USF Libraries
USF Digital Collections

Work-family conflict, eating behaviors, and the role of coping

MISSING IMAGE

Material Information

Title:
Work-family conflict, eating behaviors, and the role of coping
Physical Description:
Book
Language:
English
Creator:
Walvoord, Ashley G
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Work-interference with family
Eating
Fruits
Vegetables
Snacks
Household coping strategies
BMI
Feeding
Modeling
Dissertations, Academic -- Psychology -- Doctoral -- USF   ( lcsh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: There were two primary aims of the present study. The first aim was to examine the relationships between work-interference-with-family (WIF) and specific eating behaviors (eating vegetables, fruits, snack foods) reported by employed mothers, as it relates to health criteria such as BMI. Related to this first aim, household coping strategies were proposed as playing a significant role in the relationship between WIF and eating behaviors. The second aim was to investigate the crossover of WIF to specific child eating behaviors via mother feeding practices or mother eating behaviors. Self-report and other-report survey data were collected from working mothers and their children (recruited from the YMCA Afterschool Program in Hillsborough County), yielding a sample of 262 employed mothers and 238 mother-child dyads.Mother self-report results supported a negative relationship between WIF and mother eating vegetables on work days, but no relationships emerged for eating fruits or snack foods. Regarding the role of coping in the context of the WIF - eating behavior relationship, results were more supportive of a suppression effect than of a moderating effect of coping. There was no support for an indirect relationship between WIF and BMI via eating behaviors. Analysis of the crossover hypotheses revealed support for a negative association between WIF and the mother's feeding practices (monitoring behaviors), but no evidence was found for the hypothesized meditational relationships between mother WIF and child eating behavior (via mother eating and mother feeding) using multisource data. However, the results of supplementary analyses using only mother-report data supported several of the meditational crossover relationships.The results have implications for theoretical development and future research in the growing area of work-family and health. Major findings regarding WIF and specific eating behaviors, coping, and mother vs. child report 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 Ashley G. Walvoord.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 139 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 - 002221166
oclc - 648018840
usfldc doi - E14-SFE0002923
usfldc handle - e14.2923
System ID:
SFS0027240:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Work-Family Conflict, Eating Beha viors, and the Role of Coping by Ashley G. Walvoord A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Tammy D. Allen, Ph.D. Paul Spector, Ph.D. Michael Coovert, Ph.D. Vicky Phares, Ph.D. David Himmelgreen, Ph.D. Date of Approval: March 24, 2009 Keywords: work-interference with family, eating, fruits, vegetables, snacks, household coping strategies, BMI, feeding, modeling Copyright 2009, Ashley G. Walvoord

PAGE 2

Dedication This dissertation is dedicated to my amazing, loving, and unbelievably supportive husband, Garrett. He encouraged me thr ough the all-consuming project execution and accepted my 70 hour work weeks as temporary and necessary for my development. During data analysis, and the composition of a job talk and the wr itten dissertation, my amazing partner bolstered my efforts when I was weary (e.g., accompanied me during daily 5am study sessions at Starbucks, and la te nights in the library for consecutive months). Garrett was instrumental in my ability to keep the ‘big picture’ in site, and he helped me maintain a positive attitude. Above all, Garrett shared my perspective that temporary sacrifices would be met with pers onal growth, joy, and success at the “end of the road” in my doctoral journe y. To my dear husband – you never left my side when life rolled at 120mph, you carried me through times of exhaustion, you believed in me when my confidence waivered, and you celebrated when I experienced great success… now we begin a new journey, and the pa th is bright. Thank you, tha nk you -for the rest of our lives. I dedicate the efforts represente d in this dissertation to my pa rents: my father, for his love and professional guidance, and always-availab le long-distance support; my mother, for her positive attitude, words of encouragemen t, effort to show her love and support

PAGE 3

through countless well-timed long-distan ce actions. Thank you mom and dad for believing in me. I also dedicate this dissertation to Dr. Tammy Allen, who motivated me to challenge myself to pursue excellence and unending pe rsonal and professional development. Tammy led by example, and this example will inspire me for the rest of my life. Thank you for your patience and dedication to th e development and refinement of my professional abilities and character. Finally, this dissertation is dedi cated to God, to whom I attrib ute any ability and talent in all segments of life. My research-based doc toral training provided greater respect for the ability of science to explain major component s of human behavior (genetics, previous experiences, motives, self-efficacy, training and education, extern al constraints, availability of resources cogni tive processing, etc). More im portantly, it revealed to me the presence of God in both the “var iance accounted for” and that which is not accounted for. I am especially thankful for the “une xplained variance” in my experienced successes in my doctoral training.

PAGE 4

Acknowledgements I would like to acknowledge th e generous organizations a nd dedicated individuals who provided their resources and/or services, and who demonstrated commitment to the execution of “The Link Study: Linking Work w ith Family and Community Health.” The following organizations provided gene rous funding, assistance with recruiting research personnel, and access to a hi gh-quality participant population: USF Interdisciplinary Initiative on Sustainable Community in the Humanities and Social Sciences (Provost Khator, Dean Durham a nd Dean Petersen; research funding), Tampa Metro YMCA (Mary Anderson, Maureen “Mo” Chiodini, Pam Burke; permission for participant recruitment, us e of facilities, donation of YMCA passes), USF Office of Undergraduate Research (Dr. Naomi Yavne h, Penny Carlton; assistance with the recruitment of high-quality undergraduate research assistants, undergraduate reimbursement of research-related expenses ). I also thank the Starbucks at 50th and Fowler for their great staff and part icipation in the free refill program. I also wish to recognize the unparalleled contri butions of three gradua te students for their dedication, patience and innumerable hours of work on this research: Stephanie Melton, Mary Martinasek, and Tiffany Belliveau. My sincere appreciation is extended to the advising professors over The Link Study who provided guidance and feedback regarding

PAGE 5

the design of the project, as well as to my respected committee members who contributed to the theoretical development and evaluation in this dissert ation: Consulting faculty on the funded research project: Drs. Tammy Allen, David Himmelgreen, Rita Debate; Dissertation committee members: Drs. Paul Spector, Vicky Phares, Michael Coovert, Alan Balfour. Thank you to the fourteen undergraduate resear ch assistants who vol unteered their time on a weekly basis over a period of one to three semesters: Elizaveta Dolzhenko, Abby Bennet, Birute Gerstner, Monique Latt y, Jenna Poloni, Kim Golazewski, Yasir Abunamous, Mallory Hussin, Amanda Smyth, Alysha Ashbourne, Jennifer Kotwicki, Leatha Luttrell, Nicole Savino, and Tash a Bourget. Finally, five additional undergraduate research assistan ts deserve special recogniti on for demonstrating amazing commitment to quality, learning and initiati ve on The Link Study: Chavely Iglesias, Lori Goodwin, Charleen Maher, Bri ttney Lantzy, Philip McNab, My deepest gratitude is extended to all of th e above entities and indi viduals; without their support, this project would not have been possible.

PAGE 6

i Table of Contents List of Tables ................................................................................................................ ...... v List of Figures ............................................................................................................... ... viii Abstract ...................................................................................................................... ........ ix Chapter One: Introduction .................................................................................................. 1 Chapter Two: The WIF Health Mechanism ........................................................................ 3 Work-Family Conflict ............................................................................................. 3 WFC and Health ..................................................................................................... 5 Physical Health Symptoms ......................................................................... 6 Health-Related Conditions .......................................................................... 6 General Health Status. ................................................................................ 7 Health Behaviors: A Link Between WFC and Health Outcomes ........................... 7 WFC and Eating: A Case for WIF. ............................................................. 8 Eating Behaviors and Obesity ............................................................................... 12 WIF and Coping .................................................................................................... 14 Alternative Roles of Coping ..................................................................... 16 Chapter Three: Crossover of WIF to Child Health ........................................................... 19 Parent Eating and Feeding .................................................................................... 20 Current Study ........................................................................................................ 23

PAGE 7

ii Chapter Four: Method ....................................................................................................... 26 Participants and Sampling ..................................................................................... 26 Measures ............................................................................................................... 30 Procedure .............................................................................................................. 33 Piloting. ..................................................................................................... 33 Recruitment. .............................................................................................. 34 Survey Administration. ............................................................................. 35 Incentives. ................................................................................................. 37 Chapter Five: Results ........................................................................................................ 38 Preliminary Analyses. ........................................................................................... 38 Hypothesized Operationalization of Vari ables: Special Considerations .............. 38 On vs. Off Days in Child-Focused Hypotheses. ....................................... 39 Data Analysis ........................................................................................................ 53 Mother-Focused Hypotheses ................................................................................ 54 Hypotheses 1 and 2. .................................................................................. 54 Hypothesis 3.............................................................................................. 54 Hypotheses 4a and 4b ............................................................................... 57 Hypotheses 5 and 6. .................................................................................. 57 Mother and Child Focused Hypotheses .............................................................. 61 Hypotheses 7 and 8. .................................................................................. 61 Hypotheses 9, 10 and 11. .......................................................................... 66 Chapter Six: Supplemental Results ................................................................................... 77

PAGE 8

iii Alternate Roles of Coping .................................................................................... 78 Mother-Child Perceptions Part I: Simila rity in Mother Report and Child Report 83 Mother-Child Perceptions Part II: Self -Other Similarity Using Single Source Data. ...................................................................................................................... 84 Mother-Child Perceptions Part III: Revisiting Mother and Child-Focused Hypotheses ............................................................................................................ 86 Feeding and child’s eating be haviors (Hypothesis 8a). ............................ 86 WIF and Child Eating Mediated by F eeding Behaviors (Hypothesis 8b). 87 WIF and Child Eating Mediated by Mother Eating (Hypothesis 11). ...... 87 Chapter Seven: Discussion ............................................................................................... 94 Major Findings: WIF and Mother Healt ............................................................... 94 Major Findings: Mother-child WIF and Health Behavior .................................... 97 Supplementary Findings: Moth er-child perceptions ........................................... 102 Study Strengths and Limitations. ........................................................................ 105 Theoretical Implications and Fu ture Research Directions. ................................. 106 Implications for Practice ..................................................................................... 109 Conclusion .......................................................................................................... 110 References .................................................................................................................... ... 112 Appendices .................................................................................................................... .. 131 Appendix A. Mother self-report eati ng behavior items (Work days) ................ 132 Appendix B. Mother self-report eati ng behavior items (Off days) .................... 133 Appendix C. Child self-report eating behavior items (School days) ................. 134

PAGE 9

iv Appendix D. Child self-report eat ing behavior items (Weekends) .................... 135 Appendix E. Mother self-report work-int erference-with-family (WIF) items .. 136 Appendix F. Mother self-report Household Coping Strategies ......................... 137 Appendix G. Mother self-report ite ms Child Feeding Questionnaire ................ 138 Appendix H. Mother demographics ................................................................... 139 About the Author….……………………………………………………………...End Page

PAGE 10

v List of Tables Table 1. Demographic characterist ics of mother pa rticipants ......................................... 28 Table 2. Demographic characterist ics of child participants ............................................. 30 Table 3. Confirmatory factor analysis models: Mother feeding behaviors ..................... 33 Table 4. Normality of study variables.............................................................................. 39 Table 5. Descriptive stat istics: Mother variables (Mother self-report) ............................ 40 Table 6. Descriptive statistics: Mother variables (Child report)* .................................... 41 Table 7. Descriptive statistics: Chil d variables (Child self-report) .................................. 42 Table 8. Descriptive statistics: Ch ild variables (Mother-report)* ................................... 43 Table 9. Intercorrelations (Mother self-repor t and report of child, child self-report) ...... 44 Table 10. Supplementary intercorrela tions (Child report of mother) .............................. 49 Table 11. H3b: WIF and BMI mediated by fru its and vegetables (Mother self-report) .. 55 Table 12. H4b: WIF and BMI mediated by snack foods (Mother self-report) ................ 56 Table 13. H5: Coping as a moderator between WIF and eating vegetables .................... 58 Table 14. H5: Coping as a moderator between WIF and eating fruits ............................ 59 Table 15. H6. Coping as a moderator between WIF and snack foods ............................. 60 Table 16. H8b: Location of results .................................................................................. 61 Table 17. H8b: WIF and child eating fruit mediated by mother monitoring (All selfreport) ....................................................................................................................... 63

PAGE 11

vi Table 19. H8b: WIF and child eating snack foods mediated by m onitoring (All selfreport) ....................................................................................................................... 65 Table 20. H8b: WIF and child eating fruit mediated by pressure (All self-report) ......... 67 Table 21. H8b: WIF and child eating vegetabl es mediated by pressu re (All self-report) 68 Table 22. H8b: WIF and child eating snack foods pressure (All self-report) .................. 69 Table 23. H8b: WIF and child eating fruit me diated by restriction (All self-report) ...... 70 Table 24. H8b: WIF and child eating vegetabl es mediated by restri ction (All self-report) ............................................................................................................................... .... 71 Table 25. H8b: WIF and child eating snack foods mediated by re striction (All selfreport) ....................................................................................................................... 72 Table 26. H11: WIF and child eating fruit mediated by mother eating fruit (All selfreport) ....................................................................................................................... 74 Table 27. H11: WIF and child eating vegetabl es mediated by mother eating vegetables (All self-report) ......................................................................................................... 75 Table 28. H11: WIF and child eating snack f oods mediated by mother eating snack foods (All self-report) ......................................................................................................... 76 Table 29. Supplementary, role of coping: Direct to mother eating vegetables ............... 80 Table 30. Supplementary, role of coping: Direct to monitoring ..................................... 80 Table 31. Supplementary, role of coping: WIF and mother eating vegetables mediated by coping ................................................................................................................... 8 1 Table 32. Supplementary, role of coping: WIF and monitoring mediated by coping .... 82 Table 33. Supplementary, role of coping: Coping and mother eating vegetables mediated by WIF....................................................................................................... 88

PAGE 12

vii Table 34. Supplementary, role of coping: Coping and monitoring mediated by WIF ... 89 Table 35. Supplementary, all mother report: WIF and child eating fruit mediated by monitoring ................................................................................................................. 90 Table 36. Supplementary, all mother report: WIF and child eating vegetables mediated by monitoring ............................................................................................................ 91 Table 37. Supplementary, all mother report: WIF and child eating vegetables mediated by mother eating vegetables ...................................................................................... 92

PAGE 13

viii List of Figures Figure 1. Basic model of WIF-health. ................................................................................ 9 Figure 2. Model of WIF-heal th moderated by coping. ..................................................... 17 Figure 3. Coping as a main effect ..................................................................................... 18 Figure 4. Coping as a partial mediator .............................................................................. 18 Figure 5. Model of WIF-health cross over between parent and child ............................... 24 Figure 6. Supplementary analysis: Alternate roles of coping ........................................... 79

PAGE 14

ix Work-Family Conflict, Eating Beha viors, and the Role of Coping Ashley G. Walvoord ABSTRACT There were two primary aims of the pres ent study. The first aim was to examine the relationships between work-interfere nce-with-family (WIF) and specific eating behaviors (eating vegetables, fruits, snack foods) reported by employed mothers, as it relates to health criteria such as BMI. Related to this firs t aim, household coping strategies were proposed as playing a significant role in the relationship between WIF and eating behaviors. The second aim was to i nvestigate the crossover of WIF to specific child eating behaviors via mother feeding pr actices or mother eating behaviors. Self-report and other-report survey data were collected from working mothers and their children (recruited from the YMCA Afterschool Program in Hillsborough County), yielding a sample of 262 employed mothers a nd 238 mother-child dyads. Mother selfreport results supported a negative rela tionship between WIF and mother eating vegetables on work days, but no relationships emerged for eating fruits or snack foods. Regarding the role of coping in the contex t of the WIF – eating behavior relationship, results were more supportive of a suppression effect than of a m oderating effect of coping. There was no support for an indir ect relationship between WIF and BMI via eating behaviors.

PAGE 15

x Analysis of the crossover hypotheses reve aled support for a negative association between WIF and the mother’s feeding practi ces (monitoring behaviors), but no evidence was found for the hypothesized meditational re lationships between mother WIF and child eating behavior (via mother eating and mother feeding) using multisource data. However, the results of supplementary analyses using only mother-report data supported several of the meditational crossover relationships. The results have implications for theoretical development and future research in the grow ing area of work-family and health. Major findings regarding WIF and speci fic eating behaviors, coping, and mother vs. child report are discussed.

PAGE 16

1 Chapter One: Introduction The last two decades marked the emergen ce of a sizeable body of research that explores the interplay between work and family roles (Eby, Casper, Lockwood, Bordeaux, & Brinley, 2005). The emphasis of these issues in rese arch appropriately mirrors the steadily increasing demands of the contemporary work world -globalized and technologically advanced. The demands of the workplace have clearly manifested in longer work hours, non-traditional work hour s, overtime, and taking work home (e.g., Bond, Thompson, Galinsky, & Prottas, 2003; Br ett & Stroh, 2003). These trends are accompanied by a general increase in the number of adults who work outside of the home, including working mothers (especially mothers of young childre n; Halpern, 2004). Escalating demands and changing structures suggest an inevitable rise in the conflict between work and family responsibilitie s (Bailyn, Drago & Kochan, 2001; Baltes & Heydens-Gahir, 2003), and researchers have responded with considerable effort to identify the antecedents and consequences of work-family conflict (WFC; see reviews, Allen, Herst, Bruck & Sutton, 2000; Byron, 2005; Kossek & Ozeki, 1999). Research has provided ample evidence to support relati onships between physical and psychological health outcomes and WFC, but limited work has focused on how WFC is linked to health. A recent study unveiled the role of eating behavior s in facilitating the sp illover of work to health outcomes (Allen & Armstrong, 2006). Building upon the theory and support presented by Allen and Armstrong, the first aim of this dissertation is to further examine

PAGE 17

2 the role of eating behaviors as a link between work-interference with family (WIF) and health, and determine whether coping strate gies may influence these relationships. To this end, Chapter Two reviews relevant work -family literature w ith regard to role conflict, health and coping, followed by the hypothesized basic WIF-health behavior relationships and the role of coping as a moderator. The second aim of the present study is to extend the theoretical framework from Chapter Two, in response to numerous calls for research to examine how employment issues impact parent and child health (e.g., Cleveland, 2005; Friedman & Greenhaus, 2000; Galambos, Sears, Almeida & Kola ric, 1995; Kinnunen & Pukkinen, 2001). Specifically, crossover of WIF is expected to occur via parent feeding behaviors and corresponding parent and child eating behaviors. In Chapte r Three, support for parentchild crossover is reviewed in the context of WIF and eating behaviors, and a theoretical model of parent-child crossove r relationships is proposed. This effort marks the first examination of the relationship between parent WIF and child health behaviors. The results may provide a link to “the bottom line" wherein employers are convinced to invest in employ ee work-life balance. Scholars typically struggle to persuade organizations that the fina ncial interest of the company is served by prioritizing employee work-life balance, but health insurance is reportedly the most expensive benefit for employers, with the av erage employer paying as much as 77% of the cost of family insurance plans in recent years (Study: Em ployer share of health care costs, 2003; Trend of the month, 2004). S upport for the impact of WIF on health behaviors could provide rati onale for organizations to implement family-supportive policies and benefits to minimi ze employee WIF (Allen, 2001).

PAGE 18

3 Chapter Two: The WIF Health Mechanism Work-Family Conflict Work-family conflict (WFC) occurs when experiences in the work or family domain make it difficult to perform in the ot her domain, or simply when the demands of the two domains are incompatible (Greenhaus & Beutell, 1985). WFC is considered to be bidirectional, such that work demands conflict with the family domain, or family demands may conflict with the work-domain (t ermed ‘work interferen ce with family’ and ‘family interference with work’; WIF and FIW, respectively). There is evidence for the discriminant validity of these construc ts (Mesmer-Magnus & Viswesvaran, 2005), and research suggests that adults experience WIF to a greater degr ee than FIW (e.g., Frone, Russell, & Cooper, 1992b). According to th e domain specificity hypothesis, WIF and FIW generally have distinct antecedents and consequences such that the antecedents of WIF (FIW) usually reside in the work (fam ily) domain, while the consequences of WIF (FIW) often manifest in the family (wor k) domain (Frone, 2003). This domain specificity effect is stronger for WIF than FIW, as research dem onstrates relationships between FIW and some work antecedents and with family outcomes (Byron, 2005; Mesmer-Magnus & Viswesvaran, 2005). Ex tensive reviews of WFC consequences illustrate the penetrating reach of WFC infl uences on work, family, and well-being, such as domain satisfaction, turnover intentions, wo rk absences, performance, mental health,

PAGE 19

4 and physical health (Allen et al., 2000; Eby et al., 2003; Kossek & Ozeki, 1999; MesmerMagnus & Viswesvaran, 2005). Domain specificity can make it difficult to persuade organizations to adopt programs targeted at WIF. Because conseque nces of WIF are not typically experienced by the organization (consequences tend to mate rialize in the family domain), there may seem little reason for employers to address employee WIF issues with intervention or prevention initiatives. By contrast, FIW ha s been shown to negatively impact turnover intentions, absences, and performance (self-rat ings and supervisor ratings; e.g., Allen et al., 2000; Kossek & Ozeki, 1999; Mesmer-Magnus & Viswesvaran, 2005). Accordingly, organizations implementing work-family programs aimed at decreasing FIW (e.g., on-site day care, help with day care costs, elder care assistance, information on community day care, paid parental leave, unpaid parental leave, matern ity or paternity leave with reemployment, and flexible scheduling) report improved organizational performance (Perry-Smith & Blum, 2000). While se veral meta-analyses also demonstrate relationships between WIF and turnover intentions and certain types of absences, the relationship with more convinci ng criteria like job perf ormance is inconsiste nt at best (the few significant relationships are based on self -reported performance, while nonsignificant relationships occur with supervisor ratings or objective ratings). Regardless of direction, research has shown that employees who experi ence work family conflict are more likely to use health care resources (Duxbury & Higgins 2001). Therefore, an alternative route for securing organizational consideration of employee WIF may be via a relationship with employee and family health.

PAGE 20

5 WFC and Health To date, empirical research targets se veral health-related variables: physical health symptoms, health-related conditions, a nd general health status. Studies of physical health symptoms often use symptom checklists or frequency scales to measure symptoms such as headaches, lightheadedne ss, dizziness, sleepiness, dry mouth, chest tightness, insomnia, and sweaty palms. Hea lth-related conditions include blood pressure and overweight/obesity (e.g., body mass index ca lculated from self-report height and weight). Adult general health status is typically assessed with self-reports of overall health or psychological well-being (singl e item, “Overall, how would you rate your health at this time”, or multiple items “to what extent have you experienced/ been …able to concentrate, playing usef ul part, capable of making decisions, under stress, enjoy normal activities, feeling unhappy and depresse d, losing confidence, feeling reasonably happy”). Direct evidence and indirect evidence s upport the relationship between health and WFC. Direct support comes from research employing explicit measures of WFC (bidirectional and directio nal self-report scales). Indirect support is inferred from studies that examine contextual factors, such as part icipation in multiple roles. In general, the measurement of subjective WFC perceptions allows inference about the relationship between the experience of conf lict and other variables. Indirect research examines objective factors which signify involvement in multiple roles (e.g., an adult who has children at home and a full-time job), rather than the perception of role conflict. The following sections review the WFC-health litera ture with respect to direct and indirect evidence for each type of outcome.

PAGE 21

6 Physical Health Symptoms Bi-directional WFC demonstrates a positive relationship with somatic complaints (Allen, et al., 2000; Schmitt, Colligan & Fitzgerald, 1980; Thomas & Ganster, 1995), suggesting that perceptions of role conflict are related to health symptoms, regardless of the domai n in which the conflict originates. Studies distinguishing between WIF and FIW have not yielded consistent results. WIF demonstrates reliable positive relationship s to reported physical symptoms (Burke & Greenglass, 2001; Kinnunen & Mauno, 1998; K litzman, House, Israel & Mero, 1990; Netemeyer, Boles, & McMurrian; 1996). There is general support for a positive relationship between FIW and health sy mptoms (Burke & Greenglass, 2001; Netemeyeret al, 1996; Klitzman, et al., 1990; Grzywacz, 2000), although significant associations are not always observed with females (Kinnunen & Mauno, 1998). While both WIF and FIW exhibit significant relati onships with health symptoms, sometimes WIF is stronger (Burke & Greenglass, 2001; Grzywacz, 2000; Netemeyer, et al., 1996), but at least one study reports that the relationship with ‘non-work’ interference-withwork is stronger than with work-interfere nce-with-‘non-work’ (Klitzman, et al., 1990). The discrepancies between studies are difficu lt to interpret because the researchers did not incorporate any behavioral or perceptual factors that may also be contributing to the reported health symptoms. Health-Related Conditions. Objective health-related conditions such as blood pressure and hypertension are predicted by bi-directional WFC, and by FIW (Frone, Russell, & Cooper, 1997; Thomas & Ganster, 1995). Indirect examination of WFC through participation in multiple roles is asso ciated with decreased blood pressure from daytime to evening (presumed to represent work to home) in women with no children,

PAGE 22

7 while women with children do not experience as large of a decrease in blood pressure (Goldstein, Shapiro, Chicz-DeMet, & Guthri e, 1999). Similarly, women who reported high job strain in addition to a lot of family responsibility had higher blood pressure than women who only reported strain in one role (Brisson, Laflamme, Moisan, Milot, Masse, & Vezina, 1999). Another objec tive health outcome, obesit y, was predicted by WIF, but not FIW (Grzywacz, 2000). General Health Status. Cross-sectional and longit udinal studies yield negative correspondence between overall ratings of physical health and perceived FIW/WIF (Adams & Jex, 1999; Allen & Armstrong, 2006; Frone, Russell & Barnes, 1996; Grandey & Cropanzano, 1999; Mesmer-Magnus & Viswes varan, 2005). As noted with physical health symptoms, there is evidence for str onger relationships be tween WIF and general self-reported health, compared to FIW (Adams & Jex, 1999; Grandey & Cropanzano, 1999; Grzywacz, 2000; Judge, Boudreau & Br etz, 1994). FIW negatively predicted overall health across a four year time la g in the only known study not supporting a relationship between WIF and overall health ra tings (Frone, et al., 1997) This is perhaps attributable to the 4 year tim e lag, as other studies used cr oss-sectional or a shorter lag (five months; Grandey & Cropanzano, 1999). Health Behaviors: A Link Between WFC and Health Outcomes The research evidence for the health out comes just discussed provides guidance for framing the interplay between WFC and h ealth. However, without examination of the links through which WFC leads to these hea lth outcomes, the theory and targets for developing interventions remain elusive. Experts emphasize the need to understand the processes driving WFC-health associations rather than simply reporting simple

PAGE 23

8 relationships between WFC and health out comes (Allen & Armstrong, 2006; Greenhaus, Allen, & Spector, 2006). Examining eating be haviors exemplifies one response to this plea for process consideration, in cluding eating low-fat nutritious foods such as fruits and vegetables, and eating snack foods whose calories offer less nutritional value WFC research has dedicated very little attention to eating behaviors, w ith only several studies addressing them. Of particular interest findings recently reported by Allen and Armstrong (2006) indicate that these behavior s may play an important part in linking WFC with health outcomes. Figure 1 presents the basic WFC-health relationships proposed in the present research. The illustration re presents a portion of the mode l tested by Allen and Armstrong (2006) which hypothesized that eating beha viors and physical activity mediate the relationships between WIF and health outco mes. The first objective of the present research is to qualify the role of eating be haviors in linking WIF with health, and to determine how coping influences the process in Figure 1. What follows is a review of relevant work from the role strain, stre ss, eating, and medical science literatures, accompanied by hypotheses for spec ific paths in the model. WFC and Eating: A Case for WIF. Allen and Armstrong (2006) published the first quantitative examination of WFC and ea ting behaviors. FIW and WIF corresponded with eating fewer “healthy foods ” (i.e., fruits, fiber, and vegetables), while fatty food consumption was related to FIW only. The re lationship with healthy eating was stronger for WIF than FIW, and the authors note th at the association between WIF and healthy food consumption may indicate the influence of WIF on cer tain food choices that are connected to perceptions of time (e.g., preparing fruits or vegetables takes time and

PAGE 24

9 effort). This explanation is plausible cons idering that perceived time scarcity (Jabs & Devine, 2006), long work hours and schedule in flexibility (Byron, 2005; Eby et al., 2003) are known antecedents of WIF. In add ition, eating foods whic h suggest convenience such as ready-to-eat or prepackaged snack foods (e.g., chips, popcorn, granola bars, crackers) might be more likely to correspond wi th WIF than the fatty-food checklist that yielded a non-signifi cant relationship. Figure 1. Basic model of WIF-health. A second quantitative study addressing the link between WFC and health outcomes found that the occurrence of family dinners was negatively related to parent WIF (Allen, Shockley, & Poteat, 2008). This is consistent with Allen and Armstrongs findings, as family dinners have been found to consist of more healthy foods (e.g., vegetables) and less fried food and bad fats (Gillman, Rifas-Shiman, Frazier, Rockett, Camargo, Field, Berkey, & Colditz, 2000). The theme of time scarcity is also a

PAGE 25

10 documented contributor to a reduction in fa mily dinners and “convenience food” habits (Jabs, Devine, Bisogni, Farrell, Jastran & Wethington, 2007). Research that is indirectly related to WFC also indicates th at perceptions of limited time contribute to poor food choices (Hagdrup, Simoes, & Brownson, 1998) such as eating fewer fruits and vegetables (T rieman, Freimuth, Damron, Lasswell, Anliker, Havas, et al., 1996), buying ready-to-eat f oods, and “eating out” more often (Devine, Connors, Sobal, & Bisogni, 2003). It is importa nt to note that shortage of time is only one theme operating through WIF; stress-strai n and behavioral in fluences of WIF on eating are also suggested by related literatures. Beyond the issue of time limitation, the stress literatures suggests that eating in response to emotional stress is a complex reaction, which varies according to emotion (e.g., fear, joy, anger, sadness, tension) and pur pose of eating (e.g., to distract, to relax, to feel better, to satiate hunge r; Macht & Simons, 2000). A dults increase their overall consumption of food, and eat more high fat foods in response to feelings of stress (Cartwright, Wardle, Steggles, Simon, Croker, & Jarvis, 2003; Hellerstedt & Jeffery, 1997; McCann, Warnick, & Knopp, 1990; Ng & Jeffery, 2003; Zellner, Loaisa, Gonzalez, Pita, Morales, Pecora & Wolf, 2006) In other words, perceptions of stress predict the decision to eat and food choice st rategies (Macht & Sim ons, 2000; Zellner, et al., 2006). This yields a familiar outcome: choosing more fast/convenient food, and less fruits and vegetables (Cartw right, et al., 2003; Pak, Olsen, & Mahoney, 2000). Even the stressors that trigger perceive d stress, such as long work hours and experiencing high job demands, have been shown to influence beha viors such as fat intake and food choices,

PAGE 26

11 and outcomes such as weight gain (Devine, Jastran, Jabs, Wethington, Farell, & Bisogni, 2006; Hellerstedt & Jeffery, 1997; Shields, 1999). Turning from research which examines antecedents or components of WIF to that which addresses the perception of negative spillover from work, qualitative research reveals that adults who engage in unhealt hy eating at work (e.g., eating foods with low nutritional value, such as foods from the vending machine) so metimes perceive that these eating habits at work spill over into home life and impact decisions about what to eat, and what to cook for one’s family (Devine, et al., 2003). The employed participants in that research perceived a lack of resources such as time and energy which obstructed healthy food choices. Further, from other qualitativ e work on food choices emerges a glimpse of truth regarding causality (ami dst a sea of inference-limiti ng cross-sectiona l evidence). Employed parents have reported th at food choices were used as a tool to manage negative spillover from work to home, indicating th at food choices involving low-preparation effort were made in response to WIF (Devine et al., 2006). For example, meal preparation was perceived by participants as one more task to be done, and consequently more convenient foods were selected in an e ffort to manage feelings of stress and work fatigue and to reduce time and effort for food. In consideration of the domain-specific hypothesis, the evidence suggesting WIF-eati ng effects, and the anecdotal support for directionality cited above, the present study proceeds with a deliberate focus on the WIF direction of WFC. Eating patterns have been observed to vary between weekdays-weekends or work days-days off (Striegel-Moore, Franko, Thompson, Affenito & Kraemer, 2006; Waterhouse, Edwards & Reilly, 2005) and are likely influenced by perceptions of time

PAGE 27

12 and convenience. For example, foods requi ring preparation are less likely to be consumed on work days. As previously di scussed, eating fruits and vegetables is sometimes perceived to require more time, while ready-to-eat snack foods (e.g., chips, crackers, granola bars) are likely to be pe rceived as requiring virtually no time. Taken together, these issues prompt separate cons ideration of eating behaviors on work and non-work days. In particular, eating fruits and vegetables is more likely to be restricted on work days, whereas eating snack foods may no t vary across work days and days off. H1a. WIF is negatively related to eati ng fruits and vegetables on work days. H1b. WIF is unrelated to eating fruits and vegetab les on days off. H2. WIF is positively rela ted to eating snack foods on work days and weekends. Eating Behaviors and Obesity A fair amount of research in the medi cal sciences literat ure supports the link between eating behaviors and various health outcomes. Although a well-balanced diet includes dietary fat, dietary fat is typically studied in the context of an unhealthy behavior, similar to fast food. The recogni zed consequences of consuming too much dietary fat (typically saturate d and trans-fats) include high BMI, poorer overall health, increased incidence or risk of cardiovascu lar disease, and obes ity (Allen & Armstrong, 2006; Bray & Popkin, 1998; Hu & Willett, 200 2; Oh, Hu, Manson, St ampfer, & Willett, 2005). Similarly, fast food, food eaten away from home, snacks and convenience food are positively related to weight gain, body fa t, and BMI (Burke, Beilin, Durkin, Stritzke, Houghton, & Cameron, 2006; Gillis & Ba r-Or, 2003; Niemeier, Raynor, Lloyd-

PAGE 28

13 Richardson, Rogers & Wing, 2006; Thompson, Ballew, Resnicow, Must, Bandini, & Dietz, 2004), although one study found that a fr equency operationalizat ion of fast food was not related to overweight status (Fre nch, Story, Neumark-Sztainer, Fulkerson & Hannan, 2001). Snack foods have predicted ri sk for obesity and waist circumference (Fisher & Birch, 2002; McCarthy, Robson, Li vingstone, Kiely, Flynn, Cran, & Gibney, 2006). Fruit and vegetable consumption has been linked to reduced insomnia and weight gain; lower risks for obesity, cancer, stroke hypertension, diverticulosis, and coronary heart disease; fewer instances of cataracts; and better self-ratings of overall health (Allen & Armstrong, 2006; Block, Patterson & Subar, 1992; He, Hu, Colditz, Manson, Willett, & Liu, 2004; Hirayama, 1994; Liu, Manson, Lee, Cole, Hennekens, Willett, & Buring, 2000; Steinmetz & Potter, 1996; Va n Duyn & Pivonka, 2000). Given the solid support for the association eating behaviors wi th weight and body fat, BMI is a valuable health-related outcome. It is targeted in the present examination of the WIF-health mechanism. The following hypotheses are proposed: H3a. Consumption of fruits and vegetables on work days is negatively related to BMI. H3b. Consumption of eating fruits and vegetables on work days mediates the relationship between WIF and BMI. H4a. Consumption of snack foods, ir respective of day, predicts BMI. H4b. Consumption of snack foods, irrespective of day, mediates the relationship between WIF and BMI.

PAGE 29

14 WIF and Coping Building upon the basic relationships hypothesi zed in Figure 1, a key factor that is theoretically likely to affect the WIF-health pr ocess is coping. Resear ch shows that in the midst of perceived conflict employees attempt to satisfy demands from conflicting domains in an effort to reduce work-fam ily conflict (Voydanoff, 2002). Coping, defined as “cognitive and behavioral efforts to manage demands that are appraised as taxing or exceeding the resources of the person” (Lazarus & Folkman, 1984, p. 141), encompasses behaviors aimed at altering a stressful contex t (problem-focused copi ng) and attempts to cognitively readjust or regulat e emotional stress (emotion-focu sed coping). Coping with work-related stress has garnered appreciable research attention in the I/O-OB literature, with evidence from several studies sugges ting that coping strategies may be more effective in alleviating distress in domestic or family domains compared with the work domain (Menaghan & Merves, 1984; Perlin & Schooler, 1978; Shi nn, Rosario, Morch & Chestnut, 1984). This trend i ndicates that coping has the ca pacity to play a meaningful role with family domain variables such as eating behaviors an d health outcomes. Parkes (2000) identified two primary f unctions of coping that are observed in stress-outcome relationships: main effects and interactive effects. Coping main effects are a common hypothesis in research, typically specifying that coping and the outcome are related, and that this relationship is not affected by stress. In teractive effects are manifested in moderation hypotheses, wher e coping affects the strength of the relationship between stressor a nd outcome. Some researcher s argue that inconsistent evidence exists for how coping fits as a mode rator of the relationship between stressors and their outcomes (Fortes-Ferreira, Peiro, Gonzalez-Morales & Martin, 2006). Indeed,

PAGE 30

15 main effects, moderation effects, and mediati on effects of coping are observed in relevant literatures. As noted by Parkes (1990), the theoretical appropriateness of main, mediating, or moderating effects is dependent on the sp ecific stressors, outcomes and context of interest, not on a general coping function that is universally observed across paradigms. Findings from role-conflict re search specifically suggest a moderating effect from coping behaviors or strategies. C oping behaviors have moderate d between role conflict and emotional exhaustion and depressive symptoms; between life event stress and depression; between job disruption and depr ession, between occupational st ress or work overload and affective distress (Lam & McBride-Chang, 2007; Osipow, & Davis 1988; Parasuraman & Cleek, 1984; Pearlin, Menaghan, Lieberma n, & Mullan, 1981; Pomaki, Supeli, & Verhoeven, 2007). Problem-focused coping is also evidenced as a moderator between work demands and subjective health co mplaints (Eriksen & Ursin, 1999). Prior to hypothesizing a specifi c integration of coping in the present research, it is important to note that an appr opriate specification of the c oping construct should reflect aspects of the stressor as well as the domain of the outcomes. The stressor of interest in the present research is WIF (work interference with the family domain). The outcome of interest is eating behavior (and later the eati ng behavior of children in the home) which generally implicates non-work responsibiliti es or the family domain. Therefore, a relevant form of coping would represent cognitions and/or behaviors that aid the employee in managing multiple roles with specific implications for non-work responsibilities in the family domain. Household coping strategies (Steffy & Jones, 1988) represent both structural redefinition of one’s family or non-work role and

PAGE 31

16 personal-role redefinition, whereby adults alter expectations and personal attitudes associated with household responsibilities. St ructural role redefinition may manifest in numerous ways, such as taking action that enco urages family members to expect less, or organizing the sharing of one’s household respon sibilities among others family members. Personal role redefinition suggests prioritizi ng family role activit ies and taking on the most important activities first. In the cont ext of WIF and health, a moderating effect of coping is predicted (Figure 2). The interpretation of ‘moderation’ in this context is that coping may buffer the effects of WIF on eati ng strategies. When low household coping efforts are reported, WIF is hypothesized to demonstrate a stronger negative relationship with eating fruits and vegetables and a st ronger positive relations hip with eating snack foods. H5. The relationship between WIF and ea ting fruits and vegetables on work days is moderated by household coping strate gies, such that stronger WIF-eating associations occur when little or no hous ehold coping strategies are reported. H6. The relationship between WIF and eati ng snack foods, irrespective of day, is moderated by household coping strategies Stronger WIF-eating relationships will occur when little or no household coping strategies are reported. Alternative Roles of Coping Given the many functions of coping demonstrated in the literature, two alternate roles for coping will be considered from an exploratory perspective. First, Figure 3 presents a main effect of coping in the WIF-health mechanism. Main effects represent the mo st commonly hypothesized role of coping in stressor-strain relationships (Parkes, 1990). A comprehensive review and meta-analysis of the literature revealed that problem-focus ed coping positively related to overall health

PAGE 32

17 outcomes, such as objective weight gain, and self-reported physical health ratings (Penley, Tomaka, & Wiebe, 2002), providing suppor t for a main effect of coping. Eating behaviors themselves could function as a type of coping response to WIF, and as depicted in the model, eating fruits, vegetables a nd snack foods may be influenced by WIF and household coping strategies. Figure 2. Model of WIF-heal th moderated by coping. At least two studies have demonstrated a mediating capacity for coping between WFC and affective outcomes (Perrone, Aegi sdottir, Webb & Blalock, 2006; Voydanoff, 2002). The results reported by Perrone et al (2006) suggested that the influence of WFC on domain satisfaction was partially mediat ed by coping. In the context of health behaviors, Figure 4 models the WIF-eating behavior relationship as being partially mediated by coping. Specifically, the model predic ts that WIF is direc tly associated with both the adoption of household coping strate gies and eating behaviors. Household

PAGE 33

18 coping strategies are subsequently related to eating behaviors, constructing an indirect relationship between WIF a nd eating behaviors. Figure 3. Coping as a main effect Figure 4. Coping as a partial mediator

PAGE 34

19 Chapter Three: Crossover of WIF to Child Health Despite repeated calls for research to examine parental influence in the development of children’s health, the contextu al factors shaping the influence, and the ways in which parental work stress and employment issues affect children (e.g., Crouter & Bumpus, 2001; Davison & Birch, 2001; Gala mbos, et al., 1995; Greenhaus, et al., 2006; Kinnunen & Pukkinen, 2001; Prochaska, R odgers, & Sallis, 2002; Trost, Sallis, Pate, Freedson, Taylor, & Dowda, 2003), the re lationships between parent WIF and child physical health have not been examined. Experts argue that c ontextual factors (e.g., parent employment and WIF) ar e likely to impact parent-chi ld health behaviors (Davison & Birch, 2001). Documenting the potential cr ossover of WIF to child health carries considerable significance. Whereas less mall eable contributing fact ors to obesity (e.g., genes) are not easily targeted by policy and in tervention, other significant influences such as behavioral choices (e.g., poor eating ha bits) can be more readily managed through strategic intervention and public education. The prevalence of childhood obesity has increased significantly over th e past 20 years, and presentday estimates indicate that approximately 20% of children in the U.S. are obese (Torgan, 2002; Troiano & Flegal, 1998; U.S. Department of Health and Huma n Services, 2000). Many adult obesityrelated conditions such as high blood pressure early signs of hardening of the arteries, asthma, type 2 diabetes, and sleep apnea are now being observed in children with increasing frequency (Daniels, 2006).

PAGE 35

20 The WFC crossover literature primarily addresses the crossover of affective and cognitive outcomes between spouses (e.g., Hammer, Allen, & Grigsby, 1997; Westman & Etzion, 2005). There is some evidence of cr ossover between parent and child in which conflict or work demands influences parent behaviors and subsequently child behaviors and affect (negative spillove r from work to parent-adol escent interaction, Sallinen, Ronka, & Kinnunen, 2007; crossover of parent affect to child be haviors and affect, Stewart & Barling, 1996). Simila rly, parent stress (general stress and job-related stress) can also lead to parent-child interactions and parenting behaviors that negatively affect child/adolescent emotional outcomes (e.g., Ba rling, MacEwen, & Nolte, 1993; Galambos et al.,1995; Galinsky, 2000; Kinnunen & P ukkinen, 2001; MacEwen & Barling, 1991; McLoyd & Wilson, 1991; Perry-Jenkins, Repett i, & Crouter, 2000; Stewart & Barling, 1996). Physical health is not addressed in any of this work, allowing only theoretical inference about health-related crossover. The available findings on crossover support the domain specificity hypothesis (Frone, 2003) such that the WIF of one spouse can crossover and lead to family-related conseque nces for the child or other spouse. This basic process is in line with the theory of th e present research and is bolstered by indirect support from the parenting and obes ity literatures, described next. Parent Eating and Feeding Davison and Birch (2001) indi cate that dietary intake is one of the most proximal predictors in their ecological model of childhood overweight predictors. Next in proximity are parent influences (e.g., child f eeding practices, parent dietary intake, parent food preferences). Beyond affecting his or her ow n health, the parent plays a critical role in shaping the family eating environment (Bir ch & Fisher, 1995). It has been suggested

PAGE 36

21 that both parent eating practices and child ea ting behaviors should be considered in order to understand the impact that parent variab les can have on child health (Birch, 2006). Parent practices for child-feeding have been id entified as an environmental risk factor in childhood obesity, as they are linked to ch ild weight status (Birch & Fisher, 1998, Johnson & Birch, 1994). Furthermore, expert s emphasize that chil d feeding practices shape the child’s eating envir onment, child food preferences, child eating behaviors, and child self-regulation of ener gy intake (Birch, 2006). Th e feeding construct-domain encompasses parent restriction of foods, pre ssure to eat, and mon itoring child eating. These practices are related to child eating a nd health differently. BMI, food intake and weight are positively predicted by both restri ction of foods and monitoring child eating, but negatively predicted by pressure to eat (Birch & Fisher, 1998; Faith, Scanlon, Birch, Francis, & Sherry, 2004; Johnson & Birch, 1994; Kaur, Li, Nazir, Choi, Resnicow, Birch & Ahluwalia, 2006). Restriction of food can increase the child’ s interest in and preference for the restricted food. Further, re striction has been linke d to an increase of children eating when they aren ’t hungry (Birch, Fisher & Davison, 2003; Fisher & Birch, 1999). In the present context, the experience of work-int erference-with-family may represent work demands directly interferi ng with the parent’s family demands or responsibilities for feeding ot her family members. Feeding practices are likely to be negatively related to WIF in terms of reduced physical and psychological availability to control (pressure and restric tion), and maintain awareness of (monitoring), child eating. Beyond feeding practices, th ere is evidence that ch ildren exhibit stronger preferences for high fat foods if their parents are obese (B irch & Fisher, 1995, Klesges, Eck, Hanson, Haddock, & Klesges, 1990). Such fi ndings are typically explained by role-

PAGE 37

22 modeling eating behaviors which influence the impressionable child (speed and duration of parent eating, Agras, Berkowitz, & Ha mmer, 1988; mother’s fruit and vegetable intake, Galloway, Fiorito, Lee, & Birch, 2005; parents with high diet ary intake, Davison, Francis, & Birch, 2005; Laskarzewski, Porri son, Khoury, Kelly, Glatfelter, Larsen, & Glueck, 1980; Oliveria, Ellison, Moore, & Gillman, 1992; Patterson, Rupp, Sallis, Atkins, & Nader, 1988; Perusse, Leblanc, & Bouchard, 1988; Vauthier, Lluch, Lecomte, Artur, & Herbeth, 1996; similar parent-ch ild food preferences, Borah-Giddens & Falciglia, 1993). As hypothesized in Chapter Two, WIF is expected to be related to parent fruit, vegetable and sn ack intake behaviors, which may function as parent rolemodeling of eating behaviors to the child. Building upon the hypotheses presented in Chapter Two, the available support for parent-child crossover, and for parental influence via feeding and role-modeling, the second aim of this study is to examine pa rent-child WIF crossover to health. The theoretical framework in Figure 5 delineates a process in which parent WIF crosses over from parent feeding and eating behaviors to child ea ting behaviors. Coping is expected to function in the same capacity as hypothesized in Chapter Two. WIF is expected to be negatively related to the pare nt’s child-feeding practices, which subsequently have the opportunity to directly relate to child eating behaviors. H7. WIF is negatively related to pre ssure, monitoring and restriction feeding practices. H8a. Feeding practices are related to ch ild consumption of fruits and vegetables (positive relationships with monitoring, negative relationship with pressure), and

PAGE 38

23 snack foods (negative relationship with monitoring and positive relationship with restriction). H8b. Feeding practices mediate the re lationship between WIF and child eating behaviors. Role-modeling is represented by the top pa th in Figure 5, from WIF to parent eating behaviors to child eati ng behaviors. It is expect ed that parent healthy and unhealthy eating behaviors will exhibit a dire ct relationship with child healthy and unhealthy eating behavior s, respectively. H9. Parent fruit and vegetable consumpti on will be positively related to child fruit and vegetable consumption. H10. Parent snack food consumption will be positively related to child snack food consumption. H11. Parent healthy and convenience eati ng will mediate between WIF and child healthy and convenience eating behaviors. Current Study The present study investigated the associ ation between WIF, eating behaviors, BMI and the role of household coping strate gies in adults. Next, crossover between parent WIF and child health behaviors vi a parent feeding and eating behaviors was examined. Full explication of the hypothe sized relationships will ultimately require research targeting a number of specific populat ions that vary in ethnicity, culture (e.g., Ahye, Devine, & Odoms-Young, 2006), marita l status, dual-employment, gender of parent, and gender of child. Further, mother vs. father role-modeling and active parent involvement may influence adolescents diffe rently (Barber & Delfabbro, 2000; Patock-

PAGE 39

24 Figure 5. Model of WIF-health cros sover between parent and child

PAGE 40

25 Peckham & Morgan-Lopez, 2006), as could tradit ional vs. non-traditiona l parenting roles. For the purposes of conducting the first resear ch to evaluate the proposed spillover of WIF onto child health behaviors, dyads of employed mothers and their children were prioritized in the present study. As previous ly noted, workforce trends have indicated a substantial increase in the number of mothers in the wo rkplace (Halpern, 2004) and mothers are traditionally more closely i nvolved with a child’s feeding and eating behaviors (e.g., Harrell, 1995). Ther e is evidence that mothers tend to have or exert more influence regarding eating behaviors than do fathers (Fisher & Birch, 1999; Smolak, Levine, & Schermer, 1999). Research is cert ainly warranted for fathers as well, and father-based extension of the present effort will be described in the future research directions of Chapter Seven: Discussion. A dditionally, the child age range deemed most relevant for the hypothesized relationships was prepubescent because puberty may cloud the role of parent influen ce in child eating behaviors (e.g., eating more in relation to sporadic growth spurts and hormonal changes while unrelated to parent influence), and parent influence may be less re levant for older children who tend to have more autonomy over what they consum e (e.g., teenagers).

PAGE 41

26 Chapter Four: Method Participants and Sampling An a priori power analysis was cond ucted using Fritz a nd MacKinnon’s (2007) methods for determining adequate sample sizes in meditational analyses. To achieve statistical power of .80 with small-medium effect sizes ( = .26) for the alpha (independent variable to medi ator) and beta (mediator to dependent variable) paths, samples of 148, 162 or 196 participants were required for the bias-corrected boostrap, percentile bootstrap, and Sobel mediation procedures, respectively. The target sample size during recruitment was 200 mo thers or mother-child dyads. Approximately 509 families were recruite d from a random sample of 20 YMCA Afterschool programs in Tampa, Florida. From this recruitment sample, 334 families indicated their interest and intent to participate in the study. Although this yields an approximate 65.6% positive response to recrui tment, the estimate is conservative. The true response rate cannot be calculated due to inconsistent YMCA records across the 20 sites, therefore it is not possible to determine whether nonresponse was due to nonenrollment in the YMCA at the time of recruitment (e.g., students who may not have been enrolled at the time of recruitment, but study materials were left for the families because the child name and age were on the YM CA roster), ineligibility for the study (no mother in the family, unemployed mother, ch ild age different from YMCA records and outside of eligible range), or intentional nonresponse/disinterest in the study. The child

PAGE 42

27 age range of 8 to 11 years old was targeted in order to recruit primarily prepubescent children who were still old enough to read survey items and provide a reasonable assessment of their food inta ke and physical activity. From the 334 families successfully recrui ted, a total of 262 mother surveys were received which suggests an observed mother response rate of 78.4%, although the issues described above also render this estimate c onservative. Survey da ta was collected from 283 of the 306 children for whom parent consen t to participate was obtained (17 children were repeatedly absent during administra tions, 2 children opted out during informed assent procedures, and 4 child surveys were ad ministered but the data was lost). A final sample of 262 mothers and 283 children provided 238 matched mother-child dyads. Mother participants worked between 20 and 70 hours per week ( M = 41.72, SD = 7.09) and all mothers had at least one child living at home. The demographics of the sample of mothers are displayed in Table 1. The sample was predominantly Caucasian (52.9%), Black/African-American (22%), a nd Hispanic (18.5%). The majority of mothers reported being married (55.1%) or living with a partner (10.2%), and 65% had more than one child living at home. Age ranged from 24 to 61 years old ( M = 37.26, SD = 6.95). The modal level of education was “Som e college” (27%), and education level in the sample ranged from some high school to gr aduate degrees. A quarter of the sample reported an annual household income of $80,000 or higher (25.1%), with $30,000 – $39,999 (17.3%) and $20,000 $29,999 (14.8%) as th e brackets with subsequently highest representation. Records from the YM CA indicate that approximately 25% of all enrollees attend the Afterschool program at a re duced cost or for free. The distribution of body mass index (BMI) in the sample was examined in comparison to national and

PAGE 43

28 Table 1. Demographic characterist ics of mother participants Variable % Ethnicity Caucasian, Non Hispanic 52.9% Hispanic 18.5% Black / African American 22.0% Asian 1.5% American Indian 0.5% Multiracial 4.2% Marital Status Married 55.1% Not married but living with partner 10.2% Not married 34.8% Children One child living at home 35.0% Two children living at home 42.3% Three or more children living at home 22.7% Education Some high school 2.7% High school diploma/GED 21.2% Some College 27.0% 2-year college degree 15.4% 4-year college degree 19.3% Some graduate school or graduate degree 14.3% Note. N = 262

PAGE 44

29 Table 1. (cont’d) Variable % Annual Household Income $10,000 $19,999 7.8% $20,000 $29,999 14.8% $30,000 $39,999 17.3% $40,000 $49,999 10.3% $50,000 $59,999 10.7% $60,000 $69,999 6.2% $70,000 $79,999 7.8% $80,000 or higher 25.1% Body Mass Index (BMI) BMI < 25 (Normal weight) 54.3% BMI = 25 – 29.9 (Overweight) 25.3% BMI > 30 (Obese) 22.2% Note. N = 258 state norms. In 2005 estimates suggest that be tween 20 and 24% of adults in the state of Florida were obese (BMI > 30), and in 2006 experts estimated that approximately 23.5 % of women in the U.S. were obese (Ogden, Carroll, Curin, McDowell, Tabak & Flegal, 2006). In the sample for the present study, 20.2 % of mothers reported height and weight measurements that yielded a BMI of greater than 30 and the designation of being obese. The demographics of the child participants are displayed in Table 2. The child sample exhibited ethnic representation similar to the sample of mothers, across Caucasian (46.3%), Hispanic (13.9%) and Black/African -American (20.5%) ethn icities, with a slightly higher percentage of multiracial ethnicities reported (14.7%). Fifty-six percent of

PAGE 45

30 the child sample were girls, and the children’ s ages were eight (35.1% ), nine (32.6%), ten (27.6%) and eleven (4.7%) years old. Table 2. Demographic characteri stics of child participants Variable % Ethnicity Caucasian, Non Hispanic 46.3% Hispanic 13.9% Black / African American 20.5% Asian 1.5% American Indian .5% Multiracial 14.7% Other 2.7% Child Age 8 years old 35.1% 9 years old 32.6% 10 years old 27.6% 11 years old 4.7% Gender Boy 44% Girl 56% Note. N = 283 Measures Overview. Mother self-report was used to measure mother constructs (WIF, coping), mother eating behaviors, and mother BMI. Child self-report was used to represent child ea ting behaviors.

PAGE 46

31 Eating Behaviors. Based on pilot-tests, items were developed to represent consumption of vegetables (8 items), fruits (8 items), snack foods (8 items). Each item separately probes the frequency of consumpti on at breakfast, lunch, snacks and dinner. The breakfast-lunch-snack-dinner questions are presented twice for each food group (fruits, vegetables, snack foods), once referri ng to work days, and a second time referring to non-work days. Mothers responded to thes e items using themselves as a referent (Appendices A, B), and a second time shifting th e referent to the child (school days and weekends). Children responded to these item s in the same format (e.g., self report, Appendices C, D), and although child report of mother eating behaviors was not targeted by the hypotheses, this data was also collected for exploratory purposes. Work Interference with Family WIF was measured using five items developed by Netemeyer et al. (1996) (“Due to work-relate d duties, I have to make changes to my plans for family activities.”; “The demands of my work life interfere with my home and family life”; Appendix E). Responses were meas ured on a 5-point scal e that ranges from “no, never” to “yes, always. This measur e has demonstrated good internal consistency, strong dimensionality (differe ntiating between family interf erence with work and WIF), it does not confound the WIF construct with cons equences of WIF, and evidence supports its discriminant and convergent validity (Netem eyer et al., 1996). Mo ther self-report data indicated strong internal consistency ( = .94). Household Coping Strategies Household Coping Strategies was assessed by an adaptation of the Steffy and Jones (1988) H ousehold Coping Strate gies scale. The adapted nine-item scale uses a five-point Likert frequency scale to measure the respondent’s cognitive and behavioral effort s to handle their house hold responsibilities

PAGE 47

32 (Sample item, “Do you hire people to help w ith chores, for example, babysitters, cleaning help, yard help, etc?”; Appendix F). The moth er self-report data s uggests adequate scale reliability ( = .78). Feeding Behaviors Feeding Behaviors was measured using three mother selfreport scales from the Child Feeding Questionnaire (CFQ; Birch, Fisher, Grimm-Thomas, Markey, Sawyer, & Johnson, 2001; Appendix G). The three types of feeding behaviors assessed included: monitoring (3 items; = .94; e.g., “Do you keep track of the sweets that your child eats?”), restriction (8 items; = .76; e.g., “I have to be sure that my child does not eat too many high fat foods.”), and pressure (4 items, = .74; e.g., “If my child says ‘I'm not hungry’, I try to get him or her to eat anyway.”). In order to examine the dimensionality of the three feeding behavior subscales, competing confirmatory factor analyses were performed. A three factor, two factor, and one factor model were specified (Table 3), and resulting fit statistics examined. The chi-square test of fit was significant for all models, but reduced in size as th e number of factors specified increased, suggesting better fit of the th ree-factor model. Comple mentary fit statistics also improved as the number of factors modeled increased, including RM SEA (target: below .08), CFI (target: at or above .95) TLI (target: at or above .90). The fit of the three-factor model was not ideal, however, it demonstrat ed the best fit of the competing factor structures. Thus, the three i ndividual feeding behavior scal es were retained in their original form. Minor wording alterations were applied to the CFQ items in order to provide respondents with examples of foods mentioned by the items. Child repo rt of the mother’s feeding behaviors was not required for hypothesi s testing, but this data was collected for

PAGE 48

33 exploratory purposes by reframing the questi ons to reflect the mother’s behavior (monitoring, = .82; restriction, = .54; and pressure = .55). BMI. Mother weight (pounds) and height (feet, inches) was self-reported (Appendix H). Parent weight was converted to kilograms, height was converted to centimeters, and BMI was calculated from the mother self-report data (BMI = weight (kg) / [height (m)]2; Center for Disease Control, 2007). Table 3. Confirmatory factor analys is models: Mother feeding behaviors Model x2 df RMSEA CFI TLI One Factor 1015.79** 90 .20 .50 .42 Two Factor 726.70** 90 .16 .66 .60 Three Factor 502.03** 90 .13 .78 .74 Note. N = 263 **p < .01 Procedure Piloting. Two pilot survey sessions were conducted at a YMCA summer camp to determine the feasibility having children in our target age group respond to the survey questions and to obtain feedback from mothers regarding the adult survey. Four children (ages nine to ten years old) and their mothers pa rticipated in the first pilot. Feedback from the child administration inspired a number of item-wording changes as well as the incorporation of brief presen tations about fruit, vegetabl es, snack foods and physical activity placed directly before the beginni ng of the survey. Feedback from the pilot mothers suggested the need for additional instru ction in two areas of the mother survey. The second pilot was conducted with five childre n (ages eight to eleven years) to assess

PAGE 49

34 child response to the revised survey. The pilot session indica ted that the revised protocol and survey functioned more efficiently than the previous wording and protocol. The second pilot revealed the need to administ er the survey separa tely and with less instruction for ten and eleven year-olds, due to their hi gher reading and comprehension level compared to eight and nine year-olds. Recruitment. The Tampa Metro YMCA granted perm ission to recruit participants and conduct the proposed research in the “Afterschool” programs in Hillsborough County (41 program sites in operation, 20 randomly sele cted for recruitment). Each Afterschool site was visited one week before the sche duled child-survey administration to recruit children and parents. During th is visit, a brief introduction was made to the children to explain the purpose of the study and what child involvement and compensation entails. Parents of eligible children were approached when the parent arrived to sign out the child to go home from the program. Parents were offered a brief verbal explanation of the study and the study informed consent for child pa rticipation was presented to be read and signed or taken home to review. Eligible ch ildren had to be between the ages of eight and eleven years old (as of August 1st, 2007) and enrolled in a YMCA Afterschool Program. To be eligible a mother must have had a child in the targeted age range who attended a YMCA Afterschool program, she must have been employed at least 20 hours per week, she must have been the child’s biological mother, adoptive mother, step mother, or female legal guardian, and she had to indicate that she felt comfortable reading and writing in English without a translat or. Accompanying the consent was a parent letter, an extra copy of consent to keep for thei r records, and the mother survey to fill out in the next 14 days. Parents who did not give informed consent on the recruitment day

PAGE 50

35 had seven days to return the cons ent, in order for the child to participate. If a mother had multiple children in the target age range, she was asked to participate with the oldest child (if the mother had twins, one twin wa s randomly selected to participate with the mother). Non-selected twins and younger si blings in the age range were given the opportunity to participate in the survey themselves to ea rn the same compensation, but their data was not used when testing the hypotheses. Survey Administration. On the scheduled survey ad ministration day, a research team returned to the Afterschool site seve n days prior to collect data from child participants whose parents provided informed consent. A brief interactive presentation was made to the child participants to establish a frame of reference for the fruit, vegetables, and physical activity survey items At the end of the presentation, special care was taken to communicate two key elements to the children: 1) Child were told that the survey is about themselves; the answ ers of other children don’t matter because children are to answer about themselves. 2) Ch ildren were instructed to answer whatever was ‘true’ about themselves. The research team emphasized that there were no right or wrong answers to the survey, the children co uld only get the question right by answering what is true for them. These steps were ta ken to improve the quality of child responses, and temptation to use or seek anot her child’s answers to the survey. Informed assent was obtained from th e children, communicating that the child may stop any time and still receive compensatio n. Survey administration took place in small groups with a researcher assigned to every five (or fewer) children, depending on the overall child participant-re searcher ratio (smaller groups preferred). When possible, eight and nine year-olds were grouped toge ther, and ten and elev en year-olds grouped

PAGE 51

36 together (to better match basic comprehension ability). Survey proctors verbally accompanied all children through the entire survey, and a bank of allowable and recommended comments and clarifications was provided to encourage consistent administration across research assistants and ac ross collection sites. As a general rule, every question was read to ei ght and nine year-olds, along with a description of scale anchors every time the scale changed between it ems. Ten and eleven year-olds were read at least one question from every scale, and receiv ed a description of scale anchors every time the scale changed between items. Rec ognizing that not all children in either age group would have identical ability, the surv ey proctor was allowed to repeat any comments from the bank of accepted comments without restriction. Likewise, scale anchors could have been read additional time s throughout a scale, and each question read aloud for the older age group if the proctor de emed appropriate according to the ability displayed by the child. Children were encour aged to ‘think in their head’ and not out loud, to avoid influenc ing other children. Participant mothers were encouraged to take the survey home to complete and they were instructed to complete their que stionnaire without discus sing the content with the child. Because the mother survey was unproctored, mothers did not receive verbal instructions for specific secti ons, nor did they receive the fr ame-of-reference training. To encourage similar perceptions of the eati ng behavior items reader-friendly written instructions were provided with examples that mirror the child presentation. Reminder phone calls were made to mothers at seve n, ten, and fourteen days, with additional surveys dropped off as needed. Survey return boxes were set up at each site for staff to deposit returned materials. Follow-up visits to pick up returned materials and leave

PAGE 52

37 reminder slips were scheduled as needed. To improve the expected response rates for mothers (15 -20%) participants were co mpensated for their participation. Incentives. Children received an inexpensiv e toy for their participation, and mothers received a $15 giftcard of their choice (Walmart, Target, AMC Theaters, Starbucks) for completing and retu rning the time one survey. In an effort to benefit study participants and encourage healthy lifestyle behaviors, participant dyads who completed and returned the study materials received a “Healthy Living” pamphl et and free five-day passes for their family to visit any YMCA facility (provided by the Tampa Metro YMCA). The free YMCA passes were not an nounced prior to receipt of the completed surveys, to avoid self-selection into th e study by mothers who may have prioritized physical activity or family activities.

PAGE 53

38 Chapter Five: Results Preliminary Analyses. The data were screened for outliers and normality (skewness, kurtosis). Three of the study variables had between one and thre e outlier values, th erefore the hypothesis analyses were run with and without the outli ers and the results ex amined for agreement (Table 4). Neither the direc tion of effects nor significance differed when results were rerun without the outliers; therefore they were not removed from the final dataset. Six study variables were identified as having distribution issues of either skew and/or kurtosis (5 with a positive skew, 1 w ith a negative skew, 3 with a leptokurtotic distribution; as calculated by dividing kurtosis or skewness st atistic by its standard error and identifying variables with resulting values greater than 3.3). These vari ables (Table 4) were graphed for visual inspection, and the distribution violations deemed minor. Therefore no transformations were applied to the data. Descriptive statis tics for the study variables are presented in Tables 5 (Mothe r self-report), 6 (Chi ld report of mother, exploratory), 7 (Child self-report), a nd 8 (Mother report of child, exploratory). Hypothesized Operationalization of Variables: Special Considerations Mother Report of Child vs. Child Self-Report. Hypotheses were developed with the intent of repres enting variables about the mother by mother self-report, and the variables about the child by child self-report. The interc orrelations between mother selfreport and child self-report study variables are reported in Table (9). Data were also

PAGE 54

39 Table 4. Normality of study variables Outliers # Mother self-report of BMI 3 Mother self-report of Snack foods on Work days and Days off 2 Mother self-report of Snack foods on Work days 3 Normality +/Skewed, L/P Kurtosis Mother self-report of WIF +S Mother self-report of BMI +S, LK Mother self-report of Monitoring (Feeding) -S Mother self-report of Snack foods on Work days and Days off +S, LK Mother self-report of Snack foods on Work days +S, LK Child self-report of Snack foods on School Days and Weekends +S collected in which the mother reported about the child (also in Table 9) and in which child reported about the mother (Table 10) for exploratory pu rposes. The self vs. otherreport data will be discussed in Chapter Six: Supplementary Results. On vs. Off Days in Child-Focused Hypotheses. The hypotheses involving mother and child eating behaviors (H9, H10, H11) we re operationalized to reflect the motherfocused hypotheses involving WIF and eating be haviors (H1a, H2). Specifically, fruits and vegetables were examined with an emphasis on work/school days, and snack foods were examined across work/school days and days off/weekends. The hypotheses addressing mother feeding behaviors and chil d eating behaviors were analyzed with the child eating behaviors always opera tionalized across work/school days and days

PAGE 55

40 Table 5. Descriptive statistics: Mo ther variables (M other self-report) Variable Items Min Max Mean SD Skew Kurt BMI 16.60 45.76 25.99 5.65 1.192 1.300 WIF 5 1 5 2.09 0.92 .821 .357 .94 Coping 9 9 41 26.99 6.40 -.285 -.249 .78 Monitoring 3 3 15 11.11 3.36 -.682 -.154 .94 Pressure 4 4 20 10.44 4.59 .265 -.871 .74 Restriction 7 8 40 25.87 6.55 -.411 -.207 .76 Fruit & Vegetables (All) 16 16 72 44.28 9.55 -.085 .307 Fruit & Vegetables (Wk) 8 8 35 22.04 5.05 -.043 .118 Fruit & Vegetables (Off) 8 8 38 22.36 5.22 .009 .294 Fruit (All) 8 8 36 21.69 5.56 -.027 .064 Fruit (Wk) 4 4 20 10.58 3.16 .127 .143 Fruit (Off) 4 4 20 11.17 3.02 .012 .181 Veggies (All) 8 8 40 22.64 5.19 .186 .717 Veggies (Wk) 4 4 20 11.48 2.69 .205 .802 Veggies (Off) 4 4 20 11.19 2.91 .235 .252 Snack foods (All) 8 8 39 18.15 4.70 .777 1.889 Snack foods (Wk) 4 4 20 9.08 2.65 .792 1.434 Snack foods (Off) 4 4 20 9.05 2.52 .547 1.177 Note. N = 245-258

PAGE 56

41 Table 6. Descriptive statistics: Mother variables (Child report)* Variable Items Min Max Mean SD Skew Kurt Monitoring 3 3 15 11.72 3.31 -.888 -.049 .82 Restriction 8 8 40 27.00 5.66 -.102 .027 .54 Pressure 4 4 20 15.32 3.58 -.718 -.042 .55 Fruit & Vegetables (All) 16 18 80 49.65 12.7 .077 -.153 Fruit & Vegetables (Wk) 8 8 40 25.08 6.88 .101 -.504 Fruit & Vegetables (Off) 8 8 40 24.46 6.99 .171 -.177 Fruit (All) 8 8 40 25.79 7.16 -.115 -.431 Fruit (Wk) 4 4 20 13.03 3.97 -.066 -.520 Fruit (Off) 4 4 20 12.68 4.10 -.079 -.636 Veggies (All) 8 8 40 23.88 6.59 .197 -.117 Veggies (Wk) 4 4 20 12.12 3.67 .118 -.218 Veggies (Off) 4 4 20 11.74 3.68 .330 -.037 Snack foods (All) 8 8 40 19.15 6.17 .464 .373 Snack foods (Wk) 4 4 20 9.56 3.39 .453 .032 Snack foods (Off) 4 4 20 9.57 3.59 .543 .194 Note. N = 268 – 275 *Child report of mother data co llected for exploratory purposes.

PAGE 57

42 Table 7. Descriptive statistics: Child variables (Child self-report) Variable Items Min Max Mean SD Skew Kurt Fruit & Vegetables (All) 16 17 80 45.33 10.93 .234 -.201 Fruit & Vegetables (Sc) 8 9 40 23.35 6.46 .255 -.346 Fruit & Vegetables (We) 8 8 40 22.07 5.93 .202 -.291 Fruit (All) 8 8 40 24.38 6.20 .082 -.328 Fruit (Sc) 4 4 20 11.83 3.48 .133 -.521 Fruit (We) 4 4 20 12.60 3.86 .068 -.581 Veggies (All) 8 8 40 20.93 6.02 .286 -.168 Veggies (Sc) 4 4 20 10.26 3.29 .406 -.035 Veggies (We) 4 4 20 10.68 3.49 .343 -.243 Snack foods (All) 8 8 40 22.57 6.05 .389 .197 Snack foods (Sc) 4 4 20 11.20 3.43 .504 .002 Snack foods (We) 4 4 20 11.42 3.46 .269 -.036 Note. N = 275-281

PAGE 58

43 Table 8. Descriptive statistics: Child variables (Mother-report)* Variable Items Min Max Mean SD Skew Kurt Fruit & Vegetables (All) 16 16 75 43.90 9.16 .132 .599 Fruit & Vegetables (Sc) 8 8 38 21.96 4.67 .150 .739 Fruit & Vegetables (We) 8 8 40 21.94 5.30 .249 .343 Fruit (All) 8 8 40 23.25 5.39 .170 .564 Fruit (Sc) 4 4 20 11.65 2.84 .224 .495 Fruit (We) 4 4 20 11.60 3.19 .335 .363 Veggies (All) 8 8 37 20.63 4.97 .296 .495 Veggies (Sc) 4 4 19 10.30 2.53 .257 .329 Veggies (We) 4 4 20 10.35 2.91 .491 .793 Snack foods (All) 8 8 36 18.90 4.53 .249 .412 Snack foods (Sc) 4 4 20 9.40 2.48 .238 .850 Snack foods (We) 4 4 20 9.48 2.65 .546 .715 Note. N = 259-262 *Child report of mother data co llected for exploratory purposes.

PAGE 59

44 Table 9. Intercorrelations (Mother self-re port and report of chil d, child self-report) Fruits Mother SR Mother SR Mother report of child Child SR 1 2 3 4 5 6 7 8 9 10 11 12 13 Mother SR 1. Marital St. 2. Income .50** 3. BMI -.06 -.11 4.WIF .06 .18** -.04 5. Coping .15* .23** -.11† .19** Fruits Mother SR 6. Fruit (Tot) .03 .05 .00 -.05 .07 7. Fruit (Wk) .07 .07 -.03 -.04 .06 .91** 8. Fruit (Off) -.01 .02 .05 -.02 .08 .90** .63** Mother report of child 9. Fruit (Tot) .00 -.12† -.01 .00 .19** .53** .43** .53** 10. Fruit (S) -.10 -.18** .02 .00 .16* .42** .35** .40** .88** 11. Fruit (W) .08 -.05 -.02 -.02 .15* .53** .41** .54** .90** .60** Child SR 12. Fruit (Tot) -.01 -.10 -.07 .01 .05 .07 .06 .09 .32** .30** .27** 13. Fruit (S) -.03 -.07 -.01 .00 .02 .05 .02 .09 .24** .23** .18** .82** 14. Fruit (W) .03 -.12† -.09 .01 .06 .06 .07 .05 .31** .29** .27** .86** .42** Note. N = 202 -256. SR = Self-Report. † p < .10 *p < .05 **p < .01

PAGE 60

45 Table 9. (cont’d) Fruits Mother SR Mother SR Mother report of child Child SR 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Vegetables Mother SR 15. Veg (Tot) .05 .04 -.01 -.08 .11† .57** .50** .53** .37** .28** .37** .07 .03 .07 16. Veg (Wk) .03 .03 .00 -.14* .11† .50** .48** .42** .29** .23** .29** .02 .00 .02 17. Veg (Off) .07 .03 .00 .01 .10 .55** .45** .55** .39** .29** .39** .11 .06 .11† Mother report of child 18.Veg (Tot) .09 -.09 .02 -.08 .21** .36** .30** .36** .56** .49** .50** .25** .21** .23** 19.Veg (S) .04 -.12† .03 -.16* .16* .32** .28** .30** .51** .51** .41** .21** .20** .18** 20. Veg (W) .11† -.06 .00 -.01 .21** .34** .28** .35** .51** .40** .50** .23** .17** .24** Child SR 21.Veg (Tot) .10 .00 .05 .10 .06 -.02 -.04 .02 .18** .16* .16* .61** .50** .52** 22.Veg (S) .05 -.04 .07 .02 .06 -.08 -.11 -.03 .15* .15* .12† .54** .54** .37** 23.Veg (W) .13* .05 .02 .15* .04 .04 .03 .06 .16* .12† .16* .55** .36** .56** Snack Foods Mother SR 24.Snack (Tot) .01 .02 .07 -.09 -.06 -.03 -.06 .00 -.13* -.14* -.08 .06 .07 .03 25.Snack (Wk) .02 -.01 .04 -.08 -.08 .03 .01 .05 -.09 -.09 -.06 .05 .05 .04 26.Snack (Off) .00 .04 .09 -.09 -.04 -.10 -.12* -.06 -.15* -.16* -.10 .05 .08 .02 Mother report of child 27.Snack (Tot) .14* .08 -.06 .09 .06 -.12* -.07 -.15* -.14* -.14* -.12† -.03 -.02 -.01 28.Snack (S) .12† .08 -.10 .12† .10 -.08 -.01 '-.12† -.09 -.09 -.08 -.04 -.03 -.04 29.Snack (W) .13* .07 .00 .04 .00 -.12* -.10 -.13* -.16* -.16** -.12* -.01 -.01 .01 Child SR 30.Snack (Tot) .06 -.08 -.05 -.06 -.02 -.04 -.05 -.03 .02 .04 -.01 .23** .17** .23** 31.Snack (S) .07 -.06 -.07 -.03 -.02 -.04 -.03 -.05 .00 .02 -.02 .20** .15* .20** 32.Snack (W) .05 -.07 -.01 -.06 -.01 -.04 -.06 -.02 .03 .05 .01 .22** .16** .21** Note. N = 202 -256. SR = Self-Report. † p < .10 *p < .05 **p < .01

PAGE 61

46 Table 9. (cont’d) Vegetables Snack Foods Mother SR Mother report of child Child SR Mother SR Mother report of child Child SR 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Vegetables Mother SR 16. Veg (Wk) .92** 17. Veg (Off) .93** .71** Mother report of child 18.Veg (Tot) .47** .43** .44** 19.Veg (S) .39** .39** .34** .90** 20. Veg (W) .47** .40** .47** .93** .67** Child SR 21.Veg (Tot) -.05 -.07 -.01 .25** .23** .21** 22.Veg (S) -.03 -.05 .00 .22** .25** .16* .88** 23.Veg (W) -.06 -.09 -.02 .22** .16* .21** .90** .59** Snack Foods Mother SR 24.Snack Tot) -.06 -.08 -.04 -.04 .03 -.08 .00 .00 .00 25.Snack Wk) -.02 -.03 .00 -.03 .04 -.08 .00 -.02 .03 .92** 26.Snack Off) -.09 -.11† -.07 -.04 .02 -.06 -.01 .02 -.03 .91** .66** Mother report of child 27.Snack Tot) .06 .05 .05 -.07 -.10 -.04 -.03 -.04 -.01 .37** .29** .39** 28.Snack (S) .08 .09 .06 -.11† -.13* -.08 -.05 -.05 -.03 .32** .27** .32** .87** 29.Snack (W) .03 .01 .04 -.01 -.04 .02 .00 -.01 .02 .32** .24** .36** .89** .50** Child SR 30.Snack Tot) .01 -.05 .06 -.07 -.11† -.03 .16** .17** .12* .10 .05 .13† .17** .10 .20** 31.Snack (S) .01 -.03 .04 -.10 -.14* -.06 .07 .11† .04 .07 .02 .11 .15* .11 .15* .88** 32.Snack (W) .00 -.06 .05 -.02 -.06 .01 .22** .21** .19* .11† .08 .12† .16* .07 .20** .88** .55** Note. N = 202 -256. SR = Self-Report. † p < .10 *p < .05 **p < .01

PAGE 62

47 Table 9. (cont’d) Fruits Mother SR Mother SR Mother report of child Child SR 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Mother SR 33.Monitoring .01 -.01 .08 -.13* .18** .09 .08 .08 .18** .16* .16** .06 .12† -.01 34. Pressure .00 -.22** -.04 -.04 -.11† -.07 -.09 -.03 -.07 -.04 -.08 .02 .00 .04 35. Restriction .12* .11† .17** .06 .06 .07 .06 .07 .05 .03 .06 .02 .02 .02 Note. N = 202 -256. SR = Self-Report. † p < .10 *p < .05 **p < .01

PAGE 63

48 Table 9. (cont’d) Vegetables Mother SR Mother report of child Child SR 15 16 17 18 19 20 21 22 23 Mother SR 33.Monitoring .10 .13* .06 .25** .25** .21** .00 .02 -.02 34. Pressure -.04 -.05 -.02 .00 .01 .00 .00 .00 .00 35. Restriction .02 -.01 .06 .09 .05 .10 .01 .02 .00 Snack Foods Mother SR Mother report of child Child SR Mother SR 24 25 26 27 28 29 30 31 32 33 34 Mother SR 33.Monitoring -.04 -.02 -.05 -.07 -.06 -.05 .03 .03 .02 34. Pressure .20** .20** .15* .07 .08 .06 .09 .11 .04 .02 35. Restriction .18** .15* .17** .18** .14* .17** .11† .11 .08 .34** .14* Note. N = 202 -256. SR = Self-Report. † p < .10 *p < .05 **p < .01

PAGE 64

49 Table 10. Supplementary intercorre lations (Child report of mother) Mother SR Mother report of child Child SR WIF Coping Fruit (Tot) Fruit (Wk) Fruit (Off) Fruit (Tot) Fruit (S) Fruit (W) Fruit (Tot) Fruit (S) Fruit (W) Child report of Mother Fruit (Tot) .09 .09 .25** .26** .22** .22** .22** .17** .51** .31** .54** Fruit (Wk) .07 .14* .22** .23** .19** .15* .14* .13† .42** .27** .43** Fruit (Off) .12† .05 .21** .21** .19** .24** .25** .18** .50** .29** .53** Veg (Tot) .04 .15* .19** .15* .21** .17* .14* .17* .41** .27** .41** Veg (Wk) .03 .17* .19** .17* .20** .09 .06 .11† .38** .29** .34** Veg (Off) .06 .09 .14* .09 .16* .20** .19** .18** .35** .19** .39** Snack food (Wk) -.03 -.10 -.02 -.05 .00 .08 .05 .10 .27** .20** .25** Snack food (Wk) -.05 -.08 -.02 -.05 .01 .03 .00 .06 .24** .18** .23** Snack food (Off) .00 -.08 -.02 -.03 -.01 .11† .10 .11† .24** .18** .22** Monitoring -.06 .03 .15* .16* .12† .20** .18** .19** .21** .10 .25** Pressure -.03 -.08 .07 .07 .06 -.01 .01 -.01 .12† .10† .10 Restriction .11 .05 .02 03 .00 .03 .02 .04 .12* .07 .13* Note. N = 202 -256. SR = Self-Report. † p < .10 *p < .05 **p < .01

PAGE 65

50 Table 10. (cont’d) Mother SR Mother report of child Child SR Veg (Tot) Veg (Wk) Veg (Off) Veg (Tot) Veg (S) Veg (W) Veg (Tot) Veg (S) Veg (W) Child report of Mother Fruit (Tot) .09 .10 .08 .12† .11† .09 .46** .36** .45** Fruit (Wk) .08 .09 .06 .15* .14* .12† .41** .33** .40** Fruit (Off) .09 .08 .09 .07 .06 .06 .41** .32** .40** Veg (Tot) .20** .16* .21** .23** .20** .23** .56** .48** .52** Veg (Wk) .16* .12† .18** .19** .15* .19** .51** .44** .47** Veg (Off) .17* .14* .17* .22** .20** .21** .48** .40** .46** Snack food (Wk) -.11 -.13† -.08 .04 .04 .04 .34** .32** .28** Snack food (Wk) -.11 -.11† -.08 .06 .07 .05 .30** .30** .23** Snack food (Off) -.08 -.11 -.05 .01 .01 .01 .30** .27** .26** Monitoring .11† .13* .08 .15* .14* .12† .27** .21** .27** Pressure .00 .04 -.04 -.06 -.04 -.05 .15* .14* .12* Restriction .04 .05 .03 .03 .04 .02 .12* .12† .10 Note. N = 202 -256. SR = Self-Report. † p < .10 *p < .05 **p < .01

PAGE 66

51 Table 10. (cont’d) Mother SR Mother report of child Child SR Child report of Mother Snack (Tot) Snack (Wk) Snack (Off) Snack (Tot) Snack (S) Snack (W) Snack (Tot) Snack (S) Snack (W) Child report of Mother Fruit (Tot) -.07 -.02 -.11† -.08 -.12†† -.03 .23** .20** .21** Fruit (Wk) -.11† -.06 -.15* -.09 -.14* -.01 .16** .16* .13* Fruit (Off) -.02 .00 -.06 -.05 -.07 -.02 .25** .20** .26** Veg (Tot) -.01 -.02 .00 -.02 -.06 .04 .24** .20** .23** Veg (Wk) -.02 -.02 -.03 .00 -.04 .05 .22** .17** .22** Veg (Off) .00 -.02 .01 -.03 -.06 .01 .20** .18** .18** Snack food (Wk) .17* .13* .17** .07 .04 .09 .44** .29** .49** Snack food (Wk) .17* .14* .16* .12† .07 .15* .36** .23** .40** Snack food (Off) .13* .09 .15* .01 .01 .01 .42** .28** .46** Monitoring -.01 .03 -.04 -.05 -.04 -.05 -.06 -.05 -.05 Pressure .05 .05 .04 .03 .02 .05 .03 .01 .05 Restriction .09 .10 .06 -.02 -.04 .02 .12† .07 .15* Note. N = 202 -256. SR = Self-Report. † p < .10 *p < .05 **p < .01

PAGE 67

52 Table 10. (cont’d) Mother SR Monitoring Pressure Restriction Child report of Mother Fruit (Tot) .01 -.08 -.01 Fruit (Wk) .01 -.07 -.04 Fruit (Off) .00 -.10 .01 Veg (Tot) .00 -.03 .02 Veg (Wk) .03 .02 .04 Veg (Off) -.03 -.08 .01 Snack food (Wk) -.05 .07 .04 Snack food (Wk) .02 .14* .08 Snack food (Off) -.10 -.01 -.01 Monitoring .12† .00 .06 Pressure .01 .16* .01 Restriction .17* .08 .13† Note. N = 202 -256. SR = Self-Report. † p < .10 *p < .05 **p < .01

PAGE 68

53 off/weekends. General feeding tendencies were not predicted to vary according to on vs. off days for mothers or children, therefore feeding behaviors were examined (Hypotheses 8a and 8b) in relation to: child eating fruits, vegetables, and snack f oods irrespective of day of the week. Data Analysis Hypotheses 1a – 3a, 4a, 7, 8a, 9, and 10 were evaluated with simple correlations. Mediation hypotheses (H3b, H4b, H8b, H11) were tested using two analysis methods: a bootstrapping procedure to esti mate indirect effects (Preach er & Hayes, 2004), and the Sobel test. The Preacher and Hayes’ bootstrap methodology was employed to circumvent certain limitations of Sobel tests, namely the assumption of a normal distribution, conservative estimates, and th e need for large samples (e..g, Fritz & Mackinnon, 2007; Mackinnon, Warsi, & Dwyer 1995; Shrout & Bolger, 2002). The bootstrapping approach draws a predefined num ber of random samples from the data and calculates an indirect effect for each sample. As the process repeats, a distribution based on the bootstrap samples is formed, and this bootstrap distribution forms the basis for confidence intervals around the indirect effect for determining significance. All analyses utilizing this procedure were run with the specification of 1000 bootstrap samples, and 95% confidence intervals. Further, eac h analysis was conducted a second time, controlling for marital status and annual househ old income. In interest of triangulating the results across different analytic methods, each analysis of an indirect effect was also investigated using the So bel test (Sobel, 1982).

PAGE 69

54 Mother-Focused Hypotheses Hypotheses 1 and 2. The first three hypotheses addressed the relationship between mother WIF and mother eating beha viors. Partial support was observed for Hypothesis 1a, which predicted a negative re lationship between WIF and eating fruits and vegetables on work days. The data s upported a negative relationship between WIF and eating vegetables on work days ( r = -.14, p < .05), but there was no relationship between WIF and eating fruit on work days ( r = -.04, ns ). As predicted by Hypothesis 1b, there was no significant relationship be tween WIF and weekend consumption of vegetables ( r = .01, ns ) or fruits ( r = -.02, ns ). WIF did not demonstrate a significant relationship with eating snack foods on work days and off days ( r = .09, ns ), therefore Hypothesis 2 was not supported. Hypothesis 3. Hypothesis 3 concerned the relationship between eating fruits and vegetables on work days and BMI (H3a) and the mediating role of this work-day fruit and vegetable consumption in the relati onship between WIF and BMI (H3b). No relationship was observed between BMI and eatin g fruits and vegetables on work days ( rFruits ( Off ) = -.03, ns; rVegetables (Off) = .00, ns ), thus this hypothesis was not supported. The results for Hypothesis 3b using both th e Preacher and Hayes (3004) bootstrapping meditational procedure and the Sobel test di d not exhibit notable differences between fruits and vegetables, therefor e the results for eating fruits and vegetables are reported together here for parsimony (Table 11). Th e indirect relationship between WIF and BMI via eating fruits and vegetables on work days was not significant base d on the results of either analysis procedure (Indirect effect = .011, 95% CI [-.057, .152]), failing to provide support for Hypothesis 3b.

PAGE 70

55 Table 11. H3b: WIF and BMI mediated by fr uits and vegetables (Mother self-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.51 (.36) IV to Mediator (a) -.75* (.37) Mediator to DV (b) -.02 (.08) Mediator to DV (b) .00 (.08) Total Effect IV to DV (c) -.31 (.41) Total Effect IV to DV (c) -.23 (.43) Direct Effect IV on DV (c’) -.32 (.41) Direct Effect IV to DV (c’) -.23 (.43) DV Model R2 .003 DV Model R2 .009 .011 -.057, .152 .002 -.129, .150 N =231, 1000 bootstra p samples N = 217, 1000 bootstra p samples Note. Standardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 71

56 Table 12. H4b: WIF and BMI mediated by snack foods (Mother self-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.49 (.34) IV to Mediator (a) -.48 (.34) Mediator to DV (b) .09 (.08) Mediator to DV (b) .11 (.08) Total Effect IV to DV (c) -.24 (.40) Total Effect IV to DV (c) -.18 (.42) Direct Effect IV on DV (c’) -.20 (.40) Direct Effect IV to DV (c’) -.13 (.42) DV Model R2 .008 DV Model R2 .017 -.036 -.259, .020 -.045 .-.272, .029 N = 235, 1000 bootstra p samples N = 221, 1000 bootstra p samples Note. Standardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 72

57 Hypotheses 4a and 4b Hypothesis 4a posited a relationship between BMI and eating snack foods on work days and days o ff, and Hypothesis 4b predicted that eating snack foods, irrespective of day, would mediat e between WIF and BMI. No significant relationship emerged between eating snack foods and BMI ( r = .07, ns ). The results for Hypothesis 4b using the bootstrapping procedur e are presented in Table 12. There was no support for an indirect effect based on th e bootstrapping results (Indirect effect = .036, 95% CI [-.259, .020]) or the Sobel analysis ( zSobel = -.90, ns ). Hypotheses 5 and 6. The last two mother-focused hypotheses proposed that household coping strategies would moderate the relationships between WIF and eating fruits and vegetables on work days (H5) and between WIF and eating snack foods, irrespective of day. The m oderated regression results did not indicate significant moderation for vegetables on work days (Table 13), fruits on work days (Table 14), or snack foods (Table 15). However, WIF and c oping appeared to be meaningful predictors of eating vegetables on work days ( WIF = -.17, p < .05; Coping = .14, p < .05; Table 13), and the significance of the WIF and coping regression coefficients remained after controlling for marital status and annual household income ( WIF = -.23, p < .01; Coping = .16, p < .05). The alternative roles of coping (med iating vs. direct rela tionship with eating behaviors) will be examined as supplem ental analyses in Chapter Six.

PAGE 73

58 Table 13. H5: Coping as a moderator between WIF and eating vegetables Step 1 Model 1 Model 2 Step 1 Model 1 Model 2 Model 3 WIF -.17* -.17* Marital Status .06 .05 .05 Coping .14* .14* Annual Household Income -.01 -.01 -.01 Step 2 Step 2 WIF X Coping .02 WIF -.23** -.23** Coping .16* .16* Step 3 WIF X Coping .00 R2 .039** .040 R2 .003 .065 .065 R2 .01 R2 .062 .000 Final F 4.81** 3.21* Final F .285 3.71** 2.96* Note. Standardized regression coefficients. † p < .10 p < .05 ** p < .01 N = 237 N = 219

PAGE 74

59 Table 14. H5: Coping as a moderator between WIF and eating fruits Step 1 Model 1 Model 2 Step 1 Model 1 Model 2 Model 3 WIF -.06 -.06 Marital Status .08 .08 .08 Coping .07 .07 Annual Household Income .00 .01 .01 Step 2 Step 2 WIF X Coping .02 WIF -.11 -.11 Coping .03 .03 Step 3 WIF X Coping .03 R2 .007 .007 R2 .006 .017 .018 R2 .001 R2 .011 .001 Final F .785 .563 Final F .664 .918 .778 Note. Standardized regression coefficients. † p < .10 p < .05 ** p < .01 N = 238 N = 220

PAGE 75

60 Table 15. H6. Coping as a moderator between WIF and snack foods Step 1 Model 1 Model 2 Step 1 Model 1 Model 2 Model 3 WIF -.08 -.08 Marital Status -.06 -.05 -.05 Coping -.05 -.05 Annual Household Income .04 .06 .06 Step 2 Step 2 WIF X Coping -.05 WIF -.08 -.07 Coping -.05 -.05 Step 3 WIF X Coping -.03 R2 .010 .013 R2 .003 .012 .012 R2 .003 R2 .009 .001 Final F 1.21 1.01 Final F .279 .633 .543 Note. Standardized regression coefficients. † p < .10 p < .05 ** p < .01 N = 239 N = 221

PAGE 76

61 Table 16. H8b: Location of results Mediator Child Eating Report of Child Eating Bootstrap Method Monitoring Fruits Child self-report Table 17 Vegetables Child self-report Table 18 Snack foods Child self-report Table 19 Pressure Fruits Child self-report Table 20 Vegetables Child self-report Table 21 Snack foods Child self-report Table 22 Restriction Fruits Child self-report Table 23 Vegetables Child self-report Table 24 Snack foods Child self-report Table 25 Mother and Child Focused Hypotheses Hypotheses 7 and 8. Hypotheses 7 and 8 concerned the relationship demonstrated by feeding tendencies and WIF a nd child eating behaviors. Consistent with Hypothesis 7, the relationship betwee n WIF and monitoring was negative ( r = -.13, p < .05), but no relationship emerged between WIF and pressure feeding practices ( r = -.04, ns ) or restriction feeding practices ( r = .06, ns ). Therefore Hypothesis 7 was supported only with respect to monito ring feeding behaviors. Hypothesis 8a predicted that feeding practices would be related to child consumption of fruits and vegetables (pos itive relationships w ith monitoring, negative relationship with pressure), and snack foods (negative relationship with monitoring and positive relationship with restriction). Mother self-report of monitoring feeding practices was not related to the child self-report of eating fruits ( r = .06, ns ) or vegetables ( r = .00,

PAGE 77

62 ns ). Mother self-report of pressure feedi ng tendencies and of restriction feeding tendencies were not related to child self-report of eating fruits ( rPressure & Fruits = -.02, ns ; rRestriction & Fruits = .02, ns ) or vegetables ( rPressure & Vegetables = .00, ns ; rRestriction & Vegetables = .01, ns ). Mother report of monitori ng and of pressure were no t related to child report of child eating snack foods ( rmonitoring= .03, ns; rpressure = .09, ns ). However, mother report of restriction feeding behaviors was positively related to child self-report of eating snack foods, but the significance of th is relationship was marginal ( r = 11, p < .10). In summary, Hypothesis 8a was not supported Hypothesis 8b predicted that feeding be haviors would mediate between WIF and child eating behaviors. These results will be reported according to feeding behavior (monitoring, pressure, restricti on), and within each group, results will be presented in the following order: 1. Child eating fruits; 2.Ch ild eating vegetables; 3.Child eating snack foods. The organization of these analyses and th eir respective tables is displayed in Table 16. As reported in Table 17. the bootstrapping re sults and Sobel test did not indicate a significant indirect relationship between WIF and child self-report of eating fruits on school days and weekends via monitoring (Indirect effect = -.051, 95%CI [-.284, .027]; zSobel = -.844, ns ). The indirect relationship betw een WIF and child se lf-report of eating vegetables on school days and weekends vi a monitoring was not si gnificant based on the results of either analysis procedure (Indirect effect = -.011, 95%CI [-.157, .089]; zSobel = .248, ns ; Table 18). The indirect re lationship between WIF and child self-report of eating snack foods via monitoring also received no support (Indirect eff ect = -.010, 95%CI [.184, .083]; zSobel = -.248, ns ; Table 19).

PAGE 78

63 Table 17. H8b: WIF and child eating fruit medi ated by mother monito ring (All se lf-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.39 (.25) IV to Mediator (a) -.42 (.26) Mediator to DV (b) .12 (.13) Mediator to DV (b) .18 (.13) Total Effect IV to DV (c) .12 (.46) Total Effect IV to DV (c) .29 (.49) Direct Effect IV on DV (c’) .17 (.46) Direct Effect IV to DV (c’) .36 (.49) DV Model R2 .005 DV Model R2 .026 -.051 -.284, .027 -.077 -.352, .032 N =222, 1000 bootstra p samples N = 204, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 79

64 Table 18. H8b: WIF and child eating vegetabl es mediated by monito ring (All se lf-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.43* (.25) IV to Mediator (a) -.47† (.26) Mediator to DV (b) .03 (.12) Mediator to DV (b) .08 (.12) Total Effect IV to DV (c) .60 (.43) Total Effect IV to DV (c) .63 (.45) Direct Effect IV on DV (c’) .61 (.43) Direct Effect IV to DV (c’) .66 (.45) DV Model R2 .009 DV Model R2 .039 -.011 -.157, .089 -.032 -.275, .038 N =226, 1000 bootstra p samples N = 207, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 80

65 Table 19. H8b: WIF and child eating snack foods mediated by monitori ng (All self-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.41 (.25) IV to Mediator (a) -.45† (.26) Mediator to DV (b) .03 (.12) Mediator to DV (b) .09 (.12) Total Effect IV to DV (c) -.36 (.44) Total Effect IV to DV (c) -.19 (.46) Direct Effect IV on DV (c’) -.35 (.44) Direct Effect IV to DV (c’) -.14 (.47) DV Model R2 .003 DV Model R2 .026 -.010 -.184, .083 -.042 -.281, .038 N = 226, 1000 bootstra p samples N = 207, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 81

66 Both analyses indicated that the indirect relationship between WIF and child selfreport of eating fruits on school days and weekends via pressure was not significant (Indirect effect = -.007, 95%CI [-.079, .060]; zSobel = -.209, ns ; Table 20). The indirect relationship between WIF and ch ild self-report of eating ve getables on school days and weekends via pressure was not significant based on the results of either analysis procedure (Indirect eff ect = -.003, 95%CI [-.052, .078]; zSobel = .109, ns ;Table 21). The indirect relationship between WIF and child se lf-report of eating snac k foods via pressure also received no support (Indirect effect = -.010, 95%CI [-.191, .061]; zSobel = -.566, ns ; Table 22). As reported in Tables 23 – 25, the indire ct relationships between WIF and child self-report of eating behavior s via restriction were not si gnificant for fruits on school days and weekends (Indirect effect = .003, 95%CI [-.054, .138]; zSobel = .160, ns ), vegetables on school days and weekends (Indirect effect = .003, 95%CI [-.035, .145]; zSobel = .311, ns ), or for snack foods (Indire ct effect = .049, 95%CI [-.024, .256]; zSobel = .786, ns ). Hypotheses 9, 10 and 11. The final set of hypothese s predicted the relationships between mother eating behaviors and child ea ting behaviors, and mo ther eating behaviors as a mediator between WIF and child eating be haviors. Mother eating of fruits on work days was not related to child eating of fruits ( r = .02 ns ) or vegetables ( r = -.05 ns ) on school days, thus failing to support Hypothesi s 9. Hypothesis 10 specified a relationship between mother eating snack food with child eating snack food. This hypothesis was not supported by the child self-re port of eating snack food ( r = .10, ns ) except with child self

PAGE 82

67 Table 20. H8b: WIF and child eating fru it mediated by pressure (All self-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.07 (.33) IV to Mediator (a) .09 (.34) Mediator to DV (b) .02 (.04) Mediator to DV (b) -.03 (.10) Total Effect IV to DV (c) .10, (.46) Total Effect IV to DV (c) .26 (.49) Direct Effect IV on DV (c’) .10 (.46) Direct Effect IV to DV (c’) .26 (.49) DV Model R2 .000 DV Model R2 .016 -.007 -.079, .060 -.005 -.127, .062 N = 223, 1000 bootstra p samples N = 205, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 83

68 Table 21. H8b: WIF and child eating vegetabl es mediated by Pressu re (All se lf-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.18 (.33) IV to Mediator (a) -.01 (.34) Mediator to DV (b) -.01 (.09) Mediator to DV (b) .02 (.09) Total Effect IV to DV (c) .62 (.43) Total Effect IV to DV (c) .67 (.45) Direct Effect IV on DV (c’) .62 (.43) Direct Effect IV to DV (c’) .67 (.45) DV Model R2 .009 DV Model R2 .038† -.003 -.052, .078 -.004 -.069, .080 N = 227, 1000 bootstra p samples N = 208, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 84

69 Table 22. H8b: WIF and child eating sn ack foods pressure (All self-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.10 (.33) IV to Mediator (a) .08 (.34) Mediator to DV (b) .12 (.09) Mediator to DV (b) .09 (.09) Total Effect IV to DV (c) -.36 (.44) Total Effect IV to DV (c) -.19 (.46) Direct Effect IV on DV (c’) -.35 (.44) Direct Effect IV to DV (c’) -.20 (.46) DV Model R2 .011 DV Model R2 .028 -.010 -.191, .061 .004 -.062, .167 N = 227, 1000 bootstra p samples N = 208, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 85

70 Table 23. H8b: WIF and child eating fruit mediated by restricti on (All self-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) .50 (.49) IV to Mediator (a) .41 (.50) Mediator to DV (b) .01 (.06) Mediator to DV (b) .05 (.07) Total Effect IV to DV (c) .09 (.47) Total Effect IV to DV (c) .27 (.50) Direct Effect IV on DV (c’) .09 (.47) Direct Effect IV to DV (c’) .24 (.50) DV Model R2 .000 DV Model R2 .019 .003 -.054, .138 .013 -.023, .227 N = 222, 1000 bootstra p samples N = 204, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 86

71 Table 24. H8b: WIF and child eating vegetables mediated by restric tion (All self-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) .38 (.49) IV to Mediator (a) .30 (.50) Mediator to DV (b) .02 (.06) Mediator to DV (b) .00 (.07) Total Effect IV to DV (c) .49 (.44) Total Effect IV to DV (c) .49 (.46) Direct Effect IV on DV (c’) .49 (.4) Direct Effect IV to DV (c’) .50 (.46) DV Model R2 .006 DV Model R2 .036 .003 -.035, .145 -.007 -.089, .073 N = 226, 1000 bootstra p samples N = 207, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 87

72 Table 25. H8b: WIF and child eating snack foods mediated by restri ction (All self-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) .44 (.49) IV to Mediator (a) .36 (.50) Mediator to DV (b) .11† (.06) Mediator to DV (b) .14* (.07) Total Effect IV to DV (c) -.38 (.44) Total Effect IV to DV (c) -.21 (.47) Direct Effect IV on DV (c’) -.43 (.44) Direct Effect IV to DV (c’) -.26 (.47) DV Model R2 .017 DV Model R2 .044 .049 -.024, .256 .057 -.068. .284 N = 226, 1000 bootstra p samples N = 207, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 88

73 report of eating snack foods on weekends a nd mothers eating snack foods on off days, specifically, where it exhibited marginal significance ( r = .12, p < .10). Hypothesis 11 predicted mother consump tion of fruits, vegetables and snack foods would act as a mediator between WIF and the respective child eating behaviors. There was no support for the indirect relati onship between WIF and child eating fruit on school days using child self-report (I ndirect effect = -.009, 95%CI [-.099, .033]; zSobel = .25, ns ; Table 26). Results also did not support an indirect relations hip between WIF and child self-report of eating ve getables on school days via mother eating vegetables on work days (Indirect effect = .016, 95%CI [-.067, .119]; zSobel = .369, ns ; Table 27). Finally, there was no evidence of the hypothesi zed meditational rela tionships involving eating snack foods on work/school days (In direct effect = -.043, 95%CI [-.256, .036]; zSobel = .036, ns ; Table 28).

PAGE 89

74 Table 26. H11: WIF and child eating fruit medi ated by mother eating fruit (All self-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.20 (.23) IV to Mediator (a) -.36 (.25) Mediator to DV (b) .02 (.07) Mediator to DV (b) .02 (.08) Total Effect IV to DV (c) -.08 (.26) Total Effect IV to DV (c) -.01 (.27) Direct Effect IV on DV (c’) -.08 (.26) Direct Effect IV to DV (c’) .00 (.27) DV Model R2 .001 DV Model R2 .006 -.009 -.099, .033 -.013 -.129, .048 N = 221, 1000 bootstra p samples N = 203, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 90

75 Table 27. H11: WIF and child eating vegetables medi ated by mother eating vege tables (All self-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.51** (.19) IV to Mediator (a) -.68** (.20) Mediator to DV (b) -.03 (.08) Mediator to DV (b) -.05 (.09) Total Effect IV to DV (c) .10 (.24) Total Effect IV to DV (c) .18 (.25) Direct Effect IV on DV (c’) .09 (.25) Direct Effect IV to DV (c’) .15 (.26) DV Model R2 .002 DV Model R2 .017 .016 -.067, .119 .033 -.071,.159 N = 224, 1000 bootstra p samples N = 205, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 91

76 Table 28. H11: WIF and child eating snack foods medi ated by mother eating sn ack foods (All self-report) Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.41 (.35) IV to Mediator (a) -.40 (.37) Mediator to DV (b) .11 (.08) Mediator to DV (b) .12 (.09) Total Effect IV to DV (c) -.40 (.44) Total Effect IV to DV (c) -.26 (.46) Direct Effect IV on DV (c’) -.36 (.44) Direct Effect IV to DV (c’) -.21 (.46) DV Model R2 .010 DV Model R2 .036 -.043 -.256, .036 -.049 -.304, .037 N = 224, 1000 bootstra p samples N = 208, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 92

77 Chapter Six: Supplemental Results Supplemental analyses were conducted to allow explorati on of two issues: alternate roles of coping and mother-child perception. As noted in the introduction, the literature is not in agreement about the role of coping regarding stress ors and behaviors; it may function as a moderator, as a mediator, a nd as a separate direct effect across various contexts. Given the lack of support for the hypothesized moderating effect, two alternate models of coping were examined, the first with coping as a predictor of mother behaviors in addition to WIF, and the second with co ping as a partial mediator between WIF and mother behaviors. Supplemental analyses were also conducte d to learn more about mother-child perceptions. The agreement between mother and child report was examined first with respect to child eating behavior s and then for mother eating be haviors. In the second set of supplementary analyses of mother-child perceptions, th e similarity between singlesource reports of one’s own eating behavi ors and the eating behaviors of one’s counterpart was investigated using the child’s pe rspective and then using the mother’s perspective. Building upon these results, the third and last set of supplemental analyses related to mother-child percep tion explored the retesting of mother and child – focused hypotheses (H8 – H11) using mother-report only for all variables (only significant results tabled and reported in detail).

PAGE 93

78 Alternate Roles of Coping In the interest of explor ing the role of coping beyond the moderating hypotheses (H5 & H6), several supplementary analyses were performed. First, coping was explored as having a direct relationship with the mo ther eating or feeding behavior (“Model 1”; Figure 6), accounting for variance above and beyond that of WIF. Next, coping was examined as a mediator between WIF and the mother eating or feed ing behavior (“Model 2”). These results are reported in the followi ng sections (using mother self-report only), with tables presented for significant results only. Model 1 was supported by multiple regr ession results when the dependent variable was mother eating vegetables on work days ( WIF = -.17, p < .05; Coping = .14, p < .05, R2 = .049; Table 29), and when the dependent variable was monitoring ( WIF = -.15, p < .05; Coping = .22, p < .01, R2 = .057 ; Table 30),. Coping accounted for variance above and beyond that of WIF, and these results were still significant after controlling for marital status and household income. There was partial support fo r Model 1 from two dependent variables. Bootstrapping results for Model 2 (Figur e 6) revealed a significant positive indirect effect when the dependent variable was mother eating vegetables on work days (Indirect effect = .089, 95%CI [.016, .227], R2 = .053; Table 31). This positive indirect effect was also supported when monitoring was the dependent variable (Indirect effect = .153, 95%CI [.033, .361], R2 =.057; Table 32). However, in both cases the direct effect (IV to DV after partialling out the mediator) was negative. The opposite signs of indirect and direct effects suggest inconsistent media tion. The regression coe fficients between IV and DV were further examined with and wit hout the inclusion of the mediator, to

PAGE 94

79 Figure 6. Supplementary analysis : Alternate roles of coping

PAGE 95

80 Table 29. Supplementary, role of coping: Direct to mother eating vegetables Table 30. Supplementary, role of coping: Direct to monitoring Step 1 Model 1 Model 2 Step 1 Model 1 Model 2 Model 3 WIF -.14* -.17* Marital Status .01 .05 .02 Annual Household Income .02 .07 .02 Step 2 Step 2 Coping .14* WIF -.20* -.23** Step 3 Coping .16* R2 .020 .049 R2 .001 .04 .06 R2 .02* R2 .039* .02* Final F 4.73* 4.81** Final F .073 3.03* 3.60** Note. Standardized regression coefficients. † p < .10 p < .05 ** p < .01 Step 1 Model 1 Model 2 Step 1 Model 1 Model 2 Model 3 WIF -.11† -.15* Marital Status .02 .05 .01 Annual Household Income -.06 -.05 -.07 Step 2 Step 2 Coping .22** WIF -.14† -.17* Step 3 Coping .23** R2 .013 .057 R2 .002 .024 .067 R2 .045* R2 .018* .047* Final F 3.05† 7.21** Final F .254 1.49 3.92* Note. Standardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 96

81 Table 31. Supplementary, role of coping: WIF a nd mother eating vegetabl es mediated by coping Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) 1.46** (.46) IV to Mediator (a) 1.05* (.47) Mediator to DV (b) .06* (.03) Mediator to DV (b) .076* (.03) Total Effect IV to DV (c) -.43* (.19) Total Effect IV to DV (c) -.61** (.20) Direct Effect IV on DV (c’) -.51* (.19) Direct Effect IV to DV (c’) -.68** (.20) DV Model R2 .039** DV Model R2 .065 .089 .016, .227 .069 .006, .208 N = 237,1000 bootstra p samples N = 219, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 97

82 Table 32. Supplementary, role of coping: WIF and monitoring mediated by coping Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) 1.33** (.45) IV to Mediator (a) .94* (.45) Mediator to DV (b) .11** (.03) Mediator to DV (b) .12** (.04) Total Effect IV to DV (c) -.42† (.24) Total Effect IV to DV (c) -.49† (.25) Direct Effect IV on DV (c’) -.57* (.24) Direct Effect IV to DV (c’) -.60* (.25) DV Model R2 .057 DV Model R2 .068 .153 .033, .361 .111 .007, .296 N =241,1000 bootstra p samples N = 223, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 98

83 determine whether a suppression effect was presen t. In the case of eating vegetables as the dependent variable, the strength of the relationship with WIF (c path; bunstandardized = .43, p < .05, or standardized = -.14, p < .05) increased with th e inclusion of household coping in the model (c’ path; bunstandardized = -.51, p < .05 or standardized = -.17, p < .05). With monitoring as the dependent variable, th e strength of the relationship with WIF (c path; bunstandardized = -.42, p < .10, or standardized = -.11, p < .10) also increased with the inclusion of household copi ng in the model (c’ path; bunstandardized = -.57, p < .05, or standardized = -.15, p < .05). The pattern of increasing magnitude of the WIF relationships for both dependent variables s uggest suppression. Further, the directionality of the observed path coefficients resulting from this model were consistent with theory. All effects were still significant after controlling for marital st atus and household income. The results of mediation analyses suggest that coping functioned as a suppressor variable (a form of inconsistent mediation) rather than a partial mediator; the analyses yielded theoretically appropria te path coefficient directi ons and accounted for the more variance in the dependent variables than did th e other alternate role of coping (coping as main effect). The theoretical implications of these results for th e domain of work-family conflict and health behavior s will be considered in Chapter Seven: Discussion. Mother-Child Perceptions Part I: Similari ty in Mother Report and Child Report The similarity of perspectives across mo ther and child sources was explored first by comparing the correlations between mother and child reports of child eating behaviors. The agreement between mother and child about the child’s eating behaviors ranged from small to medium positive effect sizes. The relationship between sources was strongest for the child eating fruit across school days and weekends ( r = .32, p < .01),

PAGE 99

84 followed by the child eating vegetables across school days and weekends ( r = .25 p < .01), and the child eating snack foods ( r = .17 p < .01). Review of the similarity of mother and child perspectiv es regarding the mother’s eating behaviors also revealed small to medium positive effects sizes. As was observed in the multisource agreement for child eating be haviors, the mother eating fruits across work days and days off showed the strongest association ( r = 25, p < .01), followed by the mother eating vegetables across work days and days off ( r = .20, p < .01), and the mother eating snack foods ( r = .17 p < .05). Overall, agreement across sources tended to be slightly higher when reporting about the child’s behavior than when reporti ng about the mother’s behavior, specifically for eating fruits and for eating vegetables (a lthough snack foods demonstrated identical effect sizes for mother eating and for child eating). In other words mother and child demonstrated stronger agreement in their percep tions of the child’s behavior. Mother and child demonstrated slightly weaker agreem ent in their perceptions of the mother’s behavior. Mother-Child Perceptions Part II: Self-Other Similarity Using Single Source Data. The data were also examined for simila rity between reporte d mother and child behaviors, first according only to the child’s perspective and then according only to the mother’s perspective. The child report variables demonstrated medium to large correlations between mother and child eati ng behaviors. The largest magnitude of relationship was exhibited by mo ther and child eating vegetabl es across work days/school days and off days/weekends ( r = .56 p < .01), followed by mother and child eating fruit across work days/ school days and off days/weekends ( r = .51, p < .01), and mother and

PAGE 100

85 child eating snack foods across work days/ school days and off days / weekends ( r = .44, p < .01). The mother report variables also demonstrat ed medium to large effect sizes of the similarity between mother and child eati ng behaviors. The strongest relationships between mother report variables were observe d for mother and child eating fruit across all days ( r = .53 p < .01), followed by mother and child eating vegetables across all days ( r = .47 p < .01), and mother and child eati ng snack foods across all days ( r = .37, p < .01). These correlations essentially represen t the extent that the respondents perceive similarity between their eating behavior s and those of thei r counterpart. The observed relationships revealed that mothers tended to report their child’s eating behaviors similarly to how they repor ted their own, and the children tended to report their mother’s eating behaviors as sim ilar to their own. Interestingly, both mother and child reports indicated the most percei ved similarity (regar ding behaviors between themselves and their counterpart) on offda ys/weekends, than on work days/school days. Mother’s self report of child eating fruit and the mother’s own fruit consumption was more strongly correlated re garding off days/weekends ( rFruit Off/Weekend = .54 p < .01) than work days/school days ( rFruit Work days/School days = .35 p < .01), and child self-report of mothers eating fruit and the child’s own fru it consumption was more strongly correlated for off days/weekends ( rFruit Off/Weekend = .53 p < .01) than work days/school days ( rFruit Work days/School days = .27 p < .01). These trends were also present for eating vegetables according to mother report ( rVegetables Off/Weekend = .47 p < .01; rVegetables Work days/School days = .39 p < .01), and for snack food according to mother report ( rSnack Foods Off/Weekend = .36 p < .01; rSnack Foods Work days/School days = .27 p < .01). For child report of the variables, the same

PAGE 101

86 trend did not emerge for eating vegetables as these correlations were very similar in magnitude ( rVegetables Off/Weekend = .46 p < .01; rVegetables Work days/School days = .44 p < .01), but off day/ weekend relationship for snack foods was again stronger than the work day/school day relationship ( rSnack Foods Off/Weekend = .46 p < .01; rSnack Foods Work days/School days = .23 p < .05). In summary, sizable relationships were found between mother and child eating behaviors when single source data was used (m other-report only and child report only). Further, both sources reported more similar ity between themselves and their counterpart when answering about eating behaviors on off days/ weekends, than when reporting about work days/school days. These suppl ementary analyses provide a different perspective of the non-significant relationships observed when using multi-source self report. Mother-Child Perceptions Part III: Revi siting Mother and Child-Focused Hypotheses The last set of supplemental analyses involved the reexamining the mother and child-focused hypotheses using single-source data (only significant results tabled and reported in detail). These analyses investig ated WIF crossover to child eating behaviors via mother behaviors using only mother-report. Feeding and child’s eating behaviors (Hypothesis 8a). The relationships between feeding practices and child eating fruits, vegeta bles and snack foods were explored first. Mother self-report of monitoring feeding prac tices was positively related to mother report of child eating fruits ( r = .18, p < .01) and vegetables ( r = .25, p < .01). Mother report of restriction feeding behaviors was also positivel y related to mother report of child eating snack foods ( r = .18, p < .01), in the expected direc tion (restriction typically shown to

PAGE 102

87 lead to more child consumption of foods; e.g., Birch, Fisher, & Da vison, 2003). No other feeding behaviors demonstrated an associ ation with mother report of child eating behaviors. WIF and Child Eating Mediated by Feeding Behaviors (Hypothesis 8b). Feeding behaviors were examined as a mediator of the relationship between WIF and child eating behaviors. Bootstrapping re sults supported an indirect relationship between WIF and mother report of child eating fruits on school days and weekends (Indirect effect = -.138, 95%CI [-.399, -.016]; zSobel = -1.673, p < .10; Table 33) and this effect was still significant when controlling for marital status and household in come (Indirect effect = .161, 95%CI [-.461, -.010]). The indirect relationship between WI F and mother report of child eating vegetables on school days and weekends vi a monitoring was not significant according to the bootstrapping results (Indirect effect = -.168, 95%CI [-.428, .003]; zSobel = -1.84, p < .10; Table 34), but the relationship became signi ficant after controlling for marital status and household income (Indirect effect = .177, 95%CI [-.518, -.005]). No other indirect effects via feeding behavior s approached significance. WIF and Child Eating Mediated by Mother Eating (Hypothesis 11). Mother eating behaviors were also examined as a me diator of the relationship between WIF and child eating behaviors. The indirect relationship between WIF and child eating vegetables on school days via mother ea ting vegetables on work days was the only significant indirect effect (Indirect effect = -.145, 95%CI [-.289, -.004]; zSobel = -2.13, p < .05; Table 35) and it remained significant when marital status and household income were controlled.

PAGE 103

88 Table 33. Supplementary, role of coping: Coping and mother eating vegetables mediated by WIF Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) .03** IV to Mediator (a) .02* (.01) Mediator to DV (b) -.51* (.20) Mediator to DV (b) -.68** (.20) Total Effect IV to DV (c) .05† (.03) Total Effect IV to DV (c) .05† (.03) Direct Effect IV on DV (c’) .06* (.03) Direct Effect IV to DV (c’) .07* (.03) DV Model R2 .039 DV Model R2 .065 -.014 -.034, -.003 -.014 -.035, -.002 N = 237, 1000 bootstra p samples N = 219, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 104

89 Table 34. Supplementary, role of coping: Coping and monitoring mediated by WIF Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) .03** (.01) IV to Mediator (a) .02* (.01) Mediator to DV (b) -.57* (.24) Mediator to DV (b) -.60* (.25) Total Effect IV to DV (c) .10** (.03) Total Effect IV to DV (c) .11** (.04) Direct Effect IV on DV (c’) .11** (.03) Direct Effect IV to DV (c’) .12** (.04) DV Model R2 .057 DV Model R2 .068 -.015 -.039, -.002 -.012 -.039, -.0003 N = 241, 1000 bootstra p samples N = 223, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 105

90 Table 35. Supplementary, all mother report: WI F and child eating fruit mediated by monitoring Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.51 (.23) IV to Mediator (a) -.55* (.25) Mediator to DV (b) .28** (.10) Mediator to DV (b) .29** (.11) Total Effect IV to DV (c) .00 (.38) Total Effect IV to DV (c) .08 (.40) Direct Effect IV on DV (c’) .14 (.38) Direct Effect IV to DV (c’) .24 (.40) DV Model R2 .030* DV Model R2 .053* -.138 -.399, -.016 -.161 -461, -.010 N =249, 1000 bootstra p samples N = 229, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 106

91 Table 36. Supplementary, all mother report: WIF a nd child eating vegetables mediated by monitoring Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.46* (.23) IV to Mediator (a) -.50 (.24) Mediator to DV (b) .35** (.09) Mediator to DV (b) .36** (.09) Total Effect IV to DV (c) -.46 (.35) Total Effect IV to DV (c) -.48 (.35) Direct Effect IV on DV (c’) -.30 (.34) Direct Effect IV to DV (c’) -.30 (.35) DV Model R2 .062** DV Model R2 .094** -.168 -.428, .003 -.177 -.518, -.005 N =251, 1000 bootstra p samples N = 231, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 107

92 Table 37. Supplementary, all mother report: WIF and child eating vegetables mediated by mother eating vegetables Bootstrapped Indirect Effect Analysis Controlling for Marital Status and Annual Household Income Path Coeff (SE) Indirect Effect Corrected 95%CI Path Coeff (SE) Indirect Effect Corrected 95%CI IV to Mediator (a) -.39* (.19) IV to Mediator (a) -.52** (.20) Mediator to DV (b) .36** (.06) Mediator to DV (b) .37** (.06) Total Effect IV to DV (c) -.47* (.18) Total Effect IV to DV (c) -.42* (.19) Direct Effect IV on DV (c’) -.33† (.17) Direct Effect IV to DV (c’) -.23 (.17) DV Model R2 .169** DV Model R2 .191** -.145 -.289, -.004 -.191 -.379, -.059 N =246, 1000 bootstra p samples N = 26, 1000 bootstra p samples Note. Unstandardized regression coefficients. † p < .10 p < .05 ** p < .01

PAGE 108

93 In summary, the use of mother report for all variables when testing the study hypotheses revealed several si gnificant indirect relationships between WIF and child eating behaviors, none of which were observed when multisource data were used (mother self report and child self-re port). Specifically, these relationships occurred when monitoring was the mediating f eeding behavior, and they oc curred primarily with child eating fruits or vegetables as the dependent variable. Support was also found for mother eating vegetables on work days as a mediat or between WIF and child eating vegetables on school days. The three sets of supplementary analys es involving mothe r-child perceptions reveal an interesting pattern of self-other percep tion. The findings have implications for the theoretical tenets underlying the pres ent study and will be discussed in Chapter Seven: Discussion.

PAGE 109

94 Chapter Seven: Discussion There were two primary aims in the pr esent study. The first aim was to examine the relationships between WIF and speci fic eating behaviors reported by employed mothers, as they relate to hea lth criteria such as BMI. Rela ted to this first aim, household coping strategies were proposed as playing a significant role in the relationship between WIF and eating behaviors. The second aim of the present research was to investigate the crossover of WIF to specific child eating be haviors via mother feeding practices or mother eating behaviors. Major Findings: WIF and Mother Health Limited support was found for the first aim. Work-interferen ce-with-family was negatively associated with eati ng vegetables on work days, but it was not associated with eating fruits on work days, nor was it a ssociated with eating snack foods. The relationship between eating vegetables and WI F is consistent with previous research which observed a negative relationship between WIF and report of eating healthy foods (e.g., vegetables, fruits, fibers, whole gr ains; Allen & Armstrong, 2006). As predicted, eating vegetables on days off from work was not associated with WIF. This offers implications for improving the present theore tical framework of the relationship between WIF and health. It is not possible to determine whethe r the nonsignificant relationship between fruit and WIF in the present st udy is directly inconsistent w ith previous research. The

PAGE 110

95 only other study to directly evaluate WIF a nd healthy eating behaviors used a healthy foods dietary checklist (fruits, vegetables whole grains) that was not designed for dimensional use; therefore it was analyzed in its entirety (Allen & Armstrong, 2006). In addition to the findings from Allen and Armstrong, the hypothesis addressing WIF and fruit was developed from res earch linking perceived stress w ith eating fewer servings of fruit and vegetables (e.g., in adolescents; Cartwright et al., 2003), and from qualitative findings which associated perceptions of in compatible role demands with food choices based on anticipated preparation (Devine et al ., 2006). In the presen t study, it is possible that mothers perceived that different pr eparation effort was required for fruit vs. vegetables (e.g., most fruit requires only rins ing or peeling, while vegetables may need rinsing, peeling, cutting, cooking or prepara tion of a dip if eaten uncooked). This perception might suggest the a voidance of vegetables (but not fruit) when experiencing high WIF. Similar logic follows regarding the pa latability of fruits vs. vegetables; fruits might be perceived to be highl y palatable, while vegetables may be perceived as less palatable, leading to similar levels of fr uit consumption regardless of perceived WIF, because it ‘tastes good’. The lack of significant findings for sn ack foods across all study hypotheses is attributed to ambiguity in the snack food group operationalization, as well as potentially limited value of the food group designation itsel f. Although other studi es have reported increased snacking in res ponse to stress (e.g., Oliver & Wardle,1999), this does not necessarily implicate an increase in c onsuming snack foods as a food group, as hypothesized in the present study. These issues will be further considered in Study

PAGE 111

96 Limitations Evidence was also lacking for the first aim of the study regarding the association between eating behaviors and BMI, as well as the indirect link be tween WIF and BMI via individual eating behaviors (fruits, vegeta bles, snack foods). The absence of a significant indirect relationship between WIF an d BMI is consistent with the findings of Allen and Armstrong (2006) where self-report of WIF was only signifi cantly associated with BMI when the mediating eating variable was dietary fat; wh ile reports of eating more healthy foods (fruits, vegetables, fi ber, whole grains) did not demonstrate a significant relationship with WIF. One reason for this insignificance involves the use of general food categories which may include high and low-fat foods without assessing calories and fat (respondents reporting vegetable consump tion may have eaten zucchini uncooked alone, or as a main ingredient in a high-fat, high-calori e quiche); therefore, indicating a higher intake of vegetables would not necessarily be expected to relate to lower BMI. A second explanation is that the body mass index can vary according to a number of factors beyond eating fruits and vegetables, including (but not limited to) physical activity and genetics. BMI is limite d in its sensitivity to the ratio of lean muscle/fatty tissue (adults with considerable proportions of lean mu scle mass, and adults who are obese could have similar BMI values). Accounting for mother intake of calories or specific food items (rather than food categor ies), and mother level of physical activity in addition to their eating behaviors may cont ribute to a better understanding of WIF and BMI. For example, physical activity may interact with eating behaviors in the relationship with BMI (e .g., moderating effect).

PAGE 112

97 The final objective of the fi rst aim of the present study wa s to explore the role of household coping strategies in the relationshi p between WIF and eating behaviors. The primary moderation hypothesis was not supported. Explorator y analyses of alternate models of coping revealed support for coping as a suppressor of th e relationship between WIF and eating vegetables on work days. De spite a negative direct association between WIF and eating behavior (which remained si gnificant after partiall ing out the variance accounted for by the mediator) the indirect relationship via coping was positive. This pattern suggests the presence of a competi ng process in the relationship between WIF and mother consumption of vegetables. The indirect and direct paths are both meaningful, but care should be taken in in terpreting the total effect of WIF on eating vegetables (Shrout & Bolger, 2002); The tota l effect cannot be explained by the two additive paths from WIF to coping and c oping to eating vegetabl es (Mackinnon et al., 2000). The direct effect should be cons idered conditional, holding household coping strategies constant. In summary, these results provide only partial support for th e first aim of the study. WIF was related to one work-day eating behavior (veg etables), and coping exhibited a meaningful function when examined in an alternate model in the context of WIF and one work-day eating behavior (veget ables). However, ea ting behaviors did not relate to BMI as expected, a nd WIF did not relate indirect ly to BMI as predicted. Major Findings: Mother-child WIF and Health Behavior Investigation of the second aim of the study revealed associations between WIF and feeding behaviors, feeding behaviors a nd child eating behaviors, and supplementary analyses identified interesting relationships among mother and child perceptions. Of the

PAGE 113

98 three feeding behaviors exam ined, monitoring was negatively associated with WIF and no relationships were observed between WIF and pressure or re striction. The reason for non-significant relationships with pressure a nd restriction was not im mediately apparent. The feeding behavior dimensions vary in the ty pe of behavior they represent, such that restriction and pressure seem to represent active interaction from the mother, whereas monitoring might be active or passive. A dditionally, restriction and pressure suggest intervention by the mother, whereas monitoring represents maintaining an awareness of the child’s eating behaviors. Pressure and re striction can only occur when the mother is physically and psychologically available, a nd while WIF may be associated with how often the mother is available, it may not be associated to the extent that the mother perceives restriction and pressu re as important feeding objecti ves or values. By contrast, the awareness implicated by monitoring is lik ely to decrease if the mother experiences incompatible role demands that render her less psychologically available. Finally, the items measuring pressure and restriction focu s primarily on beliefs or values, whereas the monitoring items measure whether the mother monitors her child’s consumption of various foods. Assessing the mother’s rest riction and pressure be haviors rather than restriction and pressure beliefs may have been more relevant for the hypothesized association with WIF. The role of coping with regard to WIF a nd feeding behaviors was also relevant to the second aim of the present study according to the theoretical framework (Figure 6; although no formal hypotheses were made). Re sults of the supplementary analyses for coping and monitoring feeding practices were similar to those f ound for mother eating vegetables on work days. There was no s upport for a moderating coping effect between

PAGE 114

99 WIF and any of the feeding behaviors, but evidence of suppression by coping emerged for monitoring practices, specifically. As w ith WIF and eating vegetables on work days, opposite signs were observed between direct an d indirect effects; the indirect effect between WIF and monitoring was positive, wh ile the direct relationship between WIF and monitoring (c’) was negative. This pa ttern of relationships again suggests the presence of a competing process in th e relationship between WIF and mother consumption of vegetables, and the total eff ect must be interpreted with caution. WIF was negatively related to monitoring when coping was held constant. The crossover hypotheses from the second ai m of the study were not supported in the primary analyses utilizing multisource data. There were no significant relationships between mother-reported feeding behaviors and child self-report of eating fruits and vegetables (when evaluated on all days), beyond a significant relationship between restriction and child eating snack foods. Previous work demonstrated a positive relationship between mother se lf-report of fruit and vegetable consumption and child self-report (grades 4 6) of fruit and vege table consumption in low-income households (Sylvestre, O’Loughlin, Gray-Donald, Hanl ey, & Paradis, 2007). However, the relationship between child consumption of fruits and vegetables and childrens’ perceptions of their parents modeling fruit and vegetable consumption was found to be moderated by perceived fruit and vegetable availability (Young, Fo rs, & Hayes, 2004). Perceived availability of certain foods wa s not measured in the present study and may play a role in the lack of significant association betw een mother and child eating behaviors using multisource data.

PAGE 115

100 The positive association betw een restriction and child re port of eating snack foods in the present study was consistent with theory and research which suggest that restriction practices lead to increased consumption by th e child (Birch et al., 2003). No mediational relationships between WIF and child eating be haviors via mother feeding behaviors or mother eating behaviors were supported by multisource data. However, when these hypotheses were analyzed using mother repor t of all variables in the supplementary analyses, evidence was found for two of the hypot hesized mediation patterns. The results supported full mediation for mother eating beha viors (mother-child eating vegetables on work/school days) and monitoring (child eati ng fruits and vegetables) with an overall indirect effect that was negative between WIF and child eating behaviors. In comparison with the nonsignificant results of the multisource data, the significant relationships resu lting from single source mother-report may suggest an inconsistent frame of reference across moth er and child for repor ting the child’s eating behaviors. Supplementary analysis of mother and child perceptions revealed significant agreement between mother and child report of child eating behavior, but the effect sizes were modest ( r = .17 to .32). Previous research re ported correlations of .28 .47 were found between child self-report of fruit and ve getable consumption and parent report of child fruit and vegetable consumption, which are slightly stronger in magnitude to the agreement observed in the present study. (Ta k, te Velde, de Vies, & Brug, 2006). The variance unaccounted for between the two s ources on any given behavior may be a function of the respondent’s frame of refere nce, influenced by phenomena such as 1) differing adult-child interpretation of respons es on a Likert-type frequency scale (e.g., what are the adult and child interpretation of ‘most days’ vs. ‘some days’), 2) differing

PAGE 116

101 adult-child interpretations of what one ‘u sually’ eats behaviors (e.g., adults may hold broader or more long–term perspective on what is usual, and child may focus on the most recent behaviors), and 3) the use of different strategies for responding to the items (e.g., A mother may mentally catalogue what has be en eaten or served over the preceding two weeks, then select an appropriate answer. A child may rely on a global perception such as “my mother likes fruit, so she eats it as a snack every day”). Th e literature is not in agreement about the validity of child self-report of dietary intake, s uggesting that selfreporting of dietary intake by children is of ten subject to under-r eporting, and this phenomena is related to body weight than age group or dietary survey technique (Livingstone, Robson, & Wallace, 2004). A nother review concluded that there was higher validity of child report for specific survey techniques (e.g., food recall, food reporting) over others (e.g., food frequency que stionnaires), and no systematic difference in reporting according to age among children aging 6 10 years (McPherson, Hoelscher, Alexander, Scanlon, & Serdula, 2000). Other findings suggest that starting around the age of 8 years, children quickly develop the abil ity to report their own food intake, and that this reporting is reliable by age 10 (Food Shar e Education & Research Office of Toronto, n.d., cf Sylvestre et al., 2007). Finally, the agreement between selfand otherreport may be bound by the counterpart having a li mited opportunity to observe all of the referent’s eating behaviors, an issue that will be explored further in Supplementary Findings: Mother-Child Perceptions In summary, support was observed for se veral instances of the WIF – health crossover targeted by the sec ond aim of the study. However, this should be considered preliminary evidence because the relations hips were supported only by single-source

PAGE 117

102 data. Complementary methodology (for example, comprehensive checklists of specific food items rather than categories, other-repor t provided by another adult rather than by the child, food and/or caloric intake diaries rather than Likert-type response scales, observation rather than self-re port) is needed in order to determine whether the singlesource results represent meaningful relations hips or a methodological artifact such as common method variance or so cial-desirabil ity responding. Supplementary Findings: Mo ther-child perceptions As previously noted, the supplementary an alyses helped to id entify a meaningful pattern of coping relationships, as well as preliminary support for crossover hypotheses using single source data from mothers. The supplementary analyses also presented some interesting information about mother and chil d perceptions. First, the mother and child report of each child eating behavior were significantly correlated. Significant relationships were also observed between mo ther and child reports of each mother eating behavior (and these relationships tended to be slightly stronger than the correlations for child eating behavior). The magnitude of these correlations was moderate, suggesting that mother report of child eating behavior was corroborated by child self-report of eating behavior, and vice versa for mo ther eating behavior. Indee d, research has demonstrated that parents can report accurately about child fruit and vegetable consumption (preschool intake on the previous day, Linneman, He ssler, Nanney, Steger-May, Huynh, & YhaireJoshu, 2004). Next, individuals reported similar eating behaviors between themselves and their counterpart (mothers reported similar leve ls of eating between themselves and their children, children reported similar levels of eating between th emselves and their

PAGE 118

103 mothers). It cannot be determined from the survey data alone whet her these similarities in reporting about self and reporting one’s c ounterpart represent a true similarity in behavior between mother and child, or if it is merely perceptual (i.e., each reports similarity between themselves and their counte rpart based on an assumption or belief that they are similar to each other, but in reality the behaviors are less si milar than reported). Certainly, if the reported similarity is indi cative of objective similarities in mother-child behavior (similar behaviors reported because mother and child exhibit similar eating behaviors) it would lend credence to the in ternal validity of the mediation results supported by single-source mother-report. On the other hand, if the similarity is so lely perceptual (sel f and other behaviors reported as similar, but exhibi ted behaviors are not actually similar), the present findings are still noteworthy. Future decisions that mo thers make about what to eat and what to feed their child are likely to be influenced by their perc eption of what the child is currently eating and how similar it is to thei r own eating behaviors. From a behavior modeling perspective, the issue of whether the mother recogni zes the similarity as being causal is even more intriguing (e.g, “my child and I eat pretty much the same foods” vs. “If I eat more fruits and vegetables, then my child will eat more fruits and vegetables”). The child’s perception of similarity betw een the mother and the child’s own eating behaviors is also likely to be associated with future deci sion making about what to eat, although this is less likely to occur if the ch ild does not wield much control over what he or she eats (e.g., younger children may not have the opportunity to select or refuse foods that are served or accessible). While th ese propositions are sp eculative and cannot be

PAGE 119

104 inferred from the current data, future re search examining these issues would be worthwhile. Also of interest, the similarity betwee n selfand other-eating behavior according to mother-report, and between selfand othe r-eating behavior according to child-report, was stronger for “off-“ day eating behaviors ( non work days, weekends) than for “on-“ days (work days, school days). In other wo rds, mothers reported more similar eating behaviors between themselves and their children on off-days/weekends; children reported more similar eating behaviors between them selves and their mothers on weekends./offdays. Assuming that the survey data is valid, there are competing explanations about the lesser degree of similarity between mother and child eating beha viors on work days. First, the different locations of mother and child are likely to be asso ciated with different respective food options for each person (e.g., ch ild eats lunch in cafeteria, or mother packs the child a bag lunch, eith er of which could be quite di fferent from what the mother eats at home or at work for lunch that day). A second explanation might involve deliberate efforts by the mother on work days to maintain a certain quality of diet for the child that is prioritized over maintaining the same quality of her own diet. These possible explanations demonstrate the relevance of the issue for further theoretical development of WIF crossover to child health behaviors. Understanding this trend could assist in the development of boundary conditions for the meditational patterns on work days, as well as with developing a model of these patterns across time. However, it is also possible that the perceived stronger similarity on weekends is completely inaccurate; because child and mother are more likely to be phys ically away from each other during certain

PAGE 120

105 meal times on work/school days it is simply more difficult to accurately report about the other eats. In this case, it would be possible that mother-chi ld behaviors are actually just as similar on work days as they are on non-work days, but the behaviors are being reported inaccurately because the ‘other’ is not able to observe the referent as much. Study Strengths and Limitations. There are several strengths of the pres ent study, including a cross-disciplinary foundation in theory and research, and ev aluation of a new WIF crossover process between parent and child. The analytic proced ure (bootstrapping indire ct effects) offered relaxed assumptions regarding a normal distri bution, conservative estimates, and the need for large samples (e..g, Fritz & Macki nnon, 2007; Mackinnon, Warsi, & Dwyer, 1995; Shrout & Bolger, 2002), findings were triangulated with a more conservative type of analysis (Sobel test), and the results were generally consistent be tween the two methods across the hypotheses. The response rate s observed in recruitment and survey administration were very high, and the collection of multi-s ource data allowed comparison between results from multi-source da ta and results from single-source data. Several limitations of the pr esent study are also importa nt to note. The study was cross-sectional and no variables were mani pulated, therefore di rectional and causal inferences cannot be supported. Survey me thodology was used to examine behaviors and perceptions related to a topic with considerable social desirability. However, these data offer insight to the subjective experiences between mothers and children, and they may predict future behavior, contributing to the development of longitudinal propositions in this area. Survey methodology also offers lo w cost and high feasib ility, appropriate for initial exploration of a rela tively new domain. Another potential limitation involved the

PAGE 121

106 use of non-validated scales to measure eating be havior categories that were not sensitive to caloric or fat content. Pilot research contributed to the improvement of these measures, but validation against objective measurement is warranted. In retrospect, power may have been an issue for the media tion analyses, as effect sizes were smaller than anticipated (minimum sample size calcul ated using small-medium effect sizes (.26) but small effect estimations may have been more realistic). Fr itz and Mackinnon (2007) recommend a sample of 368 to 450 responde nts / dyads for bootstrapping and Sobel procedures, respectively, when IV to medi ator and mediator to DV effect sizes are expected to be small (.14). Perhaps the most important limitation was the operationalization of the snack foods group. The example foods fitting this category ranged from junk food (chips) to breakfast ba rs (granola bars), comprising a group that was likely ambiguous to respondents and extr emely limited in meaning for health outcomes (potentially including both high a nd low calorie, high and low fiber, high and low fat foods). The theme of conveni ence was certainly manifest in the operationalization, but the health outcome imp lications of eating snack foods could be difficult to determine; however, simila r food groupings have shown significant association to child and adult indicators of obesity in prev ious research (e.g., Fisher & Birch, 2002; McCarthy et al., 2006; “savory snacks” grouping: popcorn, potato chips, tortilla chips, puffed corn snacks, pretzels, peanuts). Theoretical Implications and Fu ture Research Directions. The results of the present study hold seve ral implications for the development of the theoretical framework in the presen t study, although replication of the findings reported here is recommended. First, coping may be reconceptualized as a suppressor

PAGE 122

107 variable, in the context of eating behavi ors and feeding behaviors. Second, the designation of work/school day in the mother role-modeling paths of the model warrants consideration as a boundary condi tion to the modeled relationships. Third, based on the results of the present study, it would appear that the vegeta bles group is the only mother eating behavior that deserves representation in the framework. However, it is not clear why the relationships manifested with vegetables It is imperative to further explore the characteristics of foods that may have driven the findings in this study, with respect to perceived effort, palatability, and cost or availability of the food. Future directions for empirical research in this area include incorporating other self-report scales of eating behaviors, different self-repo rt (e.g., experience sampling) and objective (e.g., observation) data collection me thods, and replicating the study in a more heterogeneous sample of mothers (e.g., all mo thers in the present sample had enrolled their children in a YMCA after-school pr ogram, which may be indicative of income, social support, and values towards physical activ ity). It would also be useful to target older child age ranges to obser ve the relationships in chil dren with more autonomy for choosing what to eat. Beyond replication a nd extension to isolat e boundary conditions of the proposed relationships, researchers are encouraged to deve lop the theoretical framework by incorporating the role of fathers and family structures. Given traditional gender roles, perhaps interacti on between fathers and children is more likely to elicit the role-modeling path, whereas mothers and childr en exhibit the feedi ng path. Identifying the domestic roles that are speci fic to the family unit might reveal that the ‘breadwinner’ functions as the role model, regardless of wh ether it is the mother or father. Additionally, the composition of the family unit is extremely relevant to the issue of coping (e.g., adults

PAGE 123

108 may engage in more or less coping according to the extent that family responsibilities are shared, and to the extent that social/mar ital/family support resources are available). The characteristics of work days which c ontribute to the dissimilar mother-child eating behavior perceptions must be iden tified (e.g., physical space or intentional decisions), and the dissimilarity examined with supplementary data (e.g., observation, other-report by other adults). It will also be necessary to identify other feeding practices that share some of the relevant characte ristics of monitoring, whereas pressure and restriction might be discarded as considerably less relevant to the WIF – child crossover. Regarding the mothers’ eating behaviors, the characteristics of speci fic food groups must be further explored in order to strengthen future hypotheses (e.g., determine whether the difference in WIF findings between fruits and vegetables was due to perception of required effort vs. palatability). Finally, the constructs in the present theoretical framework were intended to represent behavi ors; exploratory anal ysis of mother and child perspectives suggested that a compre hensive understanding of exhibited behaviors over time may require examining the deci sions which drive those behaviors (e.g., awareness and salience of one’s own eating be havior leading to a decision to change one’s habits or keep them the same, awaren ess and salience of the eating behaviors of one’s counterpart leading to a decision to change or stay the same). Following this, the characteristics of other relevant variable s (type of WIF experienced, coping strategy employed, and characteristic of work day vs. weekend) may be more strategically pursued in theory and empirical research.

PAGE 124

109 Implications for Practice It is arguably premature to target these findings with organiza tional initiatives. However, the preliminary evidence from moth er self-report suggest s several avenues for intervention that will be relevant upon re plicating the present results and further developing the WIF-health domain. Future or ganization efforts might adopt one or more of three primary objectives: 1) attempt to al leviate or reduce WIF, 2) educate employees, and 3) equip or train employ ees to adopt effective coping strategies. Amassing support for the association between WIF and the health of employees and their families, the bottom line impact of employer-contributions to health care (in an era of obesity in the U.S.) may provide the necessary rationale for organizations to implement family-friendly policy targeted at WIF (Allen, 2001). Beyond expensive policy interventions, organizations could easily support employ ee awareness initiatives regarding the implications of WIF for employee and family health, and the beha viors that are easily modified to improve various outcomes. Al ong the same lines, employers could sponsor in-house or external training efforts to equi p employees with the ability to engage in effective coping strategies (e.g., resource accumulation, weighing which factors to sacrifice when necessary with regard to eating and feeding others in the family). By adopting initiatives that target work -life issues, organi zations may benefit from reduced health-care premiums for em ployees with full-time benefits, and the reputation of the employer is likely to benefit from perceptio ns of work-life responsibility for employees (e.g., placement in Working Mothers Magazine “Top 100 Companies to Work For”).

PAGE 125

110 Conclusion Existing research has identified relations hips between WIF and health outcomes (e.g., overall ratings of physical health; Adams & Jex, 1999; Allen & Armstrong, 2006; Frone, Russell & Barnes, 1996; Grandey & Cropanzano, 1999; Mesmer-Magnus & Viswesvaran, 2005), but little was known a bout the relationship between WIF and behaviors that are re levant to these health outcom es. Beyond adult WIF and health, experts have called for research to examine how employment issues impact parent and child health (e.g., Cleveland, 2005; Friedman & Greenhaus, 2000; Galambos, Sears, Almeida & Kolaric, 1995; Kinnunen & Pukki nen, 2001). Building upon the one study that examined WIF and dietary behaviors ( eating healthy foods, dietary fat; Allen & Armstrong, 2006), the present study investigated the association between mother WIF and eating behaviors, the role of household copi ng strategies, and the crossover of mother WIF to child eating behaviors using multisource data. Results provided evidence of a negati ve relationship between WIF and eating healthy foods (vegetables), cl arifying the relevance of work-day (vs. off-days). A competing process emerged via household coping strategies, manifes ting as suppressor of the WIF relationships. The study findings al so revealed a negativ e association between WIF and feeding practices (monitoring behavior s), and this relationship was also subject to suppression by household coping strategies Support for the hypothesized crossover from mother WIF to child eating behavior (via mother eating and monitoring) was observed in the mother-report data. Alt hough there was significan t agreement between mother and child report of ea ting behaviors, the proposed cr ossover relationships were not supported by multisource data. Therefore, until additional research accumulates, the

PAGE 126

111 majority of these relationsh ips are confined by the boundaries of mother perception. In conclusion, the present study contributes to the work-family and health literature by further clarifying the processe s that link WIF with health, and by providing preliminary evidence of crossover between mother WIF and child health. The continued study of this area is likely to strengthen support for the rele vance of work-family i ssues to “the bottom line" (e.g., employer contributions to health insurance), provi ding even stronger rationale for organizations to implement family -supportive policies and benefits.

PAGE 127

112 References Adams, G.A., & Jex S.M. (1999). Relations hips between time management, control, work-family conflict, and strain. Journal of Occupational Health Psychology, 4, 1, 72-77. Agras, W.S., Berkowitz, R.L., Hammer, L.C ., & Kraemer (1988). Relationships between the eating behaviors of pa rents and their 18-month-ol d children: A laboratory study. International Journal of Eating Disorders 7, 461-468. Ahye, B.A., Devine, C.M., & Odoms-Young, A.M. (2006). Values expressed through intergenerational family food and nutri tion management systems among African American women Family Community Health 29 (1), 5-16. Allen, T. D. (2001). Family-supportive work environments: The role of organizational perceptions. Journal of Vocational Behavior, 58 414-435. Allen, T.D., & Armstrong, J. (2006). Furthe r examination of the link between workfamily conflict and physical health: The role of he alth-related behaviors. American Behavioral Scientist 49(9), 1204-1221. Allen, T. D., Herst, D. E. L., Bruck, C. S., & Sutton, M. (2000). Consequences associated with work-to-family conflict: A review and agenda for future research. Journal of Occupational Health Psychology 5, 278-308.

PAGE 128

113 Allen, T.D., Shockley, K., & Poteat, L.F. (2008). Family supportive policies, workfamily conflict, and the family dinner. Journal of Vocational Behavior 70(2), 336. Bailyn, L., Drago, R. and Kochan, T.A. (2001). Integrating work and family life: A holistic approach, Sloan Work–Family Policy Ne twork, MIT Sloan School of Management, Boston. Baltes, B.B. & Heydens-Gahir, H.A. (2003). Reductions of work-family conflict through the use of selection, optimization, and compensation behaviors. Journal of Applied Psychology 88(6), 1005-1018. Barber, J.G., & Delfabbro, P.H. (2000). Predicto rs of adolescent adjustment: parent-peer relationships and parent-child conflict, Child and Adolescent Social Work Journal 17, 275-288. Barling, J., MacEwen, K. E., & Nolte, M. L. (1993). Homemaker role experiences affect toddler behaviors via maternal we ll-being and parenting behavior. Journal of Abnormal Child Psychology 21 213-229. Birch, L.L. (2006). Commentary Child feed ing practices and the etiology of obesity Obesity 14, 3, 343-344. Birch, L.L., & Fisher, J.A. (1995). Appetite and eating behavior in children. In: Gaull, G.E. (Ed.) The Pediatric Clinics of North America: Pediatric Nutrition Philadelphia, PA: WB Saunders, 931–953. Birch, L.L., & Fisher, J.O. (1998). Developm ent of eating behaviors among children and adolescents. Pediatrics 101, 539–49.

PAGE 129

114 Birch, L., Fisher, J., & Davison, K. (2003) Learning to overea t: Maternal use of restrictive feeding practices promotes girls’ eating in th e absence of hunger. American Journal of Clinical Nutrition 78, 215–220. Birch, L.L., Fisher, J.O., Grimm-Thomas, K.G., Markey, C.N., Sawyer, R., & Johnson, S.L. (2001). Confirmatory factor analys is of the Child Feeding Questionnaire: a measure of parental attitudes, beliefs a nd practices about child feeding and obesity proneness. Appetite 36, 201-210. Block, Patterson, & Subar (1992). Fruit, vegeta ble, and cancer preven tion – a review of the epidemiological evidence. Nutrition and Cancer 18, 1-29. Bond, J. T., Thompson, C., Galinsky, E., & Prottas, D. (2003). Highlights of the National Study of the Changing Workforce, No. 3 New York: Families and Work Institute. Brett, J. M., & Stroh, L. K. (2003). Work ing 61 plus hours a week: Why do managers do it? Journal of Applied Psychology 88 67-78. Borah-Giddens, J. & Falciglia, G.A. (1993). A meta-analysis of the relationship in food preferences between pa rents and children. Journal of Nutrition Education, 25, 102–107. Bray, G.A., & Popkin, B.M. (1998). Diet ary fat intake does affect obesity. American Journal of Clinical Nutrition 68, 1157-1173. Brett, J. M., & Stroh, L. K. (2003). Work ing 61 plus hours a week: Why do managers do it? Journal of Applied Psychology 88 67-78. Brisson, C., Laflamme, N., Moisan, J., Milot, A., Masse, B., & Vezina, M. (1999). Effect of family responsibilities and job st rain on ambulatory blood pressure among white-collar women. Psychosomatic Medicine 61, 205-213.

PAGE 130

115 Burke, V., Beilin, L.J., Durkin, K., Stritzke, W.G., Houghton, S. & Cameron, C.A. (2006). Television, computer use, physical activity, diet and fatness in Australian adolescents. International Journal of Pediatric Obesity, 1(4), 248-255. Burke, R. J., & Greenglass, E. R. (2001). Ho spital restructuring stressors, work–family concerns and psychological we ll-being among nursing staff. Community, Work and Family 4, 49–62. Byron, K. (2005). A meta-analytic review of work-family conflict and its antecedents. Journal of Vocational Behavior 67(2), 169-198. Cartwright, M., Wardle, J., Steggles, N., Sim on, A.E., Croker, H., & Jarvis, M.J. (2003). Stress and dietary practices in adolescents. Health Psychology 22(4), 362-369. Center for Disease Control (2007). A bout BMI for adults. Retrieved from http://www.cdc.gov/healthyweight/a ssessing/bmi/adult_bmi/index.html Octoboer 13, 2007 Cleveland, J. B. (2005). What is success? Who defines it? Perspe ctives on the criterion problem as it relates to work and family. In E. E. Kossek & S. J. Lambert (Eds .), Work and life integration: Organizati onal, cultural, and in dividual perspectives (pp. 319-345). Mahwah, NJ: Lawrence Erlbaum Associates Crouter, A.C., & Bumpus, M.F. (2001). Linki ng parents’ work stress to children’s and adolescents’ psychological adjustment. Current Directions in Psychological Science (10)5, 156-159. Daniels, S.R. (2006). The consequences of childhood overweight and obesity. The Future of Children 16(1), 47-67.

PAGE 131

116 Davison, K.K., & Birch, L.L. (2001). Ch ildhood overweight: a c ontextual model and recommendations for future research. Obesity Reviews, 2, 159-171. Davison, K.K., Francis, L.A., & Birch, L.L. (2005). Reexamining Obesigenic Families: Parents' Obesity-related Behaviors Predict Girls' Change in BMI. Obesity Research, 13, 1980-1990. Devine, C.M., Connors, M.M., Sobal, J., & Bisogni, C.A. (2003). Sandwiching it in: spillover of work onto food choices and family roles in lowand moderateincome urban households. Social Science and Medicine 56, 617-690. Devine, C.M., Jastran, M., Jabs, J., Wethi ngton, E., Farell, T.J., & Bisogni, C.A. (2006). 'A lot of sacrifices'-Work-family spill over and food choice coping strategies of low-wage employed parents. Social Science and Medicine, 63, 2591-2603. Duxbury, L. & Higgins, C. (2001). Work-life ba lance in the new millennium: Where are we? Where do we need to go? CPRN Discussion Paper No. W/12. Eby, L.T., Casper, W.J., Lockwood, A., Bordeaux, C., & Brinley, A. (2005). Work and family research in IO/OB: Content an alysis and review of the literature. Journal of Vocational Behavior, 66, 124-197. Eriksen, H. R. & Ursin, H. (1999). Subjectiv e health complaints: Is coping more important than control? Work & Stress 13, 238–252. Faith, M. S., Scanlon, K. S., Francis, L. A., Sherry, B. S., & Birch, L. L. (2004). Parentchild feeding strategies and their relations hips to child eating and weight status. Obesity Researc h, 12(11) 1711-1722. Fisher, J., & Birch, L. (1999). Restricti ng access to foods and children’s eating. Appetite 32, 405–419.

PAGE 132

117 Fisher, J., & Birch, L. (2000). Pa rents’ restrictive feeding pr actices are associated with young girls’ negative self -evaluation of eating. Journal of the American Dietetic Association 100, 1346. Fisher, J., & Birch, L. (2002). Eating in th e absence of hunger and overweight in girls from 5 to 7 years of age. The American Journal of Clinical Nutrition 76(1), 226. Ford, M.T., Heinen, B.A., & Langkamer, K.L. (2007). Work and family satisfaction and conflict A meta-analysis of cross-domain relations. Journal of Applied Psychology 92(1), 57-80. Fortes-Ferreira, L., Peiro, J.M., Gonzalez -Morales, M.G., & Martin, I. (2006). Workrelated stress and well-being the roles of direct action copi ng and palliative coping. Scandinavian Journal of Psychology 47, 293-302. French, S.A., Story, M., Neumark-Sztainer, D., Fulkerson, J.A., & Hannan, P. (2001). Fast food restaurant use among adolescents : Associations with nutrient intake, food choices and behavioral and psychosocial variables. International Journal of Obesity 25, 1823-33. Friedman, S. D., & Greenhaus, J. H. (2000). Children: Unseen stakeholders at work. Work and Family – Allies or Enemies ? (pp. 69-83). Oxford. Fritz, M.S., & MacKinnon, D.P. (2007). Requi red sample size to detect the mediated effect. Psychological Science 18(3), 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). Washington, DC: American Psychological Association.

PAGE 133

118 Frone, M.R., Barnes, G.M.,& Farrell, M.P. (1994). Relationship of work-family conflict to substance use among employed mother s The role of negative affect Journal of Marriage and the Family 56, 4, 1019-1030. Frone, M. R., Russell, M., Barnes, G. M. (1996). Work–family conflict, gender, and health-related outcomes-A study of employed parents in two community samples. Journal of Occupational Health Psychology 1, 57–69. Frone, M.R., Russell, M., & Cooper, M.L. ( 1992). Antecedents and outcomes of workfamily conflictTesting a model of the work-family interface. Journal of Applied Psychology 77, 65-78. Frone, M.R., Russell, M., & Cooper, M.L. (1992b) Prevalence of work-family conflict Are work and family boundaries asymmetrically permeable. Journal of Organizational Behavior 13, 7, 723-729. Frone, M. R., Russell, M., & Cooper, M. L. (1993). Relationship of work-family conflict, gender, and alcohol expect ancies to alcohol use/abuse. Journal of Organizational Behavior, 14 545-558. Frone, M. R., Russell, M., & Cooper, M. L. ( 1997). Relation of work-family conflict to health outcomes: A four-year longit udinal study of employed parents. Journal of Occupational and Organizational Psychology, 70 325-335. Galambos, N. L., Sears, H. A., Almeida, D. M., & Kolaric, G. C. (1995). Parents’ work overload and problem behavi or in young adolescents. Journal of Research on Adolescence, 5, 201–223. Galinsky, E. (2000). Ask the Children New York: Quill.

PAGE 134

119 Galloway, A. T., Fiorito, L., Lee, Y., Birch, L. L. (2005) Parental pressure, dietary patterns, and weight status in girls who are “picky eaters”. Journal of the American Dietetic Association 105, 541-548. Gillis, L.J., & Bar-Or, O. (2003). Food away from home, sugar-sweetened drink consumption and juvenile obesity. Journal of the American College of Nutrition 22, 559-45. Gillman, M., Rifas-Shiman, S., Frazier, A., Ro ckett, H., Camargo, C., Field, A., Berkey, C., & Colditz, G. (2000). Family dinner and diet quality among older children and adolescents. Arch Family Medicine 9, 235-240. Goldstein, I.B., Shapiro, D., Chicz-DeMet, A ., & Guthrie, D. (1999) Ambulatory blood pressure, heart rate, and neuroendocrine responses in women nurses during work and off work days. Psychosomatic Medicine 61, 387-296. Grandey, A. A., Cropanzano, R. (1999). The c onservation of resources model applied to work-family conflict and strain. Journal of Vocational Behavior 54, 350-370. Greenhaus, J.H., Allen, T.D., & Spector, P.E ., (2006). Health consequences of workfamily conflict: The dark side of the work-family interface. In Employee Health, Coping and Methodologies: Research in Occupational Stress and Well Being, Vol 5. Greenhaus, J.H., & Beutell, N.J. (1985). S ources of conflict between work and family roles. The Academy of Management Review 10(1), 76-88. Grzywacz J.G. (2000). Work-family spillover and health during midlife. Is managing conflict everything. American Journal of Health Promotion 14(4), 236-243.

PAGE 135

120 Grzywacz, J.G., & Marks, N.F. (2001). Social inequalities and exercise during adulthood: Toward an ecological perspective. Journal of Health and Social Behavior 42(2), 202-220. Hagdrup, N.A., Simoes, E.J., & Brownson, R.C. (1998). Fruit and vegetable consumption in Missouri: Knowledge, barriers and benefits. American Journal of Health Behavior 22(1), 90-100. Halpern, D. (2004) Public Policy, Work, and Fam ilies: The Report of the APA Presidential Initiative on Work and Families Washington, D.C.: American Psychological Association. Hammer, L. B.; Allen, E.; & Grigsby, T. D. (1997). Work and family conflict in dualearner couples: Within-indivi dual and crossover effect s of work and family. Journal of Vocational Behavior, 50 (2), 185-203. Harrell, W.A. (1995). Husbands' involvement in housework: Effects of relative earning power and masculine orientation. Psychological Reports 77(3), 1331-1337. He, K., Hu, F.B., Colditz, G.A., Manson, J.E., Willett, W.C., & Liu, S. (2004). Changes in intake of fruits and vegetables in re lation to risk of obes ity and weight gain among middle-aged women. International Journal of Obesity 28, 1569-1574. Hellerstedt, W., & Jeffery, R. (1997). The associ ation of job strain and health behaviours in men and women. International Journal of Epidemiology 26(3), 575–583. Hirayama, T. (1994). Lifestyle and mort ality The healthiest way to live. Homeostasis in Health and Disease 35 (4-5), 168-179. Hu, F.B., & Willett, W.C. (2002). Optimal diets for prevention of coronary heart disease. Journal of the American Medical Association, 288, 2569-2578.

PAGE 136

121 Jabs, J., & Devine, C.M. (2006). Time s carcity and food choice s: An overview. Appetite 47, 196-204. Jabs, J., Devine, C.M., Bisogni, C.A., Farrell, T.J., Jastran, M., & Wethington, E. (2007). Trying to find the quickest way Employed mothers' constructions of time for food. Journal of Nutrition Education and Behavior 39, 18-25. Johnson, S.L., & Birch, L.L. (1994). Parents’ and children’s adiposit y and eating style. Pediatrics 94, 653– 61. Judge, T., Boudreau, J., & Bretz, R. (1994). J ob and life attitudes of male executives. Journal of Applied Psychology 79, 767-782. Kaur, H., Li, C., Nazir, N., Choi, W.S., Re snicow, K., Birch, L.L., & Ahluwalia, J.S. (2006). Confirmatory factor analysis of the child-feedi ng questionnaire among parents of adolescents. Appetite, 47, 36-45. Kinnunen, U., & Mauno, S. (1998). Antecedents and outcomes of work-family conflict among employed women and men in Finland. Human Relations 51, 2, 157. Kinnunen, U., & Pukkinen, L. (2001). Linking j ob characteristics to parenting behaviors via job-related affect. In Jan Gerris (ed) Dynamics of Parenting Leuven: Garant. Klesges, R.C., Eck, L.H., Hanson, C.L., Haddoc k, C., & Klesges, L.M. (1990). Effects of obesity, social interac tions, and physical environment on physical activity in preschoolers. Health Psychology 9, 435-449. Klitzman, S., House, J.S., Israel, B.A., & Mero, R.P. (1990). Work stress, nonwork stress, and health. Journal of Behavioral Medicine 13, 3, 231-243. Kossek, E.E., & Ozeki, C. (1999). Bridging th e work-family policy and productivity gap: A literature review. Community, Work, & Family 2 (1), 7-32.

PAGE 137

122 Lam, C.B., & McBride-Chang, C.A. ( 2007). Resilience in young adulthood-The moderating influences of gender-related pe rsonality traits and coping flexibility. Sex Roles 56(3-4), 159-172. Laskarzewski, P., Porrison, J. Khoury, P., Ke lly, K., Glatfelter, L., Larsen, R. and Glueck, D. (1980) Parent-child nutrient intake interrelati onships in school children ages 6 to 19: the Pr inceton School District Study. American Journal of Clinical Nutrition, 33, 2350-2355. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping New York: Springer. Linneman, C., Hessler, K., Nanney, S., Steg er-May, K., Huynh, A., Yhaire-Joshu, D. (2004). Parents are accurate reporters of their preschool ers’ fruit and vegetable consumption under limited conditions. Journal of Nutrition Education and Behavior 36, 305-308. Liu, S., Manson, J.E., Lee, I., Cole, S.R., Hennekens, C.H., Willett, W.C., & Buring, J.E. (2000). Fruit and vegetable in take and risk of cariovasc ular diseasethe Women's Health Study. American Journal of Clinical Nutrition 72, 922-928. Livingstone, M.B.E., Robson, P.J., & Wallace, J. M.W. (2004). Issues in dietary intake assessment of children and adolescents. British Journal of Nutrition 92, S213 – S222. Lukaski, C. (2001) Body mass index, bi oelectrical impedance analysis, and body composition. Nutrition, 17. 1, 55-56. Macht, M. & Simons, G. (2000). Emo tions and eating in everyday life. Appetite 35, 6571.

PAGE 138

123 MacEwen, K.E., & Barling, J. (1991). Effect s of maternal employ ment experiences on children’s behavior via mood, cognitive difficulties and parenting behavior. Journal of Marriage and the Family 53, 635-644. MacKinnon, D. P., Warsi, G., & Dwyer, J. H. (1995). A simulation study of mediated effect measures. Multivariate Behavioral Research, 30 41-62. Major, D.A., Cardenas, R.A., & Allard, C.B. (2004). Child health: A legitimate business concern. Journal of Occupational Health Psychology 9(4), 306-321. McCann, B., Warnick, R., & Knopp, R. (1990). Changes in plasma lipids and dietary intake accompanying shifts in pe rceived workload and stress. Psychosomatic Medicine 52 ,97–108. McCarthy, S.N., Robson, P.J., Livingstone, M. B., Kiely, M., Flynn, A., Crann, G.W., & Gibney, M.J. (2006). Associations be tween daily food intake and excess adiposity in Irish adults: Towards th e development of food-based dietary guidelines for reducing the prevalen ce of overweight and obesity. International Journal of Obesity 30, 993-1002. McLoyd, V.C., & Wilson, L. (1991). The strain of living poor: Pare nting, social support and child mental health. In A.C. Huston (Ed.), Children in poverty: Child development and public policy ( pp. 105-135). New York: Cambridge University Press. McPherson, R.S., Hoelscher, D.M., Alexande r, M., Scanlon, K.S., & Serdula, M.K. (2000). Dietary assessment methods among school-aged children: Validity and reliability. Preventative Medicine 31, S11-S33.

PAGE 139

124 Menaghan, E.G. & Merves, E.S., 1984. Coping with occupational problems: The limits of individual efforts. Journal of Health and Social Behavior 25, 406–423. Mesmer-Magnus, J.R., & Viswesvaran, C. ( 2005). Convergence be tween measures of work-t0-family and famnily-to-work c onflict A meta-analytic examination. Journal of Vocational Behavior 67, 215-232. Netemeyer, R.G., Boles, J.S., & McMurrian, R. (1996). Development and validation of work-family conflict and family-work conflict scales. Journal of Applied Psychology 81, 400-410. Ng, D.M., & Jeffery, R.W. (2003). Relations hips between perceived stress and health behaviors in a sample of working adults. Health Psychology 22(6), 638-642. Niemeier, H.M., Raynor, H.A., Lloyd-Richard son, E.E., Rogers, M.L., & Wing, R.R. (2006). Fast food consumption and breakfast skipping Predictors of weight gain from adolescence to adulthood in a na tionally representative sample. Journal of Adolescent Health 39, 842-849. Ogden, C.L., Carroll, M.D., Curtin, L.R., Mc Dowell, M.A., Tabak, C.J., & Flegal, K.M.. (2006). Prevalence of over weight and obesity in the United States, 1999--2004. Journal of the American Medical Association 295:1549--55. Oh, K., Hu, F.B., Manson, J.E., Stampfer, M. J.,, & Willett, W.C. (2005). Dietary fat intake and risk of coronary heart dise ase in women20 yrs of follow up of the Nurses' health study. American Journal Epidemiology 161, 7, 672-679. Oliver, G., & Wardle, J. (1999). Perceived effects on stress on food choice. Physiology and Behaviour, 66, 511–515.

PAGE 140

125 Oliveria, S., Ellison, R., Moore, L., & Gillman, M. (1992). Pare nt child relationships in nutrient intake: The Framingham children's Study. American Journal of Clinical Nutrition 56, 594–598. Osipow, S. H., & Davis, S. (1988). The relatio nship of coping resour ces to occupational stress and strain. Journal of Vocational Behavior 32(1), 1-15. Pak, S. K., Olsen, L. K., & Mahoney, B. S. ( 2000). The relationships of health behaviors to perceived stress, job satisfaction, and role modeling among health professionals in South Korea. International Journal of Community Health Education 19, 65– 76. Parasuraman, S., & Cleek, M.A. (1984). Coping behaviors and mangers’ affective reactions to role stressors. Journal of Vocational Behavior 24, 179-193. Parkes, K.R. (1990). Coping, NA, and the wo rk environment Add itive and interactive predictors of mental health. Journal of Applied Psychology 75(4), 399-409. Patock-Peckham, J.A., Morgan-Lopez, A.A. (2006). College Dr inking Behaviors: Mediational Links Between Parenting St yles, Impulse Control, and AlcoholRelated Outcomes. Psychology of Addictive Behaviors 20(2), 117-125. Patterson, T.L., Rupp, W., Sallis, I.F., Atkins, C.I., & Nader P.R. (1988). Aggregation of dietary calories, fats, and sodium in Mexican-American and Anglo families. American Journal of Preventative Medicine 8(4),75-82. Penley, J.A., Tomaka, J., & Wiebe, J.S. ( 2002). The association of coping to physical and psychological health outcome sA meta-analytic review. Journal of Behavioral Medicine 25, 6, 551-603.

PAGE 141

126 Pearlin, L.I., Menaghan, E.G., Lieberman, M. A., & Mullan, J.T. (1981). The stress process. Journal of Health and Social Behavior 22, 337-356. Perlin, L., Schooler, C. (1978) "The structure of coping." Journal of Health and Social Behavior 19, 2-21. Perrone, K.M., Aegisdottir, S., Webb, L. K. & Blalock, R.H. (2006). Work-family interface – Commitment, Conflic t, Coping, and Satisfaction. Journal of Career Development, 32(3), 286-300. Perusse, L., Leblanc, C., & Bouchard, C. (1988). Familial resemblance in lifestyle components: results from the Canada Fitn ess Survey. Canadian Journal of Public Health, 79, 201-205. Perry-Jenkins, M., Repetti, R. L., & Crouter, A. C. (2000). Work and family in the 1990s. Journal of Marriage and the Family 62(4), 981-998. Perry-Smith, J.E., & Blum, T.C. (2000). Work-family HR bundles and perceived organizational performance. Academy of Management Journal 43, 6, 1107-1117. Perusse, L., Leblanc, C., & Bouchard, C. (1988). Familial resemblance in lifestyle components: results from the Canada Fitness Survey. Canadian Journal of Public Health 79, 201-205. Pomaki, G., Supeli, A., & Verhoeven, C. (2007) Role conflict and health behaviors Moderating effects on psychological di stess and somatic complaints. Psychology and Health 22(3), 317-335. Prochaska, J.J., Rodgers, M.W., Sallis, J.F. (2002). Association of parent and peer support with adolescent physical activity. Research Quarterly for Exercise and Sport 73(2), 206-210.

PAGE 142

127 Sallinen, M., Ronka, A., & Kinnunen, U. (2007) Trajectories of depressive mood in adolescents: Does parental work or parent-adolescent relationship matter? A follow-up study through junior high school in Finland. International Journal of Behavioral Development, 31(2), 181-190. Schmitt, N., Colligan, M.J., & Fitzgerald, M. (1980). Unexplained physical symptoms in eight organizationsIndividual and organizational analyses. Journal of Occupational Psychology 53, 4, 305-317 Shields, M. (1999). Long working hours and health Health Reports, 11(2), 33–48 (Statistics Canada, Catalogue 82-003). Shinn, M., Rosario, M., Morch, H., & Chestnut D.E. (1984). Coping with job stress and burnout in the human services. Journal of Personality and Social Psychology 46(4), 864-876. Shrout, P. E., & Bolger, N. (2002). Media tion in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods, 7 422445. Smolak, L., Levine, M. P., & Schermer, F. ( 1999). Parental input and weight concerns among elementary school children. International Journal of Eating Disorders 25, 263–271. Steinmetz, K.A., & Potter J.D. (1996). Vegeta ble, fruit, and cancer preventiona review. Journal of the American Dietetic Association 96(10), 1027-1039. Steffy, B.D., & Jones, J.W. (1988). The imp act of family and career planning variables on the organizational, car eer and community commitment of professional women. Journal of Vocational Behavior 32, 196-212.

PAGE 143

128 Stewart, W., & Barling, J. (1996). Fathers’ work experiences affect children’s behaviors via job-related affect a nd parenting behaviors. Journal of Organizational Behavior 17(3), 221-232. Striegel-Moore, R.H., Franko, D.L., Thomps on, D., Affenito, S., & Kraemer, H.C. (2006). Night eating Prevalence and demographic correlates. Obesity 14(1), 139. Study: Employer share of health care costs wi ll skyrocket (2003). Business Courier of Cincinnati, Friday, May 2, 2003. Retrieved October 19, 2007 from http://cincinnati.bizjournals.com/ci ncinnati/stories/ 2003/04/28/daily53.html Sylvestre, M., O’Loughlin, J., Gray-Donald, K., Hanley, J., & Pa radis, G. (2007). Association between fruit a nd vegetable consumption in mothers and children in low-income, urban neighborhoods. Health Education & Behavior ., 34(5), 723734. Tak, N.I., te Velde, S.J., de Vies, J.H.M., & Brug, J. (2006). Parent and child reports of fruit and vegetable intakes and related family environmental factors show low levels of agreement. Human Nutrition and Dietetics 19, 275-285. Thomas, L.T., & Ganster, D.C. (1995). Imp act of family-supportiv e work variables on work-family conflict and strain A control perspective. Journal of Applied Psychology 80, 1, 6-15. Thompson, O.M., Ballew, C., Resnicow, K., Must, A., Bandini, L.G., & Dietz, W.H. (2004). Food purchased away from home as a predictor of change in BMI z-score among girls. International Journal of Obesity-Related Metabolic Disorders, 28, 282-9

PAGE 144

129 Torgan, C. (2002). Childhood obesity on the rise Retrieved November 1, 2006, from The National Institutes of Health, The Word on Health Web site : (http://www.nih.gov/news/WordonHealth /jun2002/childhoodobesity.htm). Trend of the Month: Health Insurance Is Mo st Expensive Employer-Paid Benefit (2004). Drug Benefit Trends, 16(9),443. Retrieved October 19, 2007 from the worldwide-web on : http://www.medscape.com/viewarticle/490517 Trieman, K., Freimuth, V., Damron, D., Lasswell, A., Anliker, J., Havas, S., et al. (1996). Attitudes and behaviors related to fruits and vegetables among low-income women in the WIC program. Journal of Nutritional Education 28, 149-156. Troiano, R.P., & Flegal, K.M., (1998). Overwei ght children and adol escents: Description, epidemiology, and demographics. Pediatrics, 101(3),497-504 Trost, S.G., Sallis, J.F., Pate, R.R., Freeds on, P.S., Taylor, W.C., & Dowda, M. (2003). Evaluating a model of parental influence on youth physical activity. American Journal of Preventive Medicine 25(4), 277-282. U.S. Department of Health and Human Services (2000). Healthy People 2010. Washington, DC: U.S. GPO. Van Duyn, M. & Pivonka, E. (2000). Overview of the health benefits of fruit and vegetable consumption for the dietetics professional: Selected literature. Journal of the American Dietetic Association, 100, 1511-1521. Vauthier, J.M., Lluch, A., Lecomte, E., Artur, Y., & Herbeth, B. (1996). Family resemblance in energy and macronutrient intakes: the Stanislas Family Study. International Journal of Epidemiology, 25, 1030-1037.

PAGE 145

130 Voydanoff (2002). Linkages betwen the work-family interface and work, family, and individual outcomes an integrative model. Journal of Family Issues 23, 138164. Waterhouse, J., Edwards, B, & Reilly, T. ( 2005). A comparison of eating habits between retired or semi-retired aged subjects and younger subjects in full-time work. Biological Rhythm Research 36(3), 195-218. Westman, M., & Etzion, M. (2005) The Cro ssover of Work-Family Conflict From One Spouse to the Other. Journal of Applied Social Psychology, 35(9), 1936-1957. Young, E.M., Fors, S.W., & Hayes, D.M. (2004). Associations between perceived parent behaviors and middle school student fruit and vegetable consumption. Journal of Nutrition Education Behavior 36, 2-12. Zellner, D.A., Loaiza, S., Gonzalez, Z., Pita J., Morales, J., P ecora, D., & Wolf, A. (2006). Food selection changes under stress. Physiology & Behavior 87, 789793.

PAGE 146

131 Appendices

PAGE 147

132 Appendix A. Mother self-report ea ting behavior items (Work days) Directions: We want to know what you eat on th e days that you work at your job Think about what you eat on work days while you answer the next items. Fruit means things like; apples, oranges, banana raisins, strawberries, or glass of 100% fruit juice. 1) How often do you eat fruit … at breakfast time on work days ? at lunch time on work days ? as part of a snack on work days? at dinner time on work days ? Vegetables mean things like; salad, vegetable sou p, and fresh or cooked vegetables like carrots or broccoli. DO NOT count frenc h fries, onion rings, or fried okra. 2) How often do you eat vegetables … at breakfast time on work days ? at lunch time on work days ? as part of a snack on work days? at dinner time on work days ? Snack foods means things like chips, popcor n, granola bars and crackers. 3) How often do you eat snack food … at breakfast time on work days ? at lunch time on work days ? as part of a snack on work days? at dinner time on work days ?

PAGE 148

133 Appendix B. Mother self-report ea ting behavior items (Off days) Directions: Now, we want to know what you eat on the da ys that you don’t work at your job. For some people this might be weekend, for others it might be other days of the week. Think about what you usually eat on your days off each week while you answer the next items. Fruit means things like; apples, oranges, banana raisins, strawberries, or glass of 100% fruit juice. 1) How often do you eat fruit … at breakfast time on your days off ? at lunch time on your days off ? as part of a snack on your days off? at dinner time on your days off ? Vegetables mean things like; salad, vegetable sou p, and fresh or cooked vegetables like carrots or broccoli. DO NOT count frenc h fries, onion rings, or fried okra. 2) How often do you eat vegetables … at breakfast time on your days off ? at lunch time on your days off ? as part of a snack on your days off? at dinner time on your days off ? Snack foods means things like chips, popcor n, granola bars and crackers. 3) How often do you eat snack food … at breakfast time on your days off ? at lunch time on your days off ? as part of a snack on your days off? at dinner time on your days off ?

PAGE 149

134 Appendix C. Child self-report eati ng behavior items (School days) Directions. We want to know what you eat on the days that you go to school Think about what you eat on school days while you answer the next items! Fruit means things like; apples, oranges, bananas, raisins, strawberries, or glass of 100% fruit juice. 1) How often do you eat fruit… at breakfast time on school days ? at lunch time on school days ? as part of a snack on school days? at dinner time on school days ? Vegetables mean things like; salad, vegetable sou p, and fresh or cooked vegetables like carrots or broccoli. DO NOT count frenc h fries, onion rings, or fried okra. 2) How often do you eat vegetables… at breakfast time on school days ? at lunch time on school days ? as part of a snack on school days? at dinner time on school days ? Snack foods means things like chips, popcor n, granola bars and crackers. 3) How often do you eat snack food… at breakfast time on school days ? at lunch time on school days ? as part of a snack on school days? at dinner time on school days ?

PAGE 150

135 Appendix D. Child self-report eating behavior items (Weekends) Directions. OK, NOW, we want to know wh at you eat on the WEEKEND!!! Think about what you usually eat on SATURDAY or SUNDAY while you answer the next items. Fruit means things like; apples, oranges, banana s, raisins, strawberries, or glass of 100% fruit juice. 1) How often do you eat fruit… at breakfast time on the weekend? at lunch time on the weekend? as part of a snack on the weekend? at dinner time on the weekend? Vegetables mean things like; salad, vegetable sou p, and fresh vegetables like carrots or broccoli. DO NOT count french frie s, onion rings, or fried okra. 2) How often do you eat vegetables… at breakfast time on the weekend? at lunch time on the weekend? as part of a snack on the weekend? at dinner time on the weekend? Snack foods means things like chips, popcor n, granola bars and crackers. 3) How often do you eat snack food… at breakfast time on the weekend? at lunch time on the weekend? as part of a snack on the weekend? at dinner time on the weekend?

PAGE 151

136 Appendix E. Mother self-report work -interference-with-family (WIF) items 1) The demands of my work interfere with my home and family life. 2) The amount of time my job takes up ma kes it difficult to fulfill family responsibilities. 3) Things I want to do at home do not get done because of the demands my job puts on me. 4) My job causes strain that makes it di fficult to fulfill family duties. 5) Due to work-related duties, I have to make changes to plans for family activities. Note. Netemeyer et al., 1996

PAGE 152

137 Appendix F. Mother self-repor t Household Coping Strategies 1) Do you hire people to help with chores (for example, babysitters, cleaning help, yard help, etc.)? 2) The following questions are about ways you try to manage your work and non-work responsibilities. Do you hire people to help with chores (for example, babysitters, cleaning help, yard help, etc.)? 3) Do you coordinate your household schedule with family members or with your child? 4) Do you share your family duties (for exampl e, babysitting, carpool, cleaning and yard work) with a family member, friend or your child? 5) Do you set priorities about which work or family activities are the most important? 6) Do you spend less time on less important duties? (for exampl e, regular house cleaning, activities with fr iends you aren’t close to) 7) Do you openly discuss problems in assigni ng household chores with your family? 8) Do you try to plan, schedule, and organize your work and family activities better? 9) Do you decide which work or family activities are the most important and then schedule time for each? 10) Do you lower your expectations for some activities when you can’t get everything done? (for example, allowing your house to stay kind of messy, cooking easy meals like frozen dinners) Note. Source: Steffy & Jones (1988)

PAGE 153

138 Appendix G. Mother self-report items Child Feeding Questionnaire Directions: Please answer the following questions about you and your child. Re member there are no right or wrong answers! Please select the answ er that best reflect your day to day life. Restriction 1) I have to be sure that my child does no t eat too many sweets (candy, ice cream, cake, poptarts or donuts). 2) I have to be sure that my child does no t eat too many high fat foods (for example, fried food, cheese, cheeseburgers). 3) I have to be sure that my child does not eat too much of his or her favorite foods. 4) I intentionally hide or keep some foods out of my child’s reach. 5) I offer sweets (candy, ice cream, cake, pastries poptarts or donuts) to my child as a reward for good behavior. 6) I offer my child his or her favorite foods in exchange for good behavior. 7) If I did not guide or regulate my child’s eating, he or she would eat too many junk foods. 8) If I did not guide or regulate my child’s eati ng, he or she would eat too much of his or her favorite foods. Pressure 1) My child should always eat all of the food on his or her plate. 2) I have to be especially careful to make sure that my child eats enough. 3) If my child says “I’m not hungry”, I try to get him or her to eat anyway. 4) If I did not guide or control my child’s eat ing, he or she would eat too much less than he or she should. Monitoring 1) Do you keep track of the sweets that this child eats? (For example, candy, ice cream, cake, pies, poptarts or donuts) 2) Do you keep track of the snack food that this child eats? (For example, chips, crackers, granola bars) 3) Do you keep track of the high-fat foods that this child eats? (For example, fried food, cheese, cheeseburgers) Note. Source: Birch et al. (2001)

PAGE 154

139 Appendix H. Mother demographics Please indicate your age in years. Please circle the lette r that best describes your ethni city (circle all that apply). White/Caucasian Black/African-American Hispanic/Latino Native Amercian Asian-American Other _______ In what country were you born? In what country were your parents born? Please circle the letter that best describes your child’s et hnicity (circle all that apply). White/Caucasian Black/African-American Hispanic/Latino Native Amercian Asian-American Other _______ What is your current marital status? (circle one) Not married Not married but living with partner Married What is the highest level of education you have completed? In a typical week, how many times does your ch ild spaced an entire day or night at a household other than your own? How many children do you have living in your home? How amny family members are living in your home? What is your occupation? What is your annual household income? Your weigtht _____ pounds Your height _____ feet _____ inches Your child’s weigtht _____ pounds Your child’s height _____ feet _____ inches Has anyone in your household had any special dietary needs in the last month? Have you tried to mostly eat low carb or low fat foods in the last month?

PAGE 155

About the Author Ashley Anne Marguerite Gray Walvoord co mpleted her Psychology at Louisiana State University. She earned a Ma sters in Industrial-Organi zational Psychology at the University of South Florida. Ashley’s prof essional interests include work-life issues, teams, and performance feedback. Ashley has managed large-scale funded research projects addressing multimodal virtual comm unication (Army Research Lab), and workfamily health (USF Interdisciplinary Initiative on Sustainable Communities). She enjoys teaching and mentoring students. Ashley was awarded the 2007 Eve Levine Graduate Student teaching award, was an honored finali st for the University Provost’s award for exceptional teaching, and has advised two unde rgraduate honors theses. Ashley coauthored 17 peer-reviewed resear ch papers at professional c onferences, a chapter in the Oxford Handbook of Organizati onal Well-Being, and peer-revi ewed publications in the Journal of Computers in Huma n Behavior, Human Performance, and IEEE Transactions. On the lighter side, Ashley loves music a nd regularly lets off steam by jamming in her stats professors’ rock band (“The Outliers”).


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 2200433Ka 4500
controlfield tag 001 002221166
005 20100715163932.0
007 cr bnu|||uuuuu
008 100715s2009 flua sb 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002923
035
(OCoLC)648018840
040
FHM
c FHM
049
FHMM
090
BF121 (Online)
1 100
Walvoord, Ashley G.
0 245
Work-family conflict, eating behaviors, and the role of coping
h [electronic resource] /
by Ashley G. Walvoord.
260
[Tampa, Fla.] :
b University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains 139 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2009.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
520
ABSTRACT: There were two primary aims of the present study. The first aim was to examine the relationships between work-interference-with-family (WIF) and specific eating behaviors (eating vegetables, fruits, snack foods) reported by employed mothers, as it relates to health criteria such as BMI. Related to this first aim, household coping strategies were proposed as playing a significant role in the relationship between WIF and eating behaviors. The second aim was to investigate the crossover of WIF to specific child eating behaviors via mother feeding practices or mother eating behaviors. Self-report and other-report survey data were collected from working mothers and their children (recruited from the YMCA Afterschool Program in Hillsborough County), yielding a sample of 262 employed mothers and 238 mother-child dyads.Mother self-report results supported a negative relationship between WIF and mother eating vegetables on work days, but no relationships emerged for eating fruits or snack foods. Regarding the role of coping in the context of the WIF eating behavior relationship, results were more supportive of a suppression effect than of a moderating effect of coping. There was no support for an indirect relationship between WIF and BMI via eating behaviors. Analysis of the crossover hypotheses revealed support for a negative association between WIF and the mother's feeding practices (monitoring behaviors), but no evidence was found for the hypothesized meditational relationships between mother WIF and child eating behavior (via mother eating and mother feeding) using multisource data. However, the results of supplementary analyses using only mother-report data supported several of the meditational crossover relationships.The results have implications for theoretical development and future research in the growing area of work-family and health. Major findings regarding WIF and specific eating behaviors, coping, and mother vs. child report 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-interference with family
Eating
Fruits
Vegetables
Snacks
Household coping strategies
BMI
Feeding
Modeling
690
Dissertations, Academic
z USF
x Psychology
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2923