Content uploaded by Tammy D Allen
Author content
All content in this area was uploaded by Tammy D Allen
Content may be subject to copyright.
http://abs.sagepub.com/
American Behavioral Scientist
http://abs.sagepub.com/content/49/9/1204
The online version of this article can be found at:
DOI: 10.1177/0002764206286386
2006 49: 1204American Behavioral Scientist
Tammy D. Allen and Jeremy Armstrong
Health-Related Behaviors
Further Examination of the Link Between Work-Family Conflict and Physical Health : The Role of
Published by:
http://www.sagepublications.com
can be found at:American Behavioral ScientistAdditional services and information for
http://abs.sagepub.com/cgi/alertsEmail Alerts:
http://abs.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://abs.sagepub.com/content/49/9/1204.refs.htmlCitations:
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
10.1177/0002764206286386American Behavioral ScientistAllen, Armstrong / Health-Related Behaviors
Further Examination of the
Link Between Work-Family
Conflict and Physical Health
The Role of Health-Related Behaviors
Tammy D. Allen
Jeremy Armstrong
University of South Florida
Although research consistently finds a relationship between work-family conflict and
employee health, the role of health-related behaviors such as diet and exercise have been
overlooked. The present research examines the links between both directions of work-
family conflict—family interference with work (FIW) and work interference with family
(WIF)—with several health-related behaviors (physical activity, fatty food consumption,
and healthy food consumption) and with multiple indicators of physical health (overall
health, health disorders, and body mass). Based on a sample of 246 employed individu-
als, path analysis demonstrates that FIW was associated with less physical activity and
with eating more high fat foods. In addition, WIF was associated with eating fewer
healthy foods. Fatty food consumption related to body mass and overall health, whereas
physical activity related to overall health and health disorders. The findings represent an
initial step toward a better understanding of the process linking work-family conflict with
employee physical health.
Keywords: work-family conflict; health behavior; exercise; dietary habits
During the past several decades, a considerable body of research has been gener-
ated examining the relationship between work and family roles (Eby, Casper,
Lockwood, Bordeaux, & Brinley, 2005). Much of this research investigates the issue
of work-family conflict (Bellavia & Frone, 2005). Rooted in role theory (Kahn,
Wolfe, Quinn, Snoek, & Rosenthal, 1964), work-family conflict is a specific form of
interrole conflict in which pressures from the work (family) role are incompatible with
pressures from the family (work) role. That is, participation in one role is made more
difficult by virtue of participation in the other role (Greenhaus & Beutell, 1985).
Research demonstrates that family can interfere with work (FIW) and that work can
interfere with family (WIF; Frone, Russell, & Cooper, 1992; Gutek, Searle, & Klepa,
1991). Recent meta-analytic research supports the discriminant validity of the two
types of conflict (Mesmer-Magnus & Viswesvaran, 2005).
1204
American Behavioral Scientist
Volume 49 Number 9
Month 2006 1204-1221
© 2006 Sage Publications
10.1177/0002764206286386
http://abs.sagepub.com
hosted at
http://online.sagepub.com
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
Perhaps one of the most striking findings of work-family conflict research is the
extent that work-family conflict is related to a wide variety of individual health and
well-being outcomes. For example, work-family conflict has been linked with factors
such as less job satisfaction within the work domain and less marital satisfaction
within the family domain (for reviews, see Allen, Herst, Bruck, & Sutton, 2000;
Kossek & Ozeki, 1998). Work-family conflict has also been associated with depres-
sion and clinical mental disorders (e.g., Frone, 2000; Grzywacz & Bass, 2003; Ham-
mer, Cullen, Neal, Sinclair, & Shafiro, in press). Finally, work-family conflict has
been connected with physical health outcomes (e.g., Adams & Jex, 1999; Grandey &
Cropanzano, 1999; Grzywacz, 2000).
Of interest in the present study was the relationship between work-family conflict
and worker physical health. Although there is a convincing body of research demon-
strating that work-family conflict is associated with physical health (Allen et al.,
2000), there is very little understanding of how this process occurs. That is, the mecha-
nisms that help explain why work-family conflict is associated with detrimental physi-
cal health outcomes are not well understood. The purpose of the present study was to
examine several behavioral pathways thought to help explain the link between work-
family conflict and physical health. Specifically, we examined three health-related
behavioral factors that have received limited research attention in the work and family
literature—physical activity, fatty food consumption, and health food consumption—
as behaviors thought to mediate the relationship between work-family conflict and
physical health outcomes. Three indicators of health were investigated: overall
physical health, diagnosed health problems, and body mass.
Work-Family Conflict and Physical Health Outcomes
Studies examining the relationship between work-family conflict and health
include those that assess overall or global health (e.g., Frone, Russell, & Barnes, 1996;
Frone, Russell, & Cooper, 1997; Grandey & Cropanzano, 1999; Grywacz, 2000) and
those that use symptom checklists (e.g., Adams & Jex, 1999; Burke & Greenglass,
2001; Kinnunen, & Mauno, 1998; Netemeyer, Boles,& McMurrian, 1996; Thomas &
Ganster, 1995). This research consistently demonstrates that both WIF and FIW are
associated with overall assessments of physical health and with a greater number of
physical health symptoms.
Work-family conflict has also been associated with chronic health conditions. In a
longitudinal study, Frone et al. (1997) found that FIW, but not WIF, was associated
with hypertension across a 4-year time lag. Thomas and Ganster (1995) reported a
small but significant cross-sectional relationship between WIF and diastolic blood
pressure. Research that has not measured work-family conflict directly, but that has
examined the effect of multiple roles, also suggests that the stress of managing both
work and family responsibilities may take a physical toll. Goldstein, Shapiro, Chicz-
DeMet, and Guthrie (1999) found that there was a significantly greater decrease in
heart rate from daytime to evening among women without children than for those with
children. Brisson, Laflamme, and Moisan (1999) found the combination of a high
Allen, Armstrong / Health-Related Behaviors 1205
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
stress job and child-rearing responsibilities was associated with a greater incidence of
high blood pressure among white-collar women.
We are aware of only one study that examines the relationship between work-
family conflict and body mass. Based on data from the National Survey of Midlife
Development in the United States, Grzywacz (2000) found that WIF, but not FIW, was
associated with obesity. Given that obesity has been cited as a major epidemic with
multiple causal factors (Wadden, Brownell, & Foster, 2002), including diet and physi-
cal activity, we thought the relationship between work-family conflict and body mass
deserved further attention.
Work-Family Conflict and Health-Related Behavior
Work-family conflict may relate to physical health through changes in behaviors
that in turn influence health. The effect of combining work and family responsibilities
on health-related behavior has received limited attention in the research literature
(studies examining alcohol intake are an exception, cf. Frone et al., 1996; Frone, Rus-
sell, & Cooper, 1993; Grzywacz & Marks, 2001). However, there are several theoreti-
cal perspectives that suggest a link between work-family conflict and health behavior.
The first perspective is based on stress theory. Engaging in unhealthy behaviors such
as eating comfort foods and sedentary (in)activity are thought to bring pleasure and,
thus, reduce stress (Ng & Jeffery, 2003). Engaging in these behaviors serves as a mood
management tool. Consistent with this perspective, Ng and Jeffery (2003) found that
among a sample of working men and women, high levels of perceived stress were
associated with a higher fat diet and exercising less. Thus, the stress produced by
incompatible work and family responsibilities is one reason why work-family conflict
may undermine positive health behaviors. Another perspective is based on time avail-
ability. The time availability perspective suggests that time is a finite resource with
demands from work and family roles, as well as individual needs, in competition for
its use (Nomaguchi & Bianchi, 2004). In a similar manner, Small and Riley (1990)
noted that time and psychological and physical energy may be thought of as fixed
resources. Consequently, a portion of any of these resources allocated to one pursuit
such as work or family becomes less available for another demand such as individual
activities. Therefore, it seems likely that individuals who feel that their time and
energy resources are depleted by work and family demands will be less likely to take
the additional time needed to engage in physical activity and make sound dietary food
choices. Specific research supporting the proposed links between physical activity
and food choices with work-family conflict and with physical health is briefly
reviewed in the following sections.
Physical Activity
Regular physical activity is considered an important aspect of a healthy lifestyle
and a key to disease prevention and physical well-being (Daley & Parfitt, 1996;
Dubbert, 2002). Research consistently shows that individuals who exercise experi-
1206 American Behavioral Scientist
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
ence less physical symptoms and better weight control than those who do not exercise
(Blair, 1993; Ensel & Lin, 2004).
It is surprising that there has been little research examining the relationship
between work-family conflict and physical activity. In the one study to examine work-
family conflict directly, Grzywacz and Marks (2001) reported that less WIF was asso-
ciated with greater regular vigorous physical exercise, but there was no relationship
between vigorous physical exercise and FIW.
Other research examines the effect of multiple role responsibilities on exercise.
Nomaguchi and Bianchi (2004) suggested that exercise is one form of leisure activity
that may be readily squeezed out by individuals with both work and family role
responsibilities. In support of their hypotheses, the authors found that work and family
roles did curtail exercise. Specifically, married adults reported spending less time
exercising than did unmarried adults. No relationship between number of children and
exercising was detected, but having children younger than the age of 5 was negatively
related to exercising. In a longitudinal study of young women, W. Brown and Trost
(2003) found that women who reported getting married, having their first or another
baby, becoming a single parent, or beginning paid work were significantly more likely
to be classified as inactive than were those who did not report those life events. In a
qualitative study, Backett and Davison (1995) interviewed individuals married with
children with regard to health relevant behavior. Consistent with the time availability
perspective, one primary finding in the study was that participants frequently men-
tioned that there were so many demands on their time that they found it difficult to
keep fit. In a similar manner, P. Brown, Brown, Miller, and Hansen (2001) found that
mothers of children attending child care centers reported that having no time because
of other commitments and a lack of energy were the main reasons cited for not being
more physically active.
Food Choices
There is a well-established link between specific dietary practices and physical
health. For example, dietary fat has long been linked with cardiovascular disease and
obesity (Bray & Popkin, 1998; Van Horn & Kavey, 1997). In addition, eating fruits,
vegetables, and whole grains has positive health benefits (e.g., Gerster, 1991; Hu &
Willett, 2002; Van Duyn & Pivonka, 2000).
We are not aware of any research that directly links food choices with work-family
conflict. However, the stress and time availability perspectives provide a conceptual
basis for predicting such a relationship. Research shows that individuals eat in
response to emotional stress (Macht & Simons, 2000). Cartwright et al. (2003) found
that higher perceived stress was associated with eating more fatty foods and less fruit
and vegetable consumption among adolescents. Thus, stress may have a harmful
effect on eating patterns by steering food choices away from the healthy and toward
the unhealthy. As mentioned previously, Ng and Jeffery (2003) also found that per-
ceived stress was associated with greater fatty food consumption.
Allen, Armstrong / Health-Related Behaviors 1207
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
In addition, there is evidence that consuming healthy foods such as fruits, vegeta-
bles, and fiber may be thought of as inconvenient. For example, time and effort have
been identified as primary barriers to eating fruits and vegetables (Hagdrup, Simoes,
& Brownson, 1998; Trieman et al., 1996).
A few studies do suggest that work and family issues may play a part in individuals’
food choices. In a qualitative study, Devine, Connors, Sobal, and Bisogni (2003)
reported that interview participants made poor food choices(e.g., skipping meals, eat-
ing fast food, eating junk food) because of time crunches between work and home
responsibilities (“I don’t have time to eat healthy.”). Participants also reported using
food as an “escape” from work-related stress. Schultz, Chung, and Henderson (1989)
asked participants to indicate the time-management strategies that they used to bal-
ance job and personal or family life. Buying prepared foods was mentioned by 6.7% of
the participants and eating out more was reported by 18.3%.
In sum, there is theoretical and empirical evidence to suggest that health behaviors
may link work-family conflict with physical health. Based on theory and research
from the stress, health, and work-family literatures, we tested the model shown in Fig-
ure 1, which suggests that WIF and FIW relate to health behaviors, which in turn relate
to several indicators of overall physical health.
Method
Participants
Participants were 246 individuals employed in a variety of occupations. To be eligi-
ble for inclusion in the study, participants had to work a minimum of 20 hours a week
and either be married (or living with their partner) and/or haveat least one child living
1208 American Behavioral Scientist
Food choices
Physical
activity
WIF
FIW
Physical health
Figure 1
Model of Hypothesized Relationships
Note: WIF = work interference with family; FIW = family interference with work.
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
at home. These inclusion criteria were used to ensure that each participant had at least
a moderate degree of both work and family responsibilities (Greenhaus, Collins, &
Shaw, 2003).
Participants ranged in age from 19 to 78 with a mean age of 38.61 (SD = 9.76).
They were primarily female (78%) and Caucasian (93.4%). Most were married or liv-
ing with their partner (91.4%), and 75.6% reported at least one child living at home.
The median level of education was a 4-year college degree. A variety of industries
(e.g., education, consumer goods, health care, manufacturing) and job titles (e.g.,
senior information technology–training specialist, police officer, mail carrier,
research entomologist) were represented.
Procedure
The data were collected using a snowball sampling approach. Other work and fam-
ily researchers have used snowball sampling (e.g., Carlson, Kacmar, & Williams,
2000; Westman, Etzion, & Gortler, 2004). Several methods were used to solicit partic-
ipants. First, the authors contacted personal and professional colleagues and provided
them with a link to an online survey used for data collection. Contacted colleagues
were asked to forward the survey request and link to other contacts and colleagues.
Second, we identified Internet discussion forums for parents and posted the survey
link and invitation to participate in the study on these forums. Finally, we used an e-
mail distribution list for human resource professionals to further broaden our potential
sample.
Because of the nature of our data collection method, we were unable to calculate an
estimated response rate. However, in some aspects, our sample is more representative
than many used to study employed individuals in that participants worked in a variety
of occupations and organizations.
Measures
Work-family conflict. Two directions (WIF and FIW) of work-family conflict were
assessed using the measure developed by Carlson et al. (2000). Nine items were used
to measure WIF (e.g., “I am so emotionally drained when I get home from work that it
prevents me from contributing to my family.”) and nine items were used to measure
FIW (e.g., “The time I spend with my family often causes me to not spend time in
activities at work that could be helpful to my career.”). A 5-point Likert-type scale was
used with responses that ranged from 1 = strongly disagree to 5 = strongly agree.
Higher scores indicate higher levels of WIF and FIW. Carlson et al. reported internal
consistency reliabilities ranging from .78 to .87 for each of the subscales. In the pres-
ent study, the internal consistency estimate for FIW was .87 and for WIF was .88.
Food choices. Participants were presented with two food lists. They were asked to
think about their food habits during the past year or so and to report how often they ate
the foods listed during a typical week. Six different response options were used rang-
Allen, Armstrong / Health-Related Behaviors 1209
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
ing from 0 = never to 5 = 5 or more times a week. One list consisted of 15 high fat foods
based on the dietary fat screener developed by Caan,Coates, and Schaffer (1995; e.g.,
bacon or sausage, fried chicken, doughnuts, hamburgers). The second food list con-
sisted of 11 healthy food choices (e.g., green salad, fiber cereals, carrots, dark bread)
taken from the 2003 Behavioral Risk Factor Surveillance System State Questionnaire
used by the Centers for Disease Control and Prevention. A total score for each was
computed by summing across all items. Higher scores indicate more fat consumption
and more healthy food consumption, respectively.
Physical activity. Participants were asked to indicate if during the past month, other
than their regular job, they had participated in any physical activities or exercises such
as running, calisthenics, golf, gardening, or walking for exercise. No responses were
coded 1 and yes responses were coded 2.
Overall physical health. To assess overall physical health, participants were asked
to rate their overall physical health relative to others their own age on a 5-point Likert-
type scale that ranged from 1 = poor to 5 = excellent. This measure has been used in
previous work-family and health research (Frone et al., 1996). Frone et al. (1996)
reported research indicating that single-item measures of overall physical health have
been shown to be strong predictors of mortality, hospitalization, and physical health
assessments.
Health disorders. Participants were ask to indicate if they had ever been told by a
doctor, nurse, or other health professional that they had any of the following health
disorders: high cholesterol, high blood pressure, diabetes, asthma, or ulcers. Partici-
pants were instructed to exclude a diagnosis that occurred only during a time of preg-
nancy. Affirmative responses were summed to yield a total score.
Body mass. Participant reports of their height and weight were used to calculate a
body mass index (BMI). BMI was used as a continuous variable in the analyses. To
examine how representative our sample was of the general population, we compared
the percentage of our participants that could be classified as obese based on criteria
established by the Centers for Disease Control and Prevention (BMI of 30 or greater)
with those reported in the U.S. population. The U.S. Department of Health and Human
Services (2000) has estimated that 23% of the adult population can be classified as
obese, whereas 20.6% of our sample had a BMI of 30 or greater.
Potential control variables. Gender was coded 1 = male and 2 = female. Race was
coded 0 = nonminority and 1 = minority. Parental status was coded 1 = child(ren) living
at home and 2 = no children at home. Marital status was coded 1 = not married and 2 =
married or living with partner. Age was coded in years. Education level was an 8-point
ordinal variable that ranged from no education to graduate degree. Income was a 10-
point ordinal scale that ranged from less than US$10,000 to more than US$150,000.
Participants indicated how many hours on average they worked each week.
1210 American Behavioral Scientist
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
Results
Intercorrelations, means, and standard deviations among the study variables are
shown in Table 1. Path analysis was used to determine if the relationships among the
variables were consistent with the model specified in Figure 1. To preserve power, we
included only control variables that were significantly associated with the dependent
variables. As shown in Table 1, race, income, work hours, and marital status were not
associated with any of the dependent variables and, therefore, were not included in any
of the analyses. Gender was associated with physical disorders, age was associated
with physical disorders and overall health, and education was associated with overall
health. These variables were included as controls in the analyses involving their
respective dependent variable.
A series of ordinary least squares regressions were conducted to obtain standard-
ized beta weights for each path. Each endogenous variable was treated as the criterion
and the variables hypothesized to directly affect it were entered as predictors. We also
examined whether WIF and FIW exerted a direct effect on the health outcomes. Sepa-
rate sets of analyses were conducted for each of the three health outcome variables.
The regression results for BMI are shown in Table 2. The full model accounted for
9% of the variance associated with body mass. The model received mixed support in
that four of the nine hypothesized paths were significant. Greater FIW was associated
with less physical activity and with greater fatty food consumption. Greater WIF was
associated with less healthy food consumption. Of the health behaviors examined,
fatty food consumption was significantly associated with body mass. It should also be
noted that physical activity was associated with body mass at p= .05. No direct paths
between WIF or FIW with body mass were detected.
The results for health disorders are shown in Table 3. The full model accounted for
15% of the variance associated with physical health disorders. The results again indi-
cate that four of the nine hypothesized paths were significant. The same significant
results regarding WIF, FIW, and the health behaviors apply as discussed for body
mass. With regard to the health behaviors and physical disorders, less physical activity
was associated with a greater number of reported health disorders. WIF, but not FIW,
also demonstrated a direct effect with physical disorders, indicating that its relation-
ship with physical disorders is not completely mediated by food choices and physical
activity.
The results for overall health are presented in Table 4. The full model accounted for
18% of the variance associated with overall health. Five of the hypothesized paths
were significant. The same significant results regarding WIF, FIW, and the health
behaviors apply as discussed for body mass. With regard to health behaviors and over-
all health, both greater physical activity and less fatty food consumption were associ-
ated with better overall health. No direct paths between WIF or FIW with overall
health were found.
Allen, Armstrong / Health-Related Behaviors 1211
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
1212
Table 1
Intercorrelations, Means, and Standard Deviations
12345678910111213141516
1. WIF —
2. FIW .74** —
3. Physical activity –.10 –.16* —
4. Fatty foods .03 .14* —
5. Healthy foods –.21** –.13* .22** —
6. BMI .02 .07 –.18** .25** –.12 —
7. Overall health –.21** –.24** .24** –.25** .18** –.36** —
8. Health disorders .21** .13* –.16* –.03 .00 .22** –.24** —
9. Gender –.04 –.04 –.06 –.03 –.15* –.09 –.05 –.20** —
10. Race .01 –.09 –.05 –.15* –.06 –.02 –.02 .08 .02 —
11. Age .06 .06 –.05 –.15* .15* .11 .16* .28** –.24** .01 —
12. Children –.01 –.13* .07 –.19** .05 .01 .03 .05 –.07 –.07 .33** —
13. Marital status .02 .01 .09 –.05 –.05 .03 –.08 .10 –.06 –.10 .05 .18** —
14. Education –.14* –.08 .16* –.02 .16* –.14* .18** –.05 –.06 –.03 .12 .16* .15* —
15. Income –.11 .02 .16* –.15* .02 –.10 .11 .01 –.07 –.06 .18** .13* .38** .34** —
16. Work hours .23** .09 .05 –.18** .00 –.05 .05 .08 –.30** .02 .29** .28** –.01 .13* .29** —
Mean 2.74 2.45 1.71 45.08 36.85 26.01 3.39 .47 38.61 6.50 7.53 41.43
Standard deviation .77 .72 .46 8.82 7.71 6.18 .95 .73 9.76 1.36 1.84 9.49
Note: WIF = work interference with family; FIW = family interference with work; BMI = Body Mass Index.
*p< .05. **p< .01.
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
Discussion
A great deal of research documents that both WIF and FIW are associated with
physical health. However, little research attempts to identify the mechanisms that help
explain the link between work-family conflict and health. The present study is the first
to focus on the relationship between work-family conflict and health-related behav-
iors such as food choices and physical activity. The results reveal that one reason why
work-family conflict relates to poor physical health outcomes is through its relation-
ship with health behaviors such as diet and exercise.
We found that greater FIW related to less physical activity, but there was no rela-
tionship between physical activity and WIF. This is in contrast to Grzywacz and Marks
(2001) who reported that although less WIF was associated with greater regular vigor-
ous physical exercise, there was no relationship between vigorous physical exercise
and FIW. One explanation for the different findings may be that Grzywacz and Marks
focused on vigorous physical activity, whereas our measure encompassed different
levels of activity. The sample used by Grzywacz and Marks was also limited to indi-
viduals aged 35 to 65. Regardless, the results of our study, along with those of
Grzywacz and Marks, suggest that the relationship between work-family conflict and
physical activity deserves further attention. Research that examines participation in
different levels (e.g., mild, vigorous) and/or types of exercise (e.g., walking, jogging)
Allen, Armstrong / Health-Related Behaviors 1213
Table 2
Summary of Path Analysis Regression Equations for Body Mass Index
Equation Bp
Physical activity: R2= .03, p= .035
WIF .05 .602
FIW –.20 .036
Fatty food consumption: R2= .03, p= .016
WIF –.17 .071
FIW .27 .005
Health food consumption: R2= .05, p= .003
WIF –.26 .006
FIW .06 .507
BMI: R2= .09, p= .000
Physical activity –.13 .050
Fatty food consumption .23 .000
Health food consumption –.07 .281
BMI: R2= .09, p= .001
Physical activity –.12 .057
Fatty food consumption .22 .001
Health food consumption –.07 .262
WIF –.04 .683
FIW .03 .724
Note: WIF = work interference with family;FIW = family interference with work; BMI = Body Mass Index.
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
may help further elucidate the relationship between physical activity and work-family
conflict.
We found an interesting pattern of results with regard to diet choices and direction
of work-family conflict. Specifically, the path analysis indicated that WIF was associ-
ated with eating fewer healthy foods but not with eating more high fat foods. In con-
trast, FIW was associated with eating more high fat foods but not with eating fewer
healthy foods. Thus, the results suggest that WIF steers people away from healthy
food choices, whereas FIW steers people toward unhealthy food choices. There are
several tentative explanations for the results. Individuals experiencing WIF may make
food choices that are tied to the concept of time. The factors that contribute to the
occurrence of WIF include long work hours and a lack of schedule flexibility (Byron,
2005). As noted in the introduction, one of the primary reasons individuals cite for not
1214 American Behavioral Scientist
Table 3
Summary of Path Analysis Regression Equations
for Physical Disorders (N= 243)
Equation Bp
Physical activity: R2= .036, p= .065
Gender –.08 .200
Age –.07 .309
WIF .05 .588
FIW –.20 .035
Fatty food consumption: R2= .061, p= .005
Gender –.07 .312
Age –.17 .010
WIF –.16 .083
FIW .27 .004
Health food consumption: R2= .088, p= .000
Gender –.12 .062
Age .14 .034
WIF –.27 .004
FIW .06 .541
Physical disorders: R2= .12, p= .000
Gender –.16 .016
Age .23 .000
Physical activity –.15 .016
Fatty food consumption –.02 .723
Health food consumption –.03 .632
Physical disorders: R2= .15, p= .000
Gender –.15 .020
Age .22 .001
Physical activity –.15 .015
Fatty food consumption –.01 .846
Health food consumption .02 .808
WIF .26 .005
FIW –.10 .261
Note: WIF = work interference with family; FIW = family interference with work.
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
eating more healthy foods is that their preparation is perceived to be more time con-
suming and inconvenient (Hagdrup et al., 1998; Trieman et al., 1996). Thus, individu-
als experiencing WIF may not make the effort to include more fruits, vegetables, and
fiber in their diet based on the belief that it will be another drain on their time and
energy. On the other hand, FIW may elicit more of a stress-induced response to food
choices that prompts individuals to use fatty foods as a comfort mechanism. This
seems plausible if one considers that FIW has been more highly associated with anxi-
ety disorders and depression than has WIF (Frone, 2000). It is also possible that
because FIW is associated with greater family responsibilities (Bryon, in press), these
greater responsibilities involve more opportunity to indulge in high fat foods. For
example, children’s birthday parties and family gatherings are social functions that
typically involve food, often high in fat (e.g., cake, pizza). One psychological motiva-
tion for eating involves eating for social reasons (e.g., as a way to celebrate a special
Allen, Armstrong / Health-Related Behaviors 1215
Table 4
Summary of Path Analysis Regression Equations for Overall Health
Equation Bp
Physical activity: R2= .05, p= .009
Education .16 .013
Age –.07 .289
WIF .08 .394
FIW –.20 .030
Fatty food consumption: R2= .06, p= .008
Education .00 .983
Age –.15 .018
WIF –.16 .087
FIW .27 .004
Health food consumption: R2= .09, p= .000
Education .11 .071
Age .15 .018
WIF –.25 .008
FIW .05 .558
Overall health: R2= .14, p= .000
Education .12 .050
Age .11 .074
Physical activity .17 .006
Fatty food consumption –.20 .002
Health food consumption .09 .180
Overall health: R2= .18, p= .000
Education .11 .087
Age .13 .034
Physical activity .16 .012
Fatty food consumption –.18 .004
Health food consumption .06 .371
WIF –.08 .383
FIW –.12 .192
Note: WIF = work interference with family; FIW = family interference with work.
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
occasion with friends, family, or a loved one; Jackson, Cooper, Mintz, & Albino,
2003). In future research it may be helpful to examine the psychological motivations
that individuals report for eating (e.g., social vs. coping), dietary choices, and work-
family conflict to better understand the pattern of effects observed in the present study.
Contrary to Grzywacz (2000), we found no relationship between either direction of
work-family conflict and an individual’s body mass. The greater power to detect sig-
nificant effects based on the larger samplesize (N= 1,547) used in the Grzywacz study
may be one explanation. No correlation matrix was included in the Grzywacz study to
enable a comparison of the magnitude of the zero-order correlations. Regardless, the
correlations observed in the present study were very small in magnitude, suggesting
little practical significance even with greater power to detect statistical significance.
However, it may be fruitful to conduct additional investigations examining the rela-
tionship between obesity or body mass with work-family conflict. There may be a
stronger relationship between work-family conflict and overweight or obesity status
among high-risk populations such as women with lower incomes and African
American women (U.S. Department of Health and Human Services, 2000).
WIF had a direct effect on the number of health disorders reported. Thus, the
impact of work-family conflict on physical health is not completely transmitted
through physical activity and food choices. This is in contrast to overall health, for
which complete mediation was supported. In future research it may be beneficial to
systematically unpack different aspects of physical health in relation to work-family
conflict. Most research to date uses global measures of overall health or summed
scales of physical symptoms. These types of assessments have been informative in
terms of revealing that work-family conflict does relate to physical health, but they can
mask what specific aspects of physical health may be most highly associated with
work-family conflict.
The results of the present study have several theoretical and practical implications.
The model proposed in the present study moves theory a step closer to better elucidat-
ing the process whereby work-family conflict relates to physical health. Outside of
alcohol dependency, there has been very little investigation of health-related behav-
iors in association with work-family conflict. Moreover, the research examining alco-
hol use is limited to investigating it as a dependent variable and not as a behavioral
pathway linking work-family conflict and physical health (Frone, Barnes, & Farrell,
1994; Frone et al., 1996; Frone et al., 1993; Grzywacz & Marks, 2001). In concert,our
results and those regarding alcohol use suggest that individuals may engage in
unhealthy behaviors in response to work-family conflict, perhaps as a way to manage
time (e.g., not exercising, not preparing healthy foods) and/or as a way to manage
affect (e.g., drinking, eating fatty foods). These behaviors in turn can have detrimental
consequences on physical health. These findings also have implications for health
promotion researchers who have a desire to better understand the barriers that prevent
physical activity and healthy dietary choices.
From a practical perspective, the results suggest that organizations may benefit
from integrating health promotion efforts along with family-friendly initiatives. This
is important in that Duxbury and Higgins (2001) observed that employees reporting
1216 American Behavioral Scientist
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
high levels of WIF and FIW made more use of the health care system in Canada. For
example, participants with high FIW were more likely to have been in the hospital in
the past 6 months. Thus, efforts to simultaneously promote wellness and reduce work-
family conflict may pay off for employers in terms of reduced health care costs. There
are a number of ways in which this might be accomplished. Although not practical for
every employer, on-site physical fitness facilities may make it easier for those with
family responsibilities to engage in exercise. Classes held at a convenient time geared
toward parents may promote employee fitness while also providing the basis for a
work and family balance support group. Time-management courses that also incorpo-
rate tips for preparing quick, easy, and healthy family meals may help reduce work-
family conflict and raise healthy food consumption. Ensuring that employee cafete-
rias and vending machines include healthy food selections may provide the conve-
nience needed to help time-challenged workers make sound dietary decisions. In
short, taking a whole-life employee perspective may improve employee health and in
turn save employer health care costs.
Several limitations to the present study warrant consideration. Because of the
snowball sampling approach we used, it is uncertain how representative our sample is
to any particular population. Moreover, our sample was overrepresented by women.
Additional research is needed to determine the generalizability of our results. Cause
and effect cannot be disentangled because of the cross-sectional nature of our study.
For example, although we theorized that individuals experiencing work-family con-
flict would be less likely to exercise, it is also plausible that those who exercise report
less work-family conflict. However, the longitudinal research reported by W. Brown
and Trost (2003) suggests that individuals decrease their physical activity in response
to life stage transitions. More research that examines exercise patterns across time, as
work and family demands change, would help address this issue. Finally, it is recog-
nized that the magnitude of the effects observed in the present study are small. All the
same, the findings clearly point to links between work-family conflict, health-related
behaviors, and physical health that warrant further research attention.
Future Research
Future research is needed that incorporates additional pathways that help explain
the link between work-family conflict and health outcomes. There is a considerable
body of research that links stress and poor physical health through neuroendocrine
responses (Halpern, 2003) with some research showing that these effects vary as a
function of family structure. For example, Lundberg and Frankenhaeuser (1999)
found that women with children at home had higher norepinephrine levels after work
than did both women without children and all other males. These findings were attrib-
uted to the higher overall workloads (both work at home and through paid employ-
ment) of women with children. We are not aware of any research that directly exam-
ines reports of work-family conflict and neuroendocrine responses. Such research
would likely help further explain the processes linking work-family conflict and
physical health.
Allen, Armstrong / Health-Related Behaviors 1217
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
Previous research shows that father’s WIF and job demands indirectly relate to
children’s acting out behavior (Stewart & Barling, 1996). Other research shows that
parental negative spillover from work to home is directly related to adolescent prob-
lems and grades (Voydanoff, 2004). Another promising line of research may be to
examine the effects of parental work-family conflict on children’s health behaviors.
This type of research seems particularly important given the growing rate of childhood
obesity and declining physical activity (U.S. Department of Health and Human Ser-
vices, 2000). If individuals experiencing work-family conflict find it difficult to make
healthy eating choices for themselves, these dietary practices may be passed along to
other family members.
Conclusion
Research continues to support the notion that work-family conflict relates to the
health and well-being of employed individuals. By demonstrating that work-family
conflict relates to behavioral health factors, the results of the present study provide
even more reason for being concerned about the effects of work-family conflict.
Understanding how behavioral factors play a role in the work-family conflict–health
link is a promising avenue for future research.
References
Adams, G., & Jex, S. (1999). Relationships between time management, control, work-family conflict, and
strain. Journal of Occupational Health Psychology,4, 72-77.
Allen, T. D., Herst, D. E. L., Bruck, C. S., & Sutton, M. (2000). Consequences associated with work-to-
family conflict: A reviewand agenda for future research. Journal of Occupational Health Psychology,5,
278-308.
Backett, K. C., & Davison, C. (1995). Lifecourse and lifestyle: The social and cultural location of health
behaviours. Social Science Medicine,40, 629-638.
Bellavia, G., & Frone, M. R. (2005). Work-family conflict. In J. Barling, E. K. Kelloway, & M. R. Frone
(Eds.), Handbook of work stress (pp. 113-147). Thousand Oaks, CA: Sage.
Blair, S. N. (1993). Evidence for the success of exercise in weight loss and control. Annals of Internal Medi-
cine,119, 702.
Bray, G. A., & Popkin, B. M. (1998). Dietary fat intake does affect obesity! American Journal of Clinical
Nutrition,68, 1157-1173.
Brisson, C., Laflamme, N., & Moisan, J. (1999). Effect of family responsibilities and job strain on ambula-
tory blood pressure among white-collar women. Psychosomatic Medicine,61, 205-213.
Brown, P. R., Brown, W. J., Miller, Y. D., & Hansen, V. (2001). Perceived constraints and social support for
active leisure among mothers with young children. Leisure Sciences,23, 131-144.
Brown, W. J., & Trost, S. G. (2003). Life transitions and changing physical activity patterns in young
women. American Journal of Preventative Medicine,25, 140-143.
Burke, R. J., & Greenglass, E. R. (2001). Hospital restructuring stressors, work-family concerns and psy-
chological well-being among nursing staff. Community, Work & Family,4, 49-62.
Byron, K. (2005). A meta-analytic review of work-family conflict and its antecedents. Journal of Vocational
Behavior, 67, 169-198.
1218 American Behavioral Scientist
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
Caan, B., Coates, A., & Schaffer, D. (1995). Variations in sensitivity, specificity, and predictive value of a
dietary fat screener modified from Block et al. Journal of the American Dietetic Association,95, 564-
568.
Carlson, D. S., Kacmar, K. M., & Williams,L. J. (2000). Construction and initial validation of a multidimen-
sional measure of work-family conflict. Journal of Vocational Behavior,56, 249-276.
Cartwright, M., Wardle, J., Steggles, N., Simon, A. E., Croker, H., & Jarvis, M. J. (2003). Stress and dietary
practices in adolescents. Health Psychology,22, 362-369.
Daley, A., & Parfitt, G. (1996). Good health—Is it worth it? Mood states, physical well-being, job satisfac-
tion and absenteeism in members and non-members of a British health and fitness club. Journalof Occu-
pational and Organizational Psychology,69, 121-134.
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 low- and moderate-income urban households. Social Science & Medi-
cine,56, 617-630.
Dubbert, P. M. (2002). Physical activity and exercise: Recent advances and current challenges. Journal of
Consulting and Clinical Psychology,70, 526-536.
Duxbury, L., & Higgins, C. (2001). Work-life balance in the new millennium: Where are we? Where do we
need to go? (Canadian Policy Research Network Paper No. WI12). Ottawa, ON: Canadian Policy
Research Network.
Eby, L. T., Casper, W. J., Lockwood, A., Bordeaux, C., & Brinley, A. (2005). Work and family research in
IO/OB: Content analysis and review of the literature. Journal of Vocational Behavior,66, 124-197.
Ensel, W. M., & Lin, N. (2004). Physical fitness and the stress process. Journal of Community Psychology,
32, 81-101.
Frone, M. R. (2000). Work-family conflict and employee psychiatric disorders: The national comorbidity
survey. Journal of Applied Psychology,85(6), 888-895.
Frone, M. R., Barnes, G. M., & Farrell, M. P. (1994). Relationship of work-family conflict to substance use
among employed mothers: The role of negative affect. Journal of Marriage and Family,56, 1019-1030.
Frone, M. R., Russell, M., & Barnes, G. M. (1996). Work-family conflict, gender, and health-related out-
comes: A study of employed parents in two community samples. Journal of Occupational Health Psy-
chology,1, 57-69.
Frone, M. R., Russell, M., & Cooper, M. L. (1992). Antecedents and outcomes of work-family conflict:
Testing a model of the work-family interface. Journal of Applied Psychology,77, 65-78.
Frone, M. R., Russell, M., & Cooper, M. L. (1993). Relationship of work-family conflict, gender, and alco-
hol expectancies 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 longitudinal study of employed parents. Journal of Occupational and Organizational Psy-
chology,70, 325-335.
Gerster, H. (1991). Potential role of beta-carotene in the prevention of cardiovascular disease. International
Journal of Clinical Nutrition,48,253-265.
Goldstein, I. B., Shapiro, D., Chicz-DeMet, A., & Guthrie, D. (1999). Ambulatoryblood pressure, heart rate,
and neuroendocrine responses in women nurses during work and off work days. Psychosomatic Medi-
cine,61, 387-296.
Grandey, A. A., & Cropanzano, R. (1999). The conservation of resources model applied to work-family con-
flict and strain. Journal of Vocational Behavior,54, 350-370.
Greenhaus, J. H., & Beutell, N. J. (1985). Sources of conflict between work and family roles. Academy of
Management Review,10, 76-88.
Greenhaus, J. H., Collins, K. M., & Shaw, J. D. (2003). The relation between work-family balance and qual-
ity of life. Journal of Vocational Behavior,63, 510-531.
Grzywacz, J. G. (2000). Work-family spillover and health during midlife: Is managing conflict everything?
American Journal of Health Promotion,14, 236-243.
Grzywacz, J. G., & Bass, B. L. (2003). Work, family, and mental health: Testing different models of work-
family fit. Journal of Marriage and Family,65, 248-262.
Grzywacz, J. G., & Marks, N. F. (2001). Social inequalities and exercise during adulthood: Toward an eco-
logical perspective. Journal of Health and Social Behavior,42, 202-220.
Allen, Armstrong / Health-Related Behaviors 1219
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
Gutek, B., Searle, S., & Klepa, L. (1991). Rational versus gender role explanations for work-family conflict.
Journal of Applied Psychology,76, 560-568.
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, 90-100.
Halpern, D. F. (2003, June). Psychosocial factors that link workplace/workforcemismatch and health. Paper
presented at the Workplace/Workforce Mismatch: Work, Family, Health, and Well Being Conference
sponsored by the National Institute of Child Health and Human Development, Washington, DC.
Hammer, L. B., Cullen, J. C., Neal, M. B., Sinclair, R. R., & Shafiro, M. (in press). The longitudinal effects
of work-family conflict and positive spillover on experiences of depressive symptoms among dual-
earner couples. Journal of Occupational Health Psychology.
Hu, F. B., & Willett, W. C. (2002). Optimal diets for prevention of coronary heart disease. Journal of the
American Medical Association,288, 2569-2578.
Jackson, B., Cooper, M. L., Mintz, L., & Albino, A. (2003). Motivations to eat: Scale development and vali-
dation. Journal of Research in Personality,37, 297-318.
Kahn, R. L., Wolfe, D. M., Quinn, R. P., Snoek, J. D., & Rosenthal, R. A. (1964). Organizational stress:
Studies in role conflict and ambiguity. New York: John Wiley.
Kinnunen, U., & Mauno, S. (1998). Antecedents and outcomes of work-family conflict among employed
women and men in Finland. Human Relations,51, 157-177.
Kossek, E. E., & Ozeki, C. (1998). Work-family conflict, policies, and the job-life satisfaction relationship:
A review and directions for organizational behavior–human resources research. Journal of Applied Psy-
chology,83, 139-149.
Lundberg, U., & Frankenhaeuser, M. (1999).Stress and workload of men and women in high-ranking posi-
tions. Journal of Occupational Health Psychology,4, 142-151.
Macht, M., & Simons, G. (2000). Emotions and eating in everyday life. Appetite,35, 65-71.
Mesmer-Magnus, J. R., & Viswesvaran, C. (2005). Convergence between measures of work-to-family and
family-to-work conflict: 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 con-
flict and family-work conflict scales. Journal of Applied Psychology,81, 400-410.
Ng, D. M., & Jeffery, R. W. (2003). Relationships between perceivedstress and health behaviors in a sample
of working adults. Health Psychology,22(6), 638-642.
Nomaguchi, K. M., & Bianchi, S. M. (2004). Exercise time: Gender differences in the effects of marriage,
parenthood, and employment. Journal of Marriage and Family,66, 413-430.
Schultz, J. B., Chung, Y. L., & Henderson, C. G. (1989). Work/family concernsof university faculty. In E. B.
Goldsmith (Ed.), Workand family: Theory, research,and application (pp. 249-264).Newbury Park, CA:
Sage.
Small, S. A., & Riley, D. (1990). Toward a multidimensional assessment of work spillover into family life.
Journal of Marriage and Family,52, 51-61.
Stewart, W., & Barling, J. (1996). Fathers’ work experiences effect children’s behaviors via job-related
affect and parenting behaviors. Journal of Organizational Behavior,17, 221-232.
Thomas, L. T., & Ganster, D. C. (1995). Impact of family-supportive work variables on work-family conflict
and strain: A control perspective. Journal of Applied Psychology,80,6-15.
Treiman, 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.
U.S. Department of Health and Human Services. (2000). Healthy people 2010. Washington, DC: Govern-
ment Printing Office.
Van Duyn, M., & Pivonka, E. (2000). Overview of the health benefitsof fruit and vegetable consumption for
the dietetics professional: Selected literature. Journal of the American Dietetic Association,100, 1511-
1521.
Van Horn, L., & Kavey, R. E. (1997). Diet and cardiovascular disease prevention: What works? Annals of
Behavioral Medicine,19, 197-212.
Voydanoff, P. (2004). Work, community, and parenting resources and demands as predictors of adolescent
problems and grades. Journal of Adolescent Research,19, 155-173.
1220 American Behavioral Scientist
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
Wadden, T. A., Brownell, K. D., & Foster, G. D. (2002).Obesity: Responding to the global epidemic. Jour-
nal of Consulting and Clinical Psychology,70, 510-525.
Westman, M., Etzion, D., & Gortler, E. (2004). The work-family interface and burnout. International Jour-
nal of Stress Management,11, 413-428.
TammyD. Allen is associate professor of psychology at the University of South Florida. Her research inter-
ests include work and family issues and occupational health psychology with an emphasis on how work-
family conflict relates to employee health and well-being. Her other research interests include mentoring
relationships, career development, and organizational citizenship behavior.
Jeremy Armstrong has a BA in psychology from the University of South Florida. His research interests are
in the areas of work and family issues and organizational workflow. His future plans are to obtain a graduate
degree in industrial and organizational psychology.
Allen, Armstrong / Health-Related Behaviors 1221
at UNIV OF SOUTH FLORIDA on June 20, 2011abs.sagepub.comDownloaded from
