Content uploaded by Dawn S Carlson
Author content
All content in this area was uploaded by Dawn S Carlson on Apr 07, 2017
Content may be subject to copyright.
The Moderating Effect of Work-Linked Couple Relationships and
Work–Family Integration on the Spouse Instrumental
Support-Emotional Exhaustion Relationship
Jonathon R. B. Halbesleben
University of Alabama
Kelly L. Zellars
University of North Carolina, Charlotte
Dawn S. Carlson
Baylor University
Pamela L. Perrewe´
Florida State University
Denise Rotondo
Meredith College
Based on boundary theory and conservation of resources theory, we argue that employees with
a spouse in the same occupation or workplace (referred to as work-linked couples) enhance the
effectiveness of spousal instrumental support in reducing emotional exhaustion. In the first study
of nurses (n⫽103), we found that work-linked couples have more highly integrated work and
family roles than participants whose spouses did not share an occupation or workplace. In a
second study of working adults (n⫽484), we found that work-linked couples have a stronger
relationship between spouse instrumental support and the reduction of emotional exhaustion; this
moderated relationship was mediated by their work–family integration. We discuss the implica-
tions of these findings for the study of work and family roles, social support, and emotional
exhaustion, and provide suggestions for future research.
Keywords: work–family integration, social support, emotional exhaustion, spouse support
The seminal research by Hall and Hall (1979)
examined the “two-career couple” and how to man-
age life when the woman enters the workforce. In the
three decades since their work, research examining
the links between work and family among dual-
career employees has flourished (e.g., Carlson &
Grzywacz, 2008; Edwards & Rothbard, 2000; Green-
haus & Powell, 2006; Halbesleben & Wheeler, 2007;
Janning, 2006; Kossek, Lautsch & Eaton, 2005; Voy-
danoff, 2008), especially in the stress and burnout
literature (cf., Elloy & Mackie, 2003). Like all work-
ers, as demands on both spouses in dual-career cou-
ples soar, high levels of strain emerge. Social support
plays a critical role in alleviating job strains like
emotional exhaustion (Carlson & Perrewe´, 1999;
Halbesleben, 2006). Spouse support is particularly
important for dual-career employees because it has
been associated with outcomes, such as well being,
family satisfaction, and less work–family conflict
(e.g., Aryee, Luk, Leung, & Lo, 1999; Aycan &
Eskin, 2005; Parasuraman, Greenhaus, & Granrose,
1992; Saleh et al., 2007).
As the number of dual-career couples has in-
creased (Gilbert, 1993), researchers have given
greater attention to the influence of the nature of
work done by one’s spouse (Janning, 2006). Interest
has developed in a subset of dual-career employees
that work in the same occupation or the same orga-
nization as their spouse (cf., Marshack, 1994; Moen
& Sweet, 2002; Wood & Duck, 1995) and how their
experiences differ from those who do not share a
workplace or occupation with their spouse. In the
literature, employees in this type of relationship have
been referred to as married coworkers (Janning,
2006), copreneurs (Marshack, 1994), coworking cou-
Jonathon R. B. Halbesleben, Department of Management
and Marketing, Culverhouse College of Commerce and
Business Administration, University of Alabama; Kelly L.
Zellars, Department of Management, University of North
Carolina, Charlotte; Dawn S. Carlson, Department of Man-
agement and Entrepreneurship, Hankamer School of Busi-
ness, Baylor University; Pamela L. Perrewe´ , Department of
Management, Florida State University; and Denise Rotondo,
School of Business, Meredith College.
An earlier version of this paper was presented at the 2005
meeting of the Southern Management Association in
Charleston, South Carolina.
Correspondence concerning this article should be addressed
to Jonathon R. B. Halbesleben, Department of Management
and Marketing, University of Alabama, Box 8870225, Tusca-
loosa, AL 35487-0225. E-mail: jhalbesleben@cba.ua.edu
Journal of Occupational Health Psychology
2010, Vol. 15, No. 4, 371–387
© 2010 American Psychological Association
1076-8998/10/$12.00 DOI: 10.1037/a0020521
371
ples (Moen & Sweet, 2002), and same-career couples
(Halbesleben & Wheeler, 2007; Halbesleben & Zel-
lars, 2006). With each of these labels, the authors
were referring to sharing both workplaces and occu-
pations; however, these terms do not seem to capture
both situations (e.g., the terms married coworkers and
coworking couples put the focus on sharing a work-
place). As a result, we refer to those in this situation
as work-linked couples. The idea of work linkages
with families is consistent with the work on work-
family linkages discussed by Edwards and Rothbard
(2000) as it focuses attention on how work serves as
a link between family and work roles for couples.
Work-linked couples can be linked by their work in
one of three ways: sharing only an occupation (e.g.,
both are nurses but at different hospitals), sharing
only a workplace (e.g., both work at the same hos-
pital, but one as a nurse and one as a sonographer), or
sharing both an occupation and a workplace (e.g.,
both are nurses at the same hospital).
While the literature concerning work-linked cou-
ples is relatively new, some interesting patterns have
emerged. Taking a sociological perspective, Janning
(1999, 2006; Janning & Neely, 2006) has examined
the work–family integration of work-linked couples
and the extent to which the work role and family role
are indistinguishable. It is suggested that work-linked
couples have more integrated roles and are in a better
position to be supportive of each other’s work en-
deavors. Her qualitative study (Janning & Neely,
2006) of 26 married couples found that spouses who
worked in the same workplace had more conversa-
tions about work while at home. In a similar vein,
Moen and Sweet (2002) found high levels of spill-
over (both positive and negative) between work and
family roles in couples who worked for the same
company; they interpret this finding as meaning that
the boundary between work and family is highly
integrated for work-linked couples. It is interesting to
note that while the different forms of work linkages
are conceptually different (e.g., Janning, 2006, has
noted that sharing a workplace means sharing phys-
ical space whereas sharing an occupation means
sharing mental space), the findings of these studies
tend to suggest that outcomes for these couples do
not differ if spouses share a workplace versus sharing
an occupation; however, sharing both does seem to
have added benefit (Janning, 2006).
The purpose of this research is to extend the pre-
vious research about work-linked couples by exam-
ining the impact that working in the same workplace
or occupation as one’s spouse has on an important
resource for dealing with stress (i.e., spouse instru-
mental support) and emotional exhaustion. In exam-
ining these relationships, this paper makes two con-
tributions to the literature on the work–family
interface. First, by examining spouse workplace and
occupation, we add to the limited knowledge of an
understudied population in the research (work-linked
couples; Janning, 2006). As more organizations
adopt recruitment and selection strategies supportive
of work-linked couples (Jaksic, 2008; Schiebinger,
Henderson, & Gilmartin, 2008; Wolkenbreit, 1997),
research on this group has important implications for
managers; studies of work-linked couples’ support
and strain represent a first step in that direction.
Second, our study holds implications for the inte-
gration of boundary theory (Ashforth, Kreiner, &
Fugate, 2000) and conservation of resources theory
(Hobfoll, 1988, 2001) by examining the relationship
between work-linked couples’ resources (i.e., spouse
instrumental support) and work-related strain (i.e.,
emotional exhaustion). From a boundary theory per-
spective, work-linked couples offer an opportunity to
test the theory among individuals with unique bound-
aries between their work and family roles. The
boundary theory literature has thus far supported the
positive aspects of work–family integration (e.g.,
Kreiner, 2006). By integrating boundary theory with
conservation of resources theory, we advance the
literature on the work–family boundary, its impact on
employee resources, and the utilization of resources
across boundaries to improve work-related well-
being (in the form of reduced emotional exhaustion).
To that end, we develop a model for the role of
spouse workplace, spouse occupation, and work-
family integration in the relationship between spouse
instrumental support and emotional exhaustion (see
Figure 1).
H4
Spouse
Instrumental
Work−Family
Integration
Spouse
Workplace/Occupation
(Work−Family Link)
Emotional
Exhaustion
H3
H2
H1
Figure 1. Conceptual model linking spouse workplace,
spouse occupation, work–family integration, spouse instru-
mental support, and emotional exhaustion.
372 HALBESLEBEN, ZELLARS, CARLSON, PERREWE
´, AND ROTONDO
Theoretical Foundations
Work and Family Roles Among
Work-Linked Couples: Applying
Boundary Theory
Boundary theory (Ashforth et al., 2000) helps ex-
plain the unique work–family role boundaries of
work-linked couples by focusing on the manner in
which employees manage transitions between life
roles. As we attempt to bring order to our roles, we
create psychological boundaries, or mental fences
(Nippert-Eng, 1996a; Zerubavel, 1991), to better
manage the expectations and obligations of a role.
The commonly studied roles of work and home (or
family) are defined by the boundaries surrounding
those roles (Nippert-Eng, 1996b). For example, many
employees have different physical spaces defined by
work (e.g., their office) and home in order to maintain
a boundary between these roles. Further, many em-
ployees use time to differentiate their work and fam-
ily roles (e.g., working on specific days of the week
and at specific times of the day).
The nature of the boundaries surrounding roles
differs depending on the person in the role (Nippert-
Eng, 1996a, 1996b); for example, individuals vary in
the “thickness” of their boundaries or the extent to
which one role is allowed to impact another role
(Hartmann, 1997). This suggests the existence of a
continuum of role boundary differentiation, from
segmented roles to integrated roles (Ashforth et al.,
2000). Boundary theory suggests that two compo-
nents of boundaries, flexibility and permeability, de-
termine the degree of segmentation versus integra-
tion. The first, flexibility, refers to one role that can
be enacted in another role (Voydanoff, 2008). In
other words, if one’s work role is flexible, it may not
be unusual to enact another role (e.g., family) simul-
taneously or to move between roles repeatedly during
a day in order to meet the demands of one role while
remaining in another role (Bulger, Matthews, &
Hoffman, 2007). Work-linked couples may experi-
ence greater flexibility in roles, in large part because
the people who appear in one role (e.g., work) also
appear, either physically or psychologically, in an-
other role (e.g., family; Halbesleben & Wheeler,
2007; Janning, 2006).
Permeability of the boundary refers to the extent to
which one can be “physically located in the role’s
domain but psychologically and/or behaviorally in-
volved in another role” (Ashforth et al., 2000, p. 474;
see also Voydanoff, 2008). Nippert-Eng (1996a)
noted that the more similar roles are, the more per-
meable the boundary between them. Thus, work-
linked couples’ work–family boundaries should be
more permeable because their roles are more similar
and because, as Janning (1999) documented, work-
linked couples have more symbolic and relational
representations of a permeable boundary between
work and family.
Situations of high flexibility and permeability of
boundaries represent a situation of highly integrated
roles (Ashforth et al., 2000). Taken together, the
above arguments suggest that work-linked couples
should experience higher levels of work–family role
integration than couples without a work link. Thus,
we expect that sharing a workplace and/or occupation
contributes to the integration of the work–family
boundary. Based on the findings of Janning (2006),
we do not expect differences based on whether one
shares a workplace or an occupation; however, we do
expect those who share both a workplace and occu-
pation to have higher integration because of even
greater flexibility and permeability of roles.
Hypothesis 1a: Employees who share a work-
place (but not an occupation) with their spouse
will experience greater work–family role inte-
gration than employees who do not share a
workplace with their spouse.
Hypothesis 1b: Employees who share an occu-
pation (but not a workplace) with their spouse
will experience greater work–family role inte-
gration than employees who do not share an
occupation with their spouse.
Hypothesis 1c: Employees who share both oc-
cupation and a workplace with their spouse will
experience greater work–family role integration
than employees who do not share an occupation
with their spouse or share only a workplace or
an occupation.
Moving Resources Across Role Boundaries
According to boundary theory, when roles are
highly integrated, it becomes easier to cross role
boundaries. While this has often been framed nega-
tively (e.g., the abundant research on work–family
conflict; Ahrentzen, 1990; Desrochers, Hilton, &
Larwood, 2005), integration of roles could also lead
to positive outcomes such as work–family enrich-
ment (Greenhaus & Powell, 2006; Marks, 1977;
Sieber, 1974). Greenhaus and Powell (2006) define
work-family enrichment as “the extent to which ex-
373WORK-LINKED COUPLES AND WORK–FAMILY INTEGRATION
periences in one role improve the quality of life in the
other role” (p. 73). The notion that family based
resources, particularly support from family, could
move across the work–family boundary with positive
work-related outcomes has been widely supported in
the empirical literature (cf., Adams, King, & King,
1996; Frone, Yardley, & Markel, 1997; Voydanoff,
2001).
According to boundary theory, this utilization or
reinvestment in the other role may be difficult if the
transition from roles is challenging (which is the case
when roles are segmented; Greenhaus & Powell,
2006). Alternatively, individuals with highly inte-
grated roles (i.e., work-linked couples) may transfer
resources more easily across role boundaries (Ash-
forth et al., 2000; Halbesleben & Wheeler, 2007).
Conservation of Resources Theory
While boundary theory helps explain the processes
underlying role boundaries and the transfer of re-
sources across boundaries, it does not necessarily
explain which resources should move across bound-
aries or why the movement of resources across
boundaries is important. For that foundation, we turn
to Conservation of Resources (COR) theory (Hob-
foll, 1988, 1989). The central tenant of COR theory is
that people are motivated to obtain and maintain
resources that help them attain goals. Therefore, as
individuals acquire resources, they are able to use
those resources to enrich their lives. Accordingly,
COR theory suggests that the threat of or actual loss
of resources causes people to engage in efforts to
avoid further loss (Hobfoll, 2001). A critical resource
that individuals develop and acquire to ameliorate the
depletion of other resources is social support (Hal-
besleben, 2006; Leiter, 1990). Social support in-
volves the exchange of resources between individuals
with the aim of helping the person receiving the
support (van Daalen, Willlemsen, & Sanders, 2006;
Winnubst, 1993). In this study, we focus on instru-
mental support, the availability of tangible resources
via other people (Beehr & Glazer, 2001).
Researchers have long realized the importance so-
cial support plays in reducing strain (Viswesvaran,
Sanchez, & Fisher, 1999). Consistent with COR the-
ory, supportive elements at work, as well as family,
add to the pool of personal resources that ultimately
aid the process of stress resistance (Hobfoll, 2001;
Hobfoll & Freedy, 1993). Over time, support from
family allows employees to meet demands and guard
against further resource depletion, manifested as
strain (Wright & Hobfoll, 2004).
Spouse instrumental support and emotional ex-
haustion. Emotional exhaustion, a form of work-
related strain, is characterized by energy loss and the
feeling that one’s emotional resources have been
depleted (Cordes & Dougherty, 1993). Individuals
are likely to experience emotional exhaustion when
they feel they no longer have the necessary resources
to handle the stressors confronting them (Hobfoll,
1989; Lee & Ashforth, 1996). We focus exclusively
on emotional exhaustion because it is widely seen as
the “core” dimension of job burnout (Demerouti,
Bakker, Nachreiner, & Schaufeli, 2001; Maslach,
1982; Shirom, 2003).
It is typically believed that instrumental support
will impede emotional exhaustion, as it provides the
resources needed to cope with stressful work events
(Greenglass, Fiskenbaum, & Burke, 1996; Koniarek
& Dudek, 1996; Thompson, Kirk, & Brown, 2005).
While there is evidence of an inverse relationship
between the broader concept of family support and
emotional exhaustion (Halbesleben, 2006; Lapierre
& Allen, 2006), spouse instrumental support has not
received a great deal of empirical attention (Aryee et
al., 1999).
We argue that spouse instrumental support, as a
resource, will be negatively associated with emo-
tional exhaustion. There has not been a great deal of
empirical attention given to the role that spouse in-
strumental support plays in reducing emotional ex-
haustion; however, such a relationship is consistent
with the notion of resource enrichment across roles,
whereby resources from one domain (family) help to
address issues in another domain (work) (cf., Roth-
bard, 2001; see also Barnett & Hyde, 2001). While
not a direct test of the proposed spouse instrumental
support-emotional exhaustion relationship, Neal and
Hammer (2007) reported that for employees caring
for children and their aging parents, the quality of the
spousal role was a significant predictor of work-
related well-being, even more than quality of work
role. This finding and the enrichment perspective of
roles suggest that spouse instrumental support should
help reduce emotional exhaustion.
Hypothesis 2: Spouse instrumental support will be
negatively associated with emotional exhaustion.
Work-Linked Couples and Spouse
Instrumental Support
Silberstein (1992) has noted that spouses in the
same field speak positively of spouse support and the
benefits arising from “understanding each other’s
374 HALBESLEBEN, ZELLARS, CARLSON, PERREWE
´, AND ROTONDO
work and from the camaraderie that comes from
shared professional interests” (p. 83). This suggests
that working with one’s spouse, either linked through
the same workplace or occupation, may increase the
effectiveness of spouse instrumental support by at-
tenuating the negative relationship between spouse
instrumental support and emotional exhaustion.
Spouse instrumental support reflects assistance
that is task-focused (Brown, Westbrook, & Chall-
agalla, 2005). Examples of spouse instrumental sup-
port include assistance by running errands or taking
on additional responsibilities for one’s spouse.
Spouse instrumental support is relevant to the current
study because it is most likely to differ based on
whether an employee works in the same workplace or
occupation as his of her spouse. COR theory argues
the relevance of the support in addressing the source
of strain is important. Resources only hold value if
they can be utilized appropriately (Hobfoll, 1998).
One might imagine a situation of having countless
riches, only in the wrong currency. In order for
spouse instrumental support to truly be effective in
addressing stressors (and thus, lower strains such as
emotional exhaustion), it must fit with the employ-
ee’s context (Himle, 1991; Podsakoff, Todor, &
Schuler, 1983).
Work-linked couples are in a better position to
provide spouse instrumental support because they
may have a better understanding for their spouses’
demands at work due to greater familiarity with the
challenges the spouse is facing. Work-linked couples
are likely to have a deeper appreciation of the work
component of their spouse’s life and, therefore, may
be able to offer more useful spouse instrumental
support. In particular, work-linked couples may be
more adept at helping each other with task-related
activities and technical advice, including activities at
home. Someone working at the same organization as
his or her spouse may understand that a specific time
of year is particularly busy for his or her spouse and
offer more assistance at home during this period,
allowing the spouse to focus more on work. There-
fore, the spouse instrumental support that is devel-
oped by work-linked couples is likely to be more
relevant and, thus, more effective than in couples
without a work link in addressing work demands and
reducing strain (even if the help occurs in the home;
Sieber, 1974).
We will note that at this point in our hypothesis
development, we are no longer considering the com-
bined effects of sharing both the same workplace and
same occupation. We have taken this approach for
three reasons. First, while there is literature to sup-
port the idea that outcomes may differ when both
workplace and occupation are shared (as opposed to
only sharing workplace or occupation; Janning,
2006), there is nothing in the literature to suggest that
the combination of workplace and occupation should
impact the relationship between other variables. Sec-
ond, by not combining the two work-links, we are in
a better position to empirically determine the unique
effect of each type of work-link. Combining the work
links would make it more difficult to interpret the
individual impact of sharing a workplace versus shar-
ing an occupation. Finally, our findings indicate that
the combination of same workplace and same occu-
pation do not lead to higher work–family integration
and, thus, do not seem to actually impact the vari-
ables in the study in a unique manner.
Hypothesis 3a: Working in the same workplace
(but not the same occupation) will moderate the
relationship between spouse instrumental sup-
port and emotional exhaustion such that the re-
lationship between spouse instrumental support
and emotional exhaustion will be stronger
among those who share a workplace.
Hypothesis 3b: Working in the same occupation
(but not the same workplace) will moderate the
relationship between spouse instrumental sup-
port and emotional exhaustion such that the re-
lationship between spouse instrumental support
and emotional exhaustion will be stronger
among those who share an occupation.
The Role of Work–Family Integration
The argument we propose thus far suggests that the
relationship between spouse instrumental support and
exhaustion should be moderated by working in the
same workplace or occupation as one’s spouse,
where the negative relationship between spouse in-
strumental support and exhaustion is greater for
work-linked couples. This reasoning was built
largely on the assumption that work-linked couples
would have more integrated roles than those who do
not share a workplace or occupation with their
spouse. Thus, if we combine the logic linking spouse
occupation and workplace with spouse instrumental
support and exhaustion with the logic leading to
Hypothesis 1 (linking sharing an occupation and/or
workplace with one’s spouse and its impact on work-
family integration), it would suggest that work-
family integration is the process by which sharing a
workplace or occupation impacts the spouse instru-
375WORK-LINKED COUPLES AND WORK–FAMILY INTEGRATION
mental support-emotional exhaustion relationship. In
other words, while sharing an occupation or work-
place with one’s spouse moderates the spouse instru-
mental support-emotional exhaustion relationship,
moderated effect occurs because work-linked couples
tend to have more integrated work and family roles.
As discussed earlier, boundary theory and conser-
vation of resources theory predict that high levels of
work–family integration allow support resources to
transfer from family to work (Ashforth et al., 2000;
Greenhaus & Powell, 2006; Marks, 1977; Sieber,
1974). This suggests that because it will be easier to
obtain resources from the family role to be utilized in
the work role, the negative relationship between
spouse instrumental support and emotional exhaus-
tion should be enhanced among those with high
work–family integration. Thus, we expect that work-
family integration moderates the relationship be-
tween spouse instrumental support and emotional
exhaustion.
Hypothesis 4: Work–family integration moder-
ates the relationship between spouse instrumen-
tal support and emotional exhaustion such that
high levels of work–family integration enhance
the negative relationship between spouse instru-
mental support and emotional exhaustion.
Taken together, these hypotheses suggest a medi-
ated moderation model (see Figure 1). As argued
earlier, work-linked couples will experience greater
work–family integration (H1a, 1b, and 1c) than cou-
ples without a work link. We expect both sharing a
workplace or occupation with one’s spouse (H3a and
H3b) and work–family integration (H4) will moder-
ate the relationship between spouse instrumental sup-
port and emotional exhaustion; however, we believe
that the moderation effect of sharing a workplace or
occupation with one’s spouse works through (or is
the result of) the higher level of work–family inte-
gration found in work-linked couples (H1a and H1b).
Thus, based on theory as well as our prior hypothe-
ses, we propose the following hypothesis:
Hypothesis 5: The moderating effect of working
in the same workplace or occupation with one’s
spouse (H3a, H3b) on the spouse instrumental
support-emotional exhaustion relationship will
be mediated by the effect of work–family inte-
gration (H4).
We tested the model in two studies. In Study 1, we
provide an exploratory test of Hypotheses 1a, 1b, and
1c, establishing a relationship between spouse work-
place and occupation and work–family integration
among a sample of nurses. In Study 2, we provide a
full test of the model using data from participants in
a variety of work settings using two data collection
points. We present the method and results from each
study and then integrate them in the discussion sec-
tion.
Study 1: Method
Participants and Procedure
The participants were 103 registered nurses from a
116-bed community hospital in the Midwestern
United States. The sample included 89 women and 14
men. They had an average organizational tenure of
9.38 years (SD ⫽7.35) and an average age of 41.12
years (SD ⫽9.94). Most were of Caucasian descent
(69%). A slight majority (52%) had children living at
home. Comparison with the demographic make-up of
the organization suggests that the sample is highly
consistent with the organization’s nurse population
(e.g., no significant differences in gender, age, tenure,
and race).
Paper surveys were distributed in employees’
mailboxes; upon completion, they were returned di-
rectly to the first author via business reply mail. One
hundred fifty surveys were distributed and 103 com-
plete surveys from married participants were returned
for a final response rate of 69%.
Measures
Spouse workplace and occupation. We asked
each married participant about the workplace and
occupation of their spouse. Options included “Does
not work outside the home,” “Works in the same
occupation as you, in the same company as you,”
“Works in the same occupation as you, in a different
company as you,” “Works in a different occupation
from you, but in the same company as you,” “Works
in a different occupation from you, and in a different
company as you” allowing us to differentiate be-
tween spouses that worked in the same or different
occupation and the same or different workplace. Of
the 103 participants, 20 (19%) indicated that they
worked in the same occupation and same company as
their spouse; 15 (15%) indicated they worked in the
same occupation but in a different company; and 16
(15%) indicated they work in a different occupation
but in the same company as their spouse. The remain-
ing 52 participants were categorized as not having a
376 HALBESLEBEN, ZELLARS, CARLSON, PERREWE
´, AND ROTONDO
work link; their spouse worked but in a different
occupation and different company.
In this study, we test the main effects of both
same-occupation and same-workplace situations.
While previous theoretical work has not been clear
regarding whether differences in these forms of mar-
ried coworker arrangements should be expected, an-
alyzing them separately is consistent with past re-
search (Janning, 2006, 2009) and is appropriate for
emerging research. Thus, in the analysis, we coded
the participants to contrast same occupation (1) ver-
sus different occupation or different workplace (0).
We also coded each participant for same workplace
(1) versus different workplace or different occupa-
tion (0).
Work–family integration. Work–family inte-
gration was measured using the Work–Family Inte-
gration-Blurring Scale (WFIBS) of Desrochers et al.
(2005). It is a three-item scale (e.g., “It is often
difficult to tell where my work life ends and my
family life begins”). We adapted one item slightly;
specifically “In my life, there is a clear boundary
between my career and my role as a parent” was
adapted to read “In my life, there is a clear boundary
between my career and my family role” since it was
assumed that not all participants would be parents.
Participants indicated their agreement with each item
on a five-point, Likert-type scale from strongly dis-
agree (1) to strongly agree (5). Higher scores indi-
cated higher integration of the work–family role
boundary. Desrochers et al. (2005) reported adequate
internal consistency of the scale (.73) and evidence
for the construct validity of the scale, finding that it
was significantly correlated with measures of work-
family conflict, hours worked, distractions while
working at home, and work–family transitions. In the
present study, the internal consistency reliability co-
efficient (Cronbach’s alpha) was .76.
Control variables. Following the design of Des-
rochers et al. (2005), we controlled for gender, age,
children at home, and hours worked per week.
Study 1: Results
The mean levels of work–family integration by
married coworker groups are reported in Table 1
(because the control variables were not significantly
related to work–family integration in our study, we
report nonadjusted means). We tested Hypothesis 1
using a general linear modeling approach where we
dummy coded spouse workplace (0 ⫽different work-
place and occupation;1⫽same workplace) and
spouse occupation (0 ⫽different occupation and
occupation;1⫽same occupation). We compared the
same workplace and same occupation with the group
that did not share either workplace or occupation.
This was done in order to truly compare the effect of
married coworker situations to situations where there
is no overlap in either occupation or workplace. After
entering control variables in step 1, we entered the
main effects of spouse workplace and spouse occu-
pation in the second step (see Table 2). We found a
significant main effect for spouse workplace (⫽
.52, p⬍.01), suggesting that those participants who
share a workplace with their spouse experience sig-
nificantly greater work–family integration. Similarly,
we found a significant main effect for spouse occu-
pation (⫽.46, p⬍.01), suggesting that those
participants who share an occupation with their
spouse experience significantly greater work–family
integration (see Table 2).
In step 3 of the analysis, we examined the two-way
interaction between spouse occupation and spouse
workplace on work–family integration. This test ex-
amined the potential interaction effect between
spouse occupation and spouse workplace such that
sharing both would yield a unique, interactive effect
beyond the main effects of each. That interaction was
not significant (⫽.10, ns) and adding the interac-
tion led to a nonsignificant increase in R
2
(see Table
2). Thus, Hypothesis 1c was not supported.
These findings support Hypothesis 1a and 1b that
sharing a workplace or occupation with one’s spouse
is associated with greater work–family boundary in-
tegration compared to not sharing either a workplace
or occupation. Hypothesis 1c was not supported, sug-
gesting that it is the existence of a link that is asso-
ciated with higher work–family integration but not
necessarily the number of links (e.g., two links, work
and occupation, is not necessarily better than just one
link). With the links between spouse workplace,
spouse occupation, and work–family integration es-
tablished, we tested the full mediated moderation
model from Figure 1 in Study 2.
Table 1
Mean (Standard Deviation) Levels of Work–Family
Integration by Spouse Workplace and Occupation
Same
workplace
Different
workplace
Same occupation 4.49 (.84) 3.86 (.58)
N⫽20 N ⫽15
Different occupation 4.09 (.49) 2.62 (.51)
N⫽16 N ⫽52
377WORK-LINKED COUPLES AND WORK–FAMILY INTEGRATION
Study 2: Method
Participants and Procedure
The data were collected with the assistance of 94
undergraduate business students as part of a research
experience assignment in an organizational behavior
course. The students recruited up to six working
adults to complete two online surveys. The first sur-
vey included measures of marital status, spouse in-
strumental support, work–family integration, emo-
tional exhaustion, and controls (i.e., demographics
and coworker support). The second survey, con-
ducted two months after the first survey, confirmed
the workplace and occupation of the spouse and
measured exhaustion. While there was no theoretical
rationale for a 2-month lag between surveys, we
utilized the lag in order to introduce temporal sepa-
ration between variables in the model with the goal
of reducing common method bias (Podsakoff, Mac-
Kenzie, Lee, & Podsakoff, 2003).
Given the nature of the study, the students were
asked to specifically target married, cohabiting par-
ticipants (they were not asked to specifically recruit
work-linked couples). They were asked to recruit
only one half of the married couple (e.g., only the
husband or the wife, but not both). As an incentive to
participate in the surveys, the participants entered
into a drawing for tickets to local arts events. A total
of 558 individuals were asked to complete the first
survey; 541 participants submitted surveys for an
initial response rate of 97%. For the purposes of this
study, only the participants who indicated that they
were married and who had a spouse that worked were
retained for analysis, leaving a potential sample of
519. Responses from eight participants were not an-
alyzed because of incomplete data, leaving a final
usable sample of 511 participants; a response rate
(from the initial 558 requests for surveys) of 92%.
All 511 participants were contacted via e-mail to
complete a second round survey. A total of 489
participants responded to the second round survey; of
those 487 provided complete data. We dropped one
of the 487 surveys because both spouses had an-
swered the survey. In order to maintain the indepen-
dence of the data, we used the data from the first of
the spouses to respond to the survey and dropped the
other spouse’s data. We matched the data from the
two surveys by asking the participant to provide his
or her name on the survey (this was also used for
verification purposes described below); participants
were assured that their data would remain confiden-
tial and the names were deleted from the database
once the matching occurred. In the second round
survey, one person’s marital status had changed (i.e.,
no longer married), and two participants no longer
worked in the same occupation as their spouse; their
data were dropped from further analysis. The final
sample size for Study 2 was 484, leaving a final
response and retention rate across the two data col-
lections of 87%.
To ensure that the surveys were completed by the
working adults, we randomly selected 75% of the
surveys after each round of data collection and di-
rectly contacted the participant to verify their partic-
Table 2
Study 1 Multiple Regression Results: Spouse Workplace and Spouse Occupation as Predictors of
Work–Family Integration
Step 1 Step 2 Step 3
(Std error) (Std error) (Std error)
Sex .02 .01 .02 .01 .02 .01
Age .05 .02 .04 .02 .04 .02
Children at home .02 .01 .01 .01 .04 .03
Hours per week .05 .03 .05 .03 .05 .03
Spouse workplace .52
ⴱⴱ
.13 .37
ⴱⴱ
.10
Spouse occupation .46
ⴱⴱ
.16 .36
ⴱⴱ
.08
Spouse workplace ⫻Spouse Occupation .10 .09
R
2
.02 .21 .22
⌬R
2
.19
ⴱⴱ
.01
Note. N⫽103.Dependent variable ⫽work–family integration. Spouse Workplace (0 ⫽no work link,1⫽same
workplace); Spouse Occupation (0 ⫽no work link,1⫽same occupation); Sex (0 ⫽male,1⫽female); Children at Home
(0 ⫽no children at home,1⫽children at home).
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
378 HALBESLEBEN, ZELLARS, CARLSON, PERREWE
´, AND ROTONDO
ipation. All of the contacted participants verified that
they had completed the survey.
The final sample included 225 males and 259
females with a mean age of 39.62 years. The partic-
ipants had been working for their current organiza-
tion for a mean of 10.88 years and most (77%) were
Caucasian. They worked in a primarily lower or
middle management positions (n⫽288, 60%),
though front-line (n⫽124, 26%) and executive/
ownership (n⫽59, 12%) were also represented. A
wide variety of industries were represented, with
education (22%), health care (20%), government
(17%), financial services (15%), and manufacturing
(10%) most highly represented. The participants pri-
marily lived and worked in the upper Midwestern
United States (92%).
Measures
Spouse workplace and occupation. As in Study
1, we asked each married participant about the work-
place and occupation of their spouse. Of the 484
participants, 67 (14%) indicated that they worked in
the same occupation and same workplace as their
spouse; 48 (10%) indicated they worked in the same
occupation but in a different workplace; and 41 (8%)
indicated they work in a different occupation but in
the same workplace as their spouse. The remaining
328 participants were categorized as in a couple
without a work link. In the analysis, we coded the
participants to contrast same occupation (1) versus
no work link (0). We also coded each participant for
same workplace (1) versus no work link (0).
We acknowledge that this coding procedure ex-
cluded participants in the analyses who shared both a
workplace and occupation. As we noted in the intro-
duction, we took this approach to disentangle the
effects of each type of work link and because the
combination of the two links appeared not to have a
significant effect on work–family integration in
Study 1.
Work–family integration. As in Study 1, work-
family integration was measured using the three-item
Work–Family Integration-Blurring Scale (WFIBS) of
Desrochers et al. (2005) with the adaption of one item
noted previously. This was assessed at Time 1.
Spouse instrumental support. At Time 1, we
assessed spouse instrumental support using an
adapted version of the Family Support Inventory for
Workers (King, Mattimore, King, & Adams, 1995).
This measure was used in order to provide a more
comprehensive measure of perceived social support.
The measure includes 15 items to assess instrumental
support; a sample item is, “If my job gets very
demanding, my family members will take on extra
household responsibilities.” The items were adapted
so that references to family were changed to spouse.
For example, the sample item above was changed to
“If my job gets very demanding, my spouse will take
on extra household responsibilities.” The items were
scored on a five point Likert-type scale from strongly
disagree (1) to strongly agree (5) with higher scores
indicating greater spouse instrumental support.
Emotional exhaustion. Emotional exhaustion
was measured using the emotional exhaustion sub-
scale of the Maslach Burnout Inventory–General
Scale (MBI-GS; Schaufeli, Leiter, Maslach, & Jack-
son, 1996). The items were scored on a seven-point
frequency scale from never (0) to every day (6). This
measure was completed at both Times 1 and 2.
Control variables. Based on a review of the
literature, we identified six additional variables that
may covary with spouse instrumental support and/or
emotional exhaustion and, thus, were treated as con-
trol variables in the analysis. We controlled for sex,
age, organization tenure, education level, and race
because of their potential relationships with support
resources in the workplace.
Coworker support has a significant negative rela-
tionship with emotional exhaustion (Halbesleben,
2006). We controlled for coworker instrumental sup-
port because in a same-career setting, the spouse may
be considered a coworker, potentially confounding
coworker, and spouse support for those employees.
The social support measure of Caplan, Cobb, French,
Harrison, and Pineau (1975) was utilized to measure
perceived instrumental support from each partici-
pant’s coworkers. The scale included two items; a
sample item is, “How much does each of these people
go out of their way to do things to make your life
easier for you?” Each item was followed by target
people, including coworker, and was scored on a
five-point scale from not at all (1) to very much (5).
Finally, to take advantage of the longitudinal nature
of our design, we controlled for Time 1 emo-
tional exhaustion.
Study 2: Results
The descriptive and reliability statistics and corre-
lations for all Study 2 variables are displayed in
Table 3. All measures achieved acceptable reliability.
The model we have proposed represents a medi-
ated moderation model; that is, the moderation effect
of work–family links is mediated by work–family
integration. To test the model, we employed the
379WORK-LINKED COUPLES AND WORK–FAMILY INTEGRATION
procedures outlined by Edwards and Lambert (2007)
by utilizing a path analysis framework that employs a
series of regression equations. Their approach im-
proves upon more traditional methods for testing
mediated moderation (e.g., Baron & Kenny, 1986;
Muller, Judd, & Yzerbyt, 2005) by testing the direct,
indirect, and total effects of the moderator variable in
a nested form.
1
The regression results from each model are displayed
in Table 4. We tested three nested models. The Edwards
and Lambert (2007) approach is built from a mediated
model. Model 1 was a mediated model where work-
family integration was treated as a mediator to the
spouse instrumental support-exhaustion relationship
(with the work–family links included as predictors of
exhaustion, essentially as control variables). Model 2
was the predicted mediated moderation model that in-
cluded the mediating effect of work–family integration
on the interaction between work–family links and
spouse support in predicting exhaustion. It adds the
interaction between work–family interaction and spouse
support in predicting exhaustion and paths from each
work–family link to work–family integration. Finally,
Model 3 adds the direct interaction between work-
family links and spouse support in order to test both the
direct and indirect (mediated) effects of the moderator.
In theory, the direct effect of work–family link as a
moderator should not be significant since it essentially
“working through” work–family integration. Following
current recommendations (Mooney & Duval, 1993;
Shrout & Bolger, 2002), we used bootstrapped esti-
mates from 10,000 samples to create construct bias-
corrected confidence intervals for all significance tests.
It is possible to compare the nested models by comput-
ing a generalized R
2
and comparing the models with Q
and W statistics (see Pedhauzer, 1982; Tepper, Henle,
Lambert, Giacaolone, & Duffy, 2008). The Q statistic
allows for a comparison of the generalized R
2
to its
maximum value of 1 (which would indicate that the two
models are not different in terms of generalized R
2
).
The W statistic is a conversion of the Q statistic that is
on a chi-squared distribution, thus allowing for a sig-
nificance test of the Q value.
As indicated in Table 4, the models accounted for
12% of the variance in work–family integration (p⬍
.01). Model 1 explained 49% of the variance in emo-
tional exhaustion (p⬍.01) while Models 2 and 3
1
To conserve page space, we do not report how we
derived each regression equation (for an example, see
Tepper, Henle, Lambert, Giacalone, & Duffy, 2008).
These are available from the first author.
Table 3
Study 2 Descriptive Statistics and Intercorrelations Among Study Variables
MSD 123456789101112
1. Sex 0.53 0.41 —
2. Age 39.62 9.06 ⫺.08 —
3. Tenure 10.88 7.02 ⫺.17
ⴱ
.39
ⴱⴱ
—
4. Race 1.56 1.01 ⫺.05 ⫺.08 ⫺.03 —
5. Education 3.91 1.25 ⫺.19
ⴱⴱ
.15
ⴱ
⫺.08 .00 —
6. Work–family integration 2.58 0.98 .09
ⴱ
.06 .00 .03 .04 (.81)
7. Time 1 emotional exhaustion 2.99 1.16 .08
ⴱ
.04 .01 .02 .04 ⫺.24
ⴱⴱ
(.94)
8. Spouse workplace 0.08 .07 .01 .03 .04 .02 .04 .41
ⴱⴱ
⫺.34
ⴱⴱ
—
9. Spouse occupation 0.10 .09 .05 .00 ⫺.01 .01 .04 .37
ⴱⴱ
⫺.36
ⴱⴱ
.05 —
10. Coworker instrumental support 3.45 1.19 ⫺.02 .01 .06 .03 .00 .10
ⴱ
⫺.21
ⴱⴱ
.03 .05 (.86)
11. Spouse instrumental support 3.20 0.98 ⫺.06 .02 .05 .01 .00 .26
ⴱⴱ
⫺.19
ⴱⴱ
.30
ⴱⴱ
.29
ⴱⴱ
.11
ⴱ
(.92)
12. Time 2 emotional exhaustion 2.85 1.05 .09
ⴱ
.00 .05 .06 ⫺.02 ⫺.34
ⴱⴱ
.49
ⴱⴱ
⫺.35
ⴱⴱ
⫺.40
ⴱⴱ
⫺.21
ⴱⴱ
⫺.18
ⴱⴱ
(.93)
Note.N⫽484. Internal consistency estimates (Cronbach’s alpha) appear in parentheses along the diagonal. Spouse Workplace (0 ⫽different occupation and workplace,1⫽same workplace);
Spouse Occupation (0 ⫽different occupation and workplace,1⫽same occupation); Sex (0 ⫽male,1⫽female); Education (0 ⫽some high school,1⫽high school graduate,2⫽some
college/vocational or technical degree,3⫽college graduate,4⫽some graduate studies,5⫽graduate degree); Race (0 ⫽White/Caucasian,1⫽non-White/Caucasian).
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
380 HALBESLEBEN, ZELLARS, CARLSON, PERREWE
´, AND ROTONDO
accounted for 68% of the variance in emotional exhaus-
tion (p⬍.01). Comparisons of the models reveal that
the generalized R
2
for Model 2 (R
Generalized
2
⫽.72) was
significantly different from the generalized R
2
for
Model 1 (R
Generalized
2
⫽.57; Q ⫽.63, W ⫽223.16, d⫽
1, p⬍.01). This suggests that the mediated moderation
model provides better fit compared to a more simple
mediation model; in other words, the addition of the
interaction between work–family links and work–
family integration increases the variance in emotional
exhaustion explained significantly. Additionally, the
generalized R
2
for Model 2 (R
Generalized
2
⫽.72) was
significantly different from the generalized R
2
for
Model 3 (R
Generalized
2
⫽.73; Q ⫽.97, W ⫽13.78,
d⫽1, p⬍.01). This finding suggests that the direct
moderating effect of spouse instrumental support and
work-family links is adding to the predictive power
of the model. However, when we examine the sig-
nificance of those paths (see Table 1, column for
Model 3), we find that they are not significant. In this
case, the W statistic may have indicated a significant
difference due to very high R
2
values and a large
sample size.
2
Next, we examined the path estimates for the hy-
pothesized model (Model 2 in Table 2). Spouse in-
strumental support was associated with work–family
integration (b⫽.25, p⬍.01). It was also negatively
associated with emotional exhaustion (b⫽⫺.16, p⬍
.01), supporting Hypothesis 2. Sharing a workplace
with one’s spouse was significantly negatively re-
lated to emotional exhaustion (b⫽⫺.30, p⬍.01) as
was sharing an occupation with one’s spouse (b⫽
⫺.31, p⬍.01). Work–family integration was also
significantly negatively related to emotional exhaus-
tion (b⫽⫺.30, p⬍.01). Finally, the interaction
between work–family integration and sharing a
workplace with one’s spouse was significantly re-
lated to emotional exhaustion (b⫽.34, p⬍.01). The
same held true for the interaction between work–
family integration and sharing an occupation with
one’s spouse (b⫽.28, p⬍.01).
Based on the results from the hypothesized model,
we calculated the simple effects for same workplace
couples versus couples with no work link and same
occupational couples versus with no work link (see
Table 5). Analysis of these simple effects suggests
that for same workplace participants the path from
spouse instrumental support to work–family integra-
tion was significant (p⫽.25, p⬍.01), and the path
from work–family integration and emotional exhaus-
tion was also significant (P ⫽⫺.50, p⬍.01), as was
the indirect effect of spouse instrumental support on
emotional exhaustion (P ⫽⫺.13, p⬍.05). For same
2
This was, in turn, partially explained by including Time
1 emotional exhaustion as a control variable. When we
tested the models without Time 1 emotional exhaustion, the
R
2
for emotional exhaustion in the models is much lower
and the difference between Models 2 and 3 is no longer
significant.
Table 4
Study 2 Path Analytic Tests of Hypothesized Mediated Moderation Model and Alternative Models
Path estimated Model 1 Model 2 Model 3
Spouse support 3Work–family integration .25
ⴱⴱ
.25
ⴱⴱ
.25
ⴱⴱ
Spouse support 3Emotional exhaustion ⫺.16
ⴱⴱ
⫺.16
ⴱⴱ
⫺.14
ⴱ
Work–family integration 3Emotional exhaustion ⫺.30
ⴱⴱ
⫺.30
ⴱⴱ
⫺.25
ⴱⴱ
Spouse workplace 3Emotional exhaustion ⫺.32
ⴱⴱ
⫺.30
ⴱⴱ
⫺.29
ⴱⴱ
Spouse occupation 3Emotional exhaustion ⫺.37
ⴱⴱ
⫺.31
ⴱⴱ
⫺.28
ⴱⴱ
WFI ⫻Sp. workplace 3Emotional exhaustion .34
ⴱⴱ
.32
ⴱⴱ
WFI ⫻Sp. occupation 3Emotional exhaustion .28
ⴱⴱ
.25
ⴱⴱ
Spouse workplace 3Work–family integration .47
ⴱⴱ
.43
ⴱⴱ
Spouse occupation 3Work–family integration .51
ⴱⴱ
.46
ⴱⴱ
Spouse ins. support ⫻Sp. workplace 3Emotional exhaustion .11
Spouse ins. support ⫻Sp. occupation 3Emotional exhaustion .09
R
Work–Family Integration
2
.12
ⴱⴱ
.12
ⴱⴱ
.12
ⴱⴱ
R
Emotional Exhaustion
2
.49
ⴱⴱ
.68
ⴱⴱ
.69
ⴱⴱ
R
Generalized
2
.55
ⴱⴱ
.72
ⴱⴱ
.73
ⴱⴱ
Note.N⫽484. Spouse Workplace (0 ⫽no work link,1⫽same workplace); Spouse Occupation (0 ⫽no work link,1⫽
same occupation). Table values are path estimates for each respective model test. Model 1 is the simple mediation model
(where work–family integration is the mediator). Model 2 is the hypothesized mediated moderation model. Model 3
includes the indirect moderating effect of the work–family links. WFI ⫽Work–Family Integration.
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
381WORK-LINKED COUPLES AND WORK–FAMILY INTEGRATION
occupation participants, the path from spouse instru-
mental support to work–family integration was sig-
nificant (p⫽.25, p⬍.01), and the path from work-
family integration and emotional exhaustion was also
significant (P ⫽⫺.58, p⬍.01), as was the indirect
effect of spouse instrumental support on emotional
exhaustion (P ⫽⫺.15, p⬍.05). For the participants
not in work-linked couple, the path from spouse
instrumental support to work–family integration was
also significant (p⫽.25, p⬍.01), and the path from
work–family integration and emotional exhaustion
was significant (P ⫽⫺.10, p⬍.05), though again
much lower than the same path among same occu-
pation participants. For different occupation partici-
pants, the indirect effect of spouse instrumental sup-
port on emotional exhaustion was not significant
(P ⫽⫺.03, ns). Taken together, the analysis of the
simple effects gives further credence to our proposed
mediated moderation model as the pattern of effects
is different based on whether or not a work-link
exists.
Note also in Table 5 that the direct effect of spouse
instrumental support on emotional exhaustion is also
quite different depending on the nature of the work
link, specifically whether or not a link exists in the
couple. For same workplace participants, the path is
significant (P ⫽⫺.29, p⬍.01). Further, among same
occupation participants the effect is significant (P ⫽
⫺.31, p⬍.01) while the effect is not significant for
couples without a work link (P ⫽⫺.08, ns). This
supports Hypotheses 3a and 3b.
To illustrate the mediated moderation effect, we
graphed the mediated (through work–family integra-
tion) effects of work–family links on the relationship
between spouse instrumental support and emotional
exhaustion (see Figure 2). The results suggest that
those with a work-link (either sharing a workplace or
an occupation; those sharing both are not included in
the graph) have a negative relationship between
spouse instrumental support and exhaustion. On the
other hand, those without a work link do not have a
significant relationship between spouse instrumental
support and exhaustion when their work–family in-
tegration is accounted for. Taken together, these find-
ings support Hypothesis 5 regarding the mediated
moderation effect of work–family links and work-
family integration on the relationship between spouse
instrumental support and emotional exhaustion.
Discussion
The two studies presented shed light on the spouse
instrumental support to emotional exhaustion rela-
tionship in married couples. First, we found a signif-
icant relationship between work-linked relationships
and work–family integration. Spouses that shared a
workplace and spouses that shared an occupation had
higher work–family integration than those who did
not share a workplace or occupation. These findings
were consistent with the suggestions of Janning
Table 5
Direct and Indirect Effects of Spouse Instrumental Support on Emotional Exhaustion by Married
Coworker Arrangements in Study 2
Path P
MX
P
YM
Direct Effects
(P
YX
)
Indirect Effects
(P
YM
P
MX
)
Total Effects
(P
YX
⫹P
YM
P
MX
)
Simple paths for same workplace couples .25
ⴱⴱ
⫺.50
ⴱⴱ
⫺.29
ⴱⴱ
⫺.13
ⴱ
⫺.42
ⴱⴱ
Simple paths for same occupation couples .25
ⴱⴱ
⫺.58
ⴱⴱ
⫺.31
ⴱⴱ
⫺.15
ⴱ
⫺.39
ⴱⴱ
Simple paths for couples without a work link .25
ⴱⴱ
⫺.10
ⴱ
⫺.08 ⫺.03 ⫺.10
ⴱ
Note. N⫽484. P
MX
⫽path from spouse support to work–family integration; P
YM
⫽path from work–family integration
to emotional exhaustion; P
YX
⫽path from spousal support to emotional exhaustion.
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
Emoonal Exhauson
Spouse Instrumental Support
Work−Family Link
Figure 2. Interaction between the indirect (mediated) ef-
fect of spouse instrumental support and work links on
emotional exhaustion in Study 2.
382 HALBESLEBEN, ZELLARS, CARLSON, PERREWE
´, AND ROTONDO
(2009) and Moen and Sweet (2002) that work-linked
couples would have more highly integrated work and
family roles. It is interesting to note that as demon-
strated in Study 1, the combination of sharing both a
workplace and an occupation was not significantly
different from just sharing one link, either occupation
or workplace.
Second, our findings also support the hypothesized
impact of work links and work–family integration on
the spouse instrumental support-emotional exhaus-
tion relationship, finding that work-links impact the
spouse instrumental support-emotional exhaustion
relationship indirectly through work–family integra-
tion. Such findings are consistent with recent work on
the work–family interface that suggests family expe-
riences can improve the quality of work life (Carlson
& Grzywacz, 2008; Greenhaus & Powell, 2006);
moreover, they support the hypotheses derived from
boundary and conservation of resources theories re-
garding the nature of work–family resources.
Implications for Research and Theory
This study expands the research on the understud-
ied group of employees that share a workplace or
occupation with their spouse. As noted earlier, our
findings are consistent with those of Moen and Sweet
(2002), who found high levels of spillover in couples
who worked for the same company. We extend their
work by examining the effects of spouses working in
the same occupation and by looking more specifically
at the outcomes of resources that move across role
boundaries. Moreover, Janning (1999, 2006, 2009;
Janning & Neely, 2006) has examined the work-
family integration of work-linked couples, finding
that spouses who worked in the same workplace had
more conversations about work while at home. Our
work supports these findings of high integration of
work–family roles in married coworker settings, es-
pecially regarding social support in work-linked cou-
ples (Janning, 2006). We extend this research by
looking beyond conversations to other forms of in-
strumental support in the home.
It is interesting to note, however, that we did not
find differences in work–family integration or in the
moderation of spouse instrumental support and ex-
haustion based on different types of work-linked cou-
ples (e.g., same-occupation or same-workplace). This
differs somewhat from Janning (2009), where she
noted that same-workplace employees tend to bring
more work symbols into the home whereas same-
career employees tend to bring more home symbols
into work. Our study does not necessarily contradict
what she found, however, since the work–family
integration measure we used does not necessarily
account for the direction of the integration. The
work-family integration scale focuses on psycholog-
ical perceptions of the relationship between the roles
whereas Janning was focused on the use of symbols
among work-linked couples. Finally, since our mea-
sure of support focused on home-based support, one
might expect that the nature of the work-link may not
impact such support when compared to work-specific
support. Nonetheless, future research may seek to
explore potential differences between the forms of
work-linked couples, perhaps by integrating the per-
ception of work–family integration with the behav-
iors and symbols that indicate integration.
As noted in the introduction, this study takes
boundary theory in a different direction than has been
studied in the past (see also Desrochers et al., 2005;
Winkel & Clayton, 2010). To this point, studies of
boundary theory have focused on preferences for
work–family integration versus segmentation. For
example, Kreiner (2006) found that employees’ fit
between preferred and experienced work–family in-
tegration was associated with lower stress (see also
Clark, 2000; Edwards & Rothbard, 1999). We extend
boundary theory, integrating it with COR theory, to
explain why highly integrated work and family roles
may allow for an easier transmission of resources
from one role to another and how that easier transi-
tion leads to lower strain for employees. Much of the
previous research on boundary theory has focused on
the potential costs of permeable boundaries (e.g.,
Ashforth et al., 2000; see also Kossek et al., 2005);
the current study expands this body of work by
testing a potential benefit of permeable boundaries—
greater resource sharing via spouse instrumental sup-
port. Researchers have offered some evidence for a
positive relationship between integrated roles and
higher performance (see Kossek et al., 2005). We
offer evidence for a second benefit, reduced emo-
tional exhaustion.
Limitations and Directions for
Future Research
We recognize that there are a number of method-
ological limitations to the present work that must be
addressed. We acknowledge that this was a limited
sample of participants working in one occupation or
in one organization. In Study 2, we improve upon this
limitation by studying a larger, more diverse sample
of participants. Additionally, the data from Study 1
383WORK-LINKED COUPLES AND WORK–FAMILY INTEGRATION
are single-source data collected at a single point in
time, which may lead them to susceptibility to com-
mon method bias. To test for this possibility, we
conducted Harman’s one-factor test to determine
whether the measured (nondemographic) scales were
influenced significantly by a common measurement
factor (cf., Mossholder, Bennett, Kemery, &
Weslowski, 1998). Using confirmatory factor analy-
sis, we tested a model that loaded all of the measured
variables onto one factor, finding that it provided
relatively poor fit to the data (analyses available from
first author). While this test cannot rule out the pos-
sibility of common method bias (Podsakoff et al.,
2003), the consistency of the results, especially when
we tested the model using time lagged data in Study
2, suggests that factors other than common methods
are influencing the data.
Our measure of spouse instrumental support in
Study 2 is limited in the sense that the items empha-
size the manner in which the spouse would provide
instrumental support at home rather than exclusively
work-related support behaviors. This is a result of
adapting the measure; it was created for family mem-
bers broadly, without the expectation that some fam-
ily members (e.g., spouse) would be able to help out
at work. Future research that develops a valid mea-
sure of spouse helping behavior at work could help to
address this issue.
Our study does not account for the point where
integrated roles become blurred. Blurring has been
positioned as an extreme form of integration (Ahr-
entzen, 1990; Ashforth et al., 2000) where work and
family demands appear simultaneously in the same
context (Desrochers et al., 2005). That said, it is not
clear how the two terms are entirely different. For
example, Desrochers et al. (2005) refer to integra-
tion-blurring in their hypotheses and refer to their
scale as the work–family integration-blurring scale,
but the implication is not that a low score on the scale
indicates integration and a high score indicates blur-
ring. Instead, a low score would refer to relatively
nonintegrated roles. Thus, it is not clear from their
scale at what point one could consider their roles so
integrated that they are blurred. Future research that
examines if there is a point where one’s perception
shifts from integrated to blurred boundaries would be
useful in establishing the true meaning of the term
blurring and examine whether blurred boundaries
meaningfully differ from integrated boundaries.
We recognize that there may be other variables
that help to explain the means by which work-linked
couples influence the spouse instrumental support-
emotional exhaustion relationship beyond work–
family integration. One possibility that we did not
test was the role that work-linked couples’ social
awareness or empathy might play. Given their famil-
iarity with the common stressors of an occupation,
married coworkers may have greater social aware-
ness of their spouse’s stress. Through greater social
awareness of their partner’s stressors, work-linked
couples may have greater empathy and be more
likely to be spurred to action, engaging in instrumen-
tal support at home.
Practical Implications
Despite the present study’s limitations, it may hold
important practical implications for managers seek-
ing to manage the work–family interface in a manner
that maximizes both employee and organizational
outcomes. Indeed, a number of organizations (e.g.,
DuPont) and universities have adopted policies to
hire and support work-linked couples, recognizing
that such relationships could be helpful in terms of
recruiting and retaining employees (Farley, 2007;
Wolkenbreit, 1997). The present research suggests
that beyond recruitment and retention, the nature of
the resources held by work-linked couples may pos-
itively contribute to their well-being. While we have
demonstrated that these additional resources may be
helpful in reducing strain, these benefits may extend
to other outcomes as well. For example, COR theory
suggests that having greater resources should allow
an employee to invest those resources in work, which
could lead to higher work engagement and job per-
formance (Gorgievski & Hobfoll, 2008). Indeed,
meta-analytic investigations of work engagement,
frequently conceptualized as opposite to burnout
(Salanova & Schaufeli, 2008), have found that social
support is positively associated with engagement
(Halbesleben, 2010). Moreover, studies of exhaus-
tion, engagement, and performance support the no-
tion of resource investment in work when resources
are available (e.g., Halbesleben & Bowler, 2007;
Halbesleben & Wheeler, 2008). Taken together, the
literature suggests that active recruitment of work-
linked couples, through policies to support such
work–family integration, could lead to lower strain,
higher work engagement, and higher job performance
(Halbesleben & Rotondo, 2007).
Practical barriers may exist that prevent organiza-
tions from hiring work-linked couples (e.g., a limited
number of positions). This study and the emerging
research on work-linked couples can offer sugges-
tions for all employees. For example, if instrumental
support cannot be garnered from a spouse, perhaps it
384 HALBESLEBEN, ZELLARS, CARLSON, PERREWE
´, AND ROTONDO
could be developed by supervisors (Halbesleben &
Rotondo, 2007). Programs that involve family in
work may put family members in a better position to
provide the meaningful support that work-linked cou-
ples experience regularly.
Conclusion
The present studies provide an initial examination
of the role that working with one’s spouse (either in
the same occupation or workplace) plays on the na-
ture of spouse instrumental support and emotional
exhaustion. Together, our studies expand the stress
and work–family literatures by testing for a benefit of
integrated work–family role boundaries. We believe,
pending future research, that the notion of married
coworker relationships holds promise in understand-
ing the complex dynamics of work–family roles and
social support, both practically and theoretically.
References
Adams, G. A., King, L. A., & King, D. W. (1996). Rela-
tionships of job and family involvement, family social
support, and work-family conflict with job and life sat-
isfaction. Journal of Applied Psychology, 81, 411– 420.
Ahrentzen, S. B. (1990). Managing conflict by managing
boundaries: How professional homeworkers cope with
multiple roles at home. Environment and Behavior, 22,
723–752.
Aryee, S., Luk, V., Leung, A., & Lo, S. (1999). Role
stressors, interrole conflict, and well-being: The moder-
ating influence of spousal support and coping behaviors
among employed parents in Hong Kong. Journal of Vo-
cational Behavior, 54, 259 –278.
Ashforth, B. E., Kreiner, G. E., & Fugate, M. (2000). All in
a day’s work: Boundaries and micro-role transitions.
Academy of Management Review, 25, 472– 491.
Aycan, Z., & Eskin, M. (2005). Relative contributions of
childcare, spousal support, and organizational support in
reducing work–family conflict for men and women: The
case of Turkey. Sex Roles, 53, 453– 471.
Barnett, R. C., & Hyde, J. S. (2001). Women, men, work,
and family: An expansionist theory. American Psychol-
ogist, 56, 781–796.
Baron, R. M., & Kenny, D. A. (1986). The moderator-
mediator distinction in social psychological research:
Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51, 1173–
1182.
Beehr, T. A., & Glazer, S. (2001). A cultural perspective of
social support in relation to occupational stress. In P.
Perrewe´ & D. C. Ganster (Eds.), Research in occupa-
tional stress and well-being (pp. 97–142). Greenwich,
CT: JAI Press.
Brown, S. P., Westbrook, R. A., & Challagall, G. (2005).
Good cope, bad cope: Adaptive and maladaptive coping
strategies following a critical negative work event. Jour-
nal of Applied Psychology, 90, 792–798.
Bulger, C. E., Matthews, R. A., & Hoffman, M. E. (2007).
Work and personal life boundary management: Boundary
strength, work/personal life balance, and the segmenta-
tion-integration continuum. Journal of Occupational
Health Psychology, 12, 365–375.
Caplan, R. D., Cobb, S., French, J. R., Harrison, R. U., &
Pinneau, S. R. (1975). Job demands and work health
(U.S. Department of Health, Education, and Welfare
Publication No. 75–160). Washington, DC: Institute for
Social Research.
Carlson, D. S., & Grzywacz, J. G. (2008). Reflections and
future directions on measurement in work-family re-
search. In K. Korabik, D. S. Lero, & D. L. Whitehead
(Eds.), Handbook of work-family integration: Research,
theory, and best practices (pp. 57–74). London, England:
Academic Press.
Carlson, D. S., & Perrewe´, P. L. (1999). The role of social
support in the stressor-strain relationship: An examina-
tion of work-family conflict. Journal of Management, 25,
513–540.
Clark, S. C. (2000). Work/family border theory: A new
theory of work/family balance. Human Relations, 53,
747–770.
Cordes, C., & Dougherty, T. W. (1993). A review and an
integration of research on job burnout. Academy of Man-
agement Review, 18, 621– 656.
Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli,
W. B. (2001). The job demands-resources model of burn-
out. Journal of Applied Psychology, 86, 499 –512.
Desrochers, S., Hilton, J. M., & Larwood, L. (2005). Pre-
liminary validation of the work-family integration-
blurring scale. Journal of Family Issues, 26, 442– 466.
Edwards, J. R., & Lambert, L. S. (2007). Methods for
integrating moderation and mediation: A general analytic
framework using path analysis. Psychological Methods,
12, 1–22.
Edwards, J. R., & Rothbard, N. P. (1999). Work and family
stress and well-being: An examination of person-
environment fit in the work and family domains. Orga-
nizational Behavior and Human Decision Processes, 77,
85–129.
Edwards, J. R., & Rothbard, N. P. (2000). Mechanisms
linking work and family: Clarifying the relationship be-
tween work and family constructs. Academy of Manage-
ment Review, 25, 178 –199.
Elloy, D., & Mackie, B. (2003). Overload and work-family
conflict among Australian dual-career families: Moder-
ating effects of support. Psychological Reports, 91, 907–
913.
Farley, D. (2007, April 27). In defense of spousal hiring.
The Chronicle of Higher Education, pp. C1.
Frone, M. R., Yardley, J. K., & Markel, K. S. (1997).
Developing and testing an integrative model of the work-
family interface. Journal of Vocational Behavior, 50,
145–167.
Gilbert, L. A. (1993). Two careers/one family. Newbury
Park, CA: Sage.
Gorgievski, M. J., & Hobfoll, S. E. (2008). Work can burn
us out or fire us up: Conservation of resources in burnout
and engagement (pp. 7–22). In J. R. B. Halbesleben
(Ed.), Handbook of stress and burnout in health care.
Hauppauge, NY: Nova Science Publishers.
Greenglass, E., Fiskenbaum, L., & Burke, R. J. (1996).
Components of social support, buffering effects and
385WORK-LINKED COUPLES AND WORK–FAMILY INTEGRATION
burnout: Implications for psychological functioning.
Anxiety, Stress, & Coping, 9, 185–197.
Greenhaus, J. H., & Powell, G. N. (2006). When work and
family are allies: A theory of work-family enrichment.
Academy of Management Review, 31, 72–92.
Halbesleben, J. R. B. (2006). Sources of social support and
burnout: A meta-analytic integration. Journal of Applied
Psychology, 91, 1134 –1145.
Halbesleben, J. R. B. (2010). A meta-analysis of work
engagement: Relationships with burnout, demands, re-
sources and consequences. In A. Bakker & M. P. Leiter
(Eds.), Work engagement: A handbook of essential the-
ory and research (pp. 102–117). London, England: Rou-
tledge.
Halbesleben, J. R. B., & Bowler, W. M. (2007). Emotional
exhaustion and job performance: The mediating role of
motivation. Journal of Applied Psychology, 91, 93–106.
Halbesleben, J. R. B., & Rotondo, D. (2007). Developing
social support in employees: Human resource develop-
ment lessons from same-career couples. Advances in
Developing Human Resources, 9, 544 –555.
Halbesleben, J. R. B., & Wheeler, A. R. (2007). The costs
and benefits of working with those you love: A demand/
resource perspective on working with family. In P. L.
Perrewé and D. C. Ganster (Eds.), Exploring the work
and non-work interface. Research in occupational stress
and well being (Vol. 6, pp. 119 –169). Greenwich, CT:
Elsevier.
Halbesleben, J. R. B., & Wheeler, A. R. (2008). The relative
roles of engagement and embeddedness in predicting job
performance and intention to leave. Work & Stress, 22,
242–256.
Halbesleben, J. R. B., & Zellars, K. L. (2006). Stress and the
work-family interface. In A. M. Rossi, P. Perrewe´,&S.
Sauter (Eds.), Stress & quality of working life: Current
perspectives in occupational health (pp. 53– 68). Green-
wich, CT: Information Age Publishing.
Hall, F. S., & Hall, D. R. (1979). The two-career couple: He
works, she works, but how does the relationship work?
Reading, MA: Addison Wesley Publishing Co.
Hartmann, E. (1997). The concept of boundaries in coun-
seling and psychotherapy. British Journal of Guidance
and Counseling, 25, 147–162.
Himle, D. P. (1991). Buffering effects of four social support
types on burnout among social workers. Social Work
Research and Abstracts, 27, 22–27.
Hobfoll, S. E. (1988). The ecology of stress. New York, NY:
Hemisphere.
Hobfoll, S. E. (1989). Conservation of resources: A new
attempt at conceptualizing stress. American Psychologist,
44, 513–524.
Hobfoll, S. E. (1998). Stress, culture, and community. New
York: Plenum.
Hobfoll, S. E. (2001). The influence of culture, community,
and the nested self in the stress process: Advancing
conservation of resources theory. Applied Psychology:
An International Review, 50, 337–370.
Hobfoll, S. E., & Freedy, J. (1993). Conservation of re-
sources: A general stress theory applied to burnout. In
W. B. Schaufeli, C. Maslach, & T. Marek (Eds.), Pro-
fessional burnout: Recent developments in theory and
research (pp. 115–129). Washington, DC: Taylor &
Francis.
Jaksic, V. (2008, September 11). Law schools hiring faculty
couples. The National Law Journal. Retrieved from:
http://www.law.com/jsp/nlj/PubArticleNLJ.jsp?id⫽
1202424426321&slreturn⫽1&hbxlogin⫽1
Janning, M. (1999). A conceptual framework for examining
work-family boundary permeability for professional mar-
ried co-workers. Women and Work, 1, 41–57.
Janning, M. (2006). Put yourself in my work shoes. Varia-
tions in work-related spousal support for professional
married coworkers. Journal of Family Issues, 27, 85–
109.
Janning, M. (2009). The efficacy of symbolic work-family
integration for married professionals who share paid
work–a descriptive study. Journal of Humanities and
Social Sciences, 3, 1–17.
Janning, M., & Neely, B. E. (2006). Work-family integra-
tion for professional married co-workers: An examina-
tion of cross-realm conversations. International Journal
of Sociology of the Family, 32, 79 – 86.
King, L. A., Mattimore, L. K., King, D. W., & Adams, G. A.
(1995). Family Support Inventory for Workers: A new
measure of perceived social support from family mem-
bers. Journal of Organizational Behavior, 16, 235–258.
Koniarek, J., & Dudek, B. (1996). Social support as a buffer
in the stress-burnout relationship. International Journal
of Stress Management, 3, 99 –106.
Kossek, E. E., Lautsch, B. A., & Eaton, S. C. (2005).
Flexibility enactment theory: Implications of flexibility
type, control, and boundary management for work-family
effectiveness. In E. E. Kossek & S. J. Lambert (Eds.),
Work and life integration (pp. 243–261). Mawhaw, NJ:
LEA.
Kreiner, G. E. (2006). Consequences of work-home seg-
mentation or integration: A person-environment fit per-
spective. Journal of Organizational Behavior, 27, 485–
507.
Lapierre, L. M., & Allen, T. M. (2006). Work-supportive
family, family-supportive supervision, use of organiza-
tional benefits, and problem focused coping: Implications
of work-family conflict and employee well-being. Jour-
nal of Occupational Health Psychology, 11, 169 –181.
Lee, R. T., & Ashforth, B. E. (1996). A meta-analytic
examination of the correlates of the three dimensions of
job burnout. Journal of Applied Psychology, 81, 123–
133.
Leiter, M. (1990). The impact of family resources, control
coping, and skill utilization on the development of burn-
out: A longitudinal study. Human Relations, 43, 1067–
1083.
Marks, S. R. (1977). Multiple roles and role strain: Some
notes on human energy, time and commitment. American
Sociological Review, 42, 921–936.
Marshack, K. J. (1994). Co-preneurs and dual-career cou-
ples: Are they different? Entrepreneurship, 19, 46 – 69.
Maslach, C. (1982). Burnout: The cost of caring. Engle-
wood Cliffs, NJ: Prentice Hall.
Moen, P., & Sweet, S. (2002). Two careers, one employer:
Couples working for the same corporation. Journal of
Vocational Behavior, 61, 466 – 483.
Mooney, C. Z., & Duval, R. D. (1993). Bootstrapping: A
nonparametric approach to statistical inference. New-
bury Park, CA: Sage.
Mossholder, K. W., Bennett, N., Kemery, E. R., &
Wesolowski, M. A. (1998). Relationships between bases
386 HALBESLEBEN, ZELLARS, CARLSON, PERREWE
´, AND ROTONDO
of power and work reactions: The mediational role of
procedural justice. Journal of Management, 24, 533–552.
Muller, D., Judd, C. M., & Yzerbyt, V. Y. (2005). When
moderation is mediated and mediation is moderated.
Journal of Personality and Social Psychology, 89, 852–
863.
Neal, M. B., & Hammer, L. B. (2007). Working couples
caring for children and aging parents: Effects on work
and well-being Mahwah, NJ: Erlbaum.
Nippert-Eng, C. (1996a). Home and work: Negotiating
boundaries through everyday life. Chicago, IL: Univer-
sity of Chicago Press.
Nippert-Eng, C. (1996b). Calendars and keys: The classifi-
cation of “home” and “work.” Sociological Forum, 11,
563–582.
Parasuraman, S., Greenhaus, J. H., & Granrose, C. S.
(1992). Role stressors, social support, and well-being
among two-career couples. Journal of Organizational
Behavior, 13, 339 –356.
Pedhauzer, E. J. (1982). Multiple regression in behavioral
research: Explanation and prediction (2nd ed.). New
York, NY: Hold, Rinehart, & Winston.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsa-
koff, N. P. (2003). Common method biases in behavioral
research: A critical review of the literature and recom-
mended remedies. Journal of Applied Psychology, 88,
879 –903.
Podsakoff, P. M., Todor, W. D., & Schuler, R. S. (1983).
Leader expertise as a moderator of the effects of instru-
mental and supportive leader behaviors. Journal of Man-
agement, 9, 173–185.
Rothbard, N. P. (2001). Enriching or depleting? The dy-
namics of engagement in work and family roles. Admin-
istrative Science Quarterly, 46, 655– 684.
Salanova, M., & Schaufeli, W. B. (2008). A cross-national
study of work engagement as a mediator between job
resources and proactive behavior. The International
Journal of Human Resource Management, 19(1), 116 –
131.
Saleh, K. J., Quick, J. C., Conaway, M., Sime, W. E.,
Martin, W., Hurwitz, S., & Einhorn, T. A. (2007). The
prevalence and severity of burnout among academic or-
thopedic department leaders. Journal of Bone and Joint
Surgery, 89, 896 –903.
Schaufeli, W. B., Leiter, M. P., Maslach, C., & Jackson,
S. E. (1996). The Maslach Burnout Inventory-General
Survey (MBI-GS). In C. Maslach, S. E. Jackson, & M. P.
Leiter (Eds.), Maslach Burnout Inventory Manual (3rd
ed.). (p. 21).Palo Alto, CA: Consulting Psychologists
Press.
Schiebinger, L., Henderson, A., & Gilmartin, S. (2008).
Dual-career academic couples: What universities need to
know. Stanford, CA: Clayman Institute.
Shirom, A. (2003). Job-related burnout: A review. In J. C.
Quick & L. E. Tetrick (Eds.), Handbook of occupational
health psychology (pp. 245–264). Washington, DC:
American Psychological Association.
Shrout, P. E., & Bolger, N. (2002). Mediation in experi-
mental and non-experimental studies: New procedures
and recommendations. Psychological Methods, 7, 422–
445.
Sieber, S. D. (1974). Toward a theory of role accumulation.
American Sociological Review, 39, 567–578.
Silberstein, L. R. (1992). Dual-career marriage: A system
in transition. Hillsdale, NJ: Erlbaum.
Tepper, B. J., Henle, C. A., Lambert, L. S., Giacalone,
R. A., & Duffy, M. K. (2008). Abusive supervision and
subordinate’s organizational deviance. Journal of Ap-
plied Psychology, 93, 721–732.
Thompson, B. M., Kirk, A., & Brown, D. F. (2005). Work
based support, emotional exhaustion, and spillover of
work stress to the family environment: A study of po-
licewomen. Stress and Health, 21, 199 –207.
Van Daalen, G., Willemsen, T. M., & Sanders, K. (2006).
Reducing work-family conflict through different sources
of social support. Journal of Vocational Behavior, 69,
462– 476.
Viswesvaran, C., Sanchez, J. I., & Fisher, J. (1999). The
role of social support in the process of work stress: A
meta-analysis. Journal of Vocational Behavior, 54, 314 –
334.
Voydanoff, P. (2001). Incorporating community into work
and family research: A review of basic relationships.
Human Relations, 54, 1609 –1637.
Voydanoff, P. (2008). A conceptual model of the work-
family interface. In K. Korabik, D. S. Lero, & D. L.
Whitehead (Eds.). Handbook of work-family integration:
Research, theory, and best practices (pp. 37–56). Lon-
don, England: Academic Press.
Winkel, D. E., & Clayton, R. W. (2010). Transitioning
between work and family roles as a function of boundary
flexibility and role salience. Journal of Vocational Be-
havior, 76, 336 –343.
Winnubst, J. A. M. (1993). Organizational structure, social
support, and burnout. In W. B. Schaufeli, C. Maslach, &
T. Marek (Eds.), Professional burnout: Recent develop-
ments in theory and research (pp. 151–162). Washing-
ton, DC: Taylor & Francis.
Wolkenbreit, R. (1997). In order to form a more perfect
union: Applying no-spouse rules to employees who meet
at work. Columbia Journal of Law and Social Problems,
31, 119 –165.
Wood, J. T., & Duck, S. (1995). Under-studied relation-
ships: Off the beaten track. Thousand Oaks, CA: Sage.
Wright, T. A., & Hobfoll, S. E. (2004). Commitment, psy-
chological well-being, and job performance: An exami-
nation of conservation of resources (COR) theory and job
burnout. Journal of Business and Management, 9, 389 –
406.
Zerubavel, E. (1991). The fine line: Making distinctions in
everyday life. New York, NY: Free Press.
Received February 17, 2009
Revision received April 15, 2010
Accepted May 3, 2010 y
387WORK-LINKED COUPLES AND WORK–FAMILY INTEGRATION
A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Occupational Health Psychology
This content is subject to copyright. Terms and conditions apply.