiving this summary on the survey itself)! If you have any questions or concerns about this study please contact Jay M. Dorio at jdorio@mail.usf.edu Your contrib ution to our research effort is greatly appreciated! Thank you in advance for your time and consideration! Sincerely, Jay M. Dorio, M.Ed., M.A. Tammy D. Allen, Ph.D. Peter J. Fabri, M.D., Ph.D.

PAGE 152

141 Appendix I : Physician Invitation Email (Time 1) Hello, W ith the recognition that the divorce rate among physicians is 10% to 20% higher than that of the general population, and considering the significant stress experienced by physicians and their families, a collaborative investigation between the University o f This investigation involves an online research study focusing on the role that stress and social support play in the lives of physicians and their spouses/partners. Your participation in this study would be greatly appreciated! To be eligible to participate you must be married or living permanently with your partner. You and your spouse/partner will be asked to complete two very brief surveys (approximately 7 minutes eac h) one now and one in several months. By participating, you will help make an extremely valuable contribution to the understanding of these issues. If you are willing to participate, please click on the link below: LINK If you are NOT willing to partic ipate, please click on the link below: LINK To thank you for participating in this study, we would like to make a monetary donation to a charity of your choosing, and we would like to offer you a summary of the results from this study (You will be able t o select a charity and indicate that you are interested in receiving this summary on the survey itself)! If you have any questions or concerns about this study please contact Jay M. Dorio at jdorio@mail.usf.edu Y our contribution to our research effort is greatly appreciated! Thank you in advance for your time and consideration! Sincerely, Jay M. Dorio, M.Ed., M.A. Tammy D. Allen, Ph.D. Peter J. Fabri, M.D., Ph.D.

PAGE 153

142 Appendix J : Spouse/Partner Invitation Email (T ime 1) Hello, Recently your spouse/partner (at physician e mail address) participated in an online research study and supplied us with your e mail address. This study was initiated by the Psychology, and is designed to investigate the role that stress and social support play in the lives of physicians and their spouses/partners. A major limitation of previous research examining stressors within medical families is a lack of information pr ovided from the spouse/partners of the physicians. Therefore, because your spouse/partner has already participated, your participation is critical to the success of this project! To participate, you will be asked to complete two very brief surveys (approx imately 7 minutes each) one now and one in several months. By participating, you will help make an extremely valuable contribution to the understanding of these issues. If you are willing to participate, please click on the link below: LINK If you are NOT willing to participate, please click on the link below: LINK To thank you for participating in this study, we would like to make a monetary donation to a charity of your choosing, and we would like to offer you a summary of the results from this stu dy (You will be able to select a charity and indicate that you are interested in receiving this summary on the survey itself)! If you have any questions or concerns about this study please contact Jay M. Dorio at jd orio@mail.usf.edu Your contribution to our research effort is greatly appreciated! Thank you in advance for your time and consideration! Sincerely, Jay M. Dorio, M.Ed., M.A. Tammy D. Allen, Ph.D. Peter J. Fabri, M.D., Ph.D.

PAGE 154

143 Appendix K : Spouse/Partner Reminder Email (Time 1) Hello, Recently I contacted you regarding participating in a brief survey designed to investigate the role that stress and social support play in the lives of physicians and their spouses/partners. It is critical to the success of the project to obtain information from both spouses/partners, therefore, because your spouse/partner has already participated, your participation is essential! If you are willing to participate, please click on the link below: LINK If you are NOT willi ng to participate, please click on the link below: LINK If you have any questions or concerns about this study please contact me at jdorio@mail.usf.edu. Thank you in advance for your time and consideration! Sincerely, Jay M. Dorio, M.Ed., M.A. Tammy D. Allen, Ph.D. Peter J. Fabri, M.D., Ph.D.

PAGE 155

144 Appendix L : Physician and Spouse/Partner Invitation Email (Time 2) Hello, Approximately three months ago, you and your spouse/partner participated in Part 1 of a two part online research study initiated by the Medicine and Department of Psychology, designed to investigate the role that stress and social support play in the lives of physicians and their spouses/partners. We would now greatly appreciate your participation in Part 2 of this study. A major limitation of previous research examining stressors within medical families is a lack of information provided over time. Therefore, your participation at this time is critical to the success of this project! To particip ate in Part 2 of the study, you will be asked to complete a very brief survey (approximately 5 minutes) that is very similar to the survey you completed in Part 1. If you are willing to participate, please click on the link below: LINK If you are NOT wi lling to participate, please click on the link below: LINK If you have any questions or concerns about this study please contact me at jdorio@mail.usf.edu. Your contribution to our research effort is greatly appreciated! Thank you for your time and part icipation! Sincerely, Jay M. Dorio, M.Ed., M.A. Tammy D. Allen, Ph.D. Peter J. Fabri, M.D., Ph.D.

PAGE 156

145 Appendix M : Physician and Spouse/Partner Reminder Email (Time 2) Hello, Last week I contacted you regarding participating in a Part 2 of a study designed to investigate the role that stress and social support play in the lives of physicians and their spouses/partners. It is critical to the success of the project to obtain information at both time periods of the study; therefore, because both you and your s pouse/partner have already participated at Time 1, your participation at Time 2 is essential! If you are willing to participate, please click on the link below: LINK If you are NOT willing to participate, please click on the link below: LINK If you ha ve any questions or concerns about this study please contact me at jdorio@mail.usf.edu. Thank you in advance for your time and consideration! Sincerely, Jay M. Dorio, M.Ed., M.A. Tammy D. Allen, Ph.D. Peter J. Fabri, M.D., Ph.D.

PAGE 157

About the Author Ja y M. Dorio serves as a consultant for specializing in survey based organizational development and change. Jay plays an active role in the execution of global survey projects to help organizations drive employee engagement to achieve tangible business outcomes. Prior to Kenexa, Jay gained extensive experience in both internal and external consulting and has worked in a variety of industry segments. Jay has published work on the work family interface and task performance and is a frequent presenter at professional conferences. Jay currently holds a Master of Arts in Industrial and Organizational Psychology from the University of South Florida and a Master of Education in Counselor Training from the University of Massachusetts, B oston. Jay is a member of the Society for Industrial and Organizational Psychology, Society for Human Resource Management and Academy of Management.