ArticlePDF Available

Abstract and Figures

Studies investigating the crossover of job stress and strain between partners have shown that job demands are transmitted from job incumbents to their partners, affecting their psychological and physical health. Based on the crossover literature and on models of job stress and the work-family interface, this study develops a comprehensive framework to integrate the literature conceptually, delineating the mechanisms that underlie the crossover process. Key constructs include job stress, life events, strain, personal attributes and interpersonal factors. The literature pertaining to each construct in the model is reviewed and summarized. Gaps in the literature are identified, recommendations for future research are proposed, and the implications for organizational theory and practice are discussed.
Content may be subject to copyright.
Stress and strain crossover
Mina Westman
ABSTRACT Studies investigating the crossover of job stress and strain between
partners have shown that job demands are transmitted from job
incumbents to their partners, affecting their psychological and physi-
cal health. Based on the crossover literature and on models of job
stress and the work–family interface, this study develops a compre-
hensive framework to integrate the literature conceptually, delineat-
ing the mechanisms that underlie the crossover process. Key
constructs include job stress, life events, strain, personal attributes
and interpersonal factors.The literature pertaining to each construct
in the model is reviewed and summarized.Gaps in the literature are
identified, recommendations for future research are proposed, and
the implications for organizational theory and practice are discussed.
KEYWORDS crossover mechanisms
job stress
psychological strain
There is ample evidence that job stress has an impact on workers’ mental and
physical well-being (Kahn & Byosiere, 1992). However, less attention has
been paid to crossover: the reaction of individuals to the job stress experienced
by those with whom they interact regularly. In the present article, we follow
the distinction of Bolger et al. (1989) between two situations in which stress
is carried over: spillover and crossover. In spillover, stress experienced in one
domain of life results in stress in the other domain for the same individual; in
crossover, stress experienced in the workplace by the individual leads to stress
being experienced by the individual’s spouse at home. We broaden Bolger et
al.’s definition of crossover to include stress and strain experienced by the indi-
Human Relations
Volume 54(6):717–751:017470
Copyright © 2001
The Tavistock Institute ®
SAGE Publications
London, Thousand Oaks CA,
New Delhi
02westman (ds) 26/4/01 9:23 am Page 717
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
vidual at home leading to stress and strain experienced by the spouse in the
workplace. Whereas spillover is an intra-individual transmission of stress,
crossover is a dyadic, inter-individual transmission of stress or strain.
Our objectives are threefold: to integrate prior research and stimulate
additional inquiry into the phenomenon of crossover; to propose a compre-
hensive theoretical framework relating to the crossover process that extends
our understanding of work and family life; and to derive testable proposi-
tions from the framework. This article integrates crossover research into a
job-stress model in two stages. First, the theoretical perspective for studying
crossover is introduced. Second, core mechanisms of the model are proposed,
giving the empirical support. Methodological and theoretical implications are
then discussed.
Current models of work–family interaction
Scholars in the work–family domain rely on models such as segmentation, com-
pensation and spillover to characterize the process by which work and family
are linked (Evans & Bartolome, 1980; Piotrkowski, 1979; Wilensky, 1960;
Zedeck, 1992). The crossover model adds another level of analysis to previous
approaches by adding the intra-individual level and the dyad as an additional
focus. According to current research, spillover occurs from home to work and
from work to home within the individual, whereas crossover occurs from an
individual at the workplace to his or her spouse at home or from the spouse at
home to the individual in the workplace. Therefore, in this article, crossover is
treated as an inter-individual dyadic process where stress and strain experi-
enced by an individual generate similar reactions in another individual.
With the accumulated findings of crossover research (Jones & Fletcher,
1993a; Long & Voges, 1987), it is reasonable to posit that variables reflect-
ing job and family demands are the main antecedents of the crossover
process. Whereas workers operate in different interdependent systems we
focus only on the work and family systems. Thus we focus on the individual’s
job and family demands that trigger this process, thereby defining the bound-
aries of the model. Other non-work domains such as social network and the
community, though important, are beyond the scope of this article but
suggest opportunities for future research.
The proposed framework, set up to guide research and formulate
hypotheses, clarifies how the crossover process is initiated and maintained.
A theoretical perspective
Several experts in the field of work–family research have criticized the ‘state
of the art’ of work–family models. Barnett (1998) and Zedeck (1992) have
Human Relations 54(6)
02westman (ds) 26/4/01 9:23 am Page 718
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
criticized their atheoretical basis, and Lambert (1990) found work–family
theories inadequately conceptualized, neither delineating well-specified
causal models nor capturing all processes that link work and home.
Using a system theory, Bronfenbrenner (1977) asserted that no sys-
tematic research on work–family incorporated the dynamic, interpersonal,
social system perspectives. He made the point that components within the
system tend to interrelate and affect each other, which means that processes
operating in different settings are not independent of each other. Neverthe-
less, most researchers in the work–family arena focus on individual out-
comes. The proposed model starts with the individual level and moves to the
dyadic level. Examining the crossover effects of stress and strain using the
dyad as the unit of analysis may increase our understanding of the complex-
ities of multiple roles in different domains.
The proposed model integrates crossover research into a job-stress
model and anchors it in role theory (Kahn et al., 1964). The usefulness of
role theory as a basis for crossover research is that it underscores the inter-
relations between a focal person and his or her role senders in the work set-
tings and in other settings where the individual functions. Role theory is a
sound basis for crossover research, as it relates both to the person and to his
or her role senders, thus encompassing spouses and co-workers and the inter-
action between them and because it focuses on a wider role stress paradigm
than the work–family interface models.
Kahn et al. (1964) defined ‘role’ as a set of expectations applied by the
role senders to the incumbent within and beyond the organizational bound-
aries. The expectations can result in perceived role conflict, role overload and
role ambiguity. As can be seen from the review
(Table 1), role stressors are
the main antecedents of crossover.
According to role theory, work and family settings are involved in
elaborate interchanges over time with their social environments. This con-
ceptualization allows us to view family members as intimately connected to
others in the workplace and vice versa. The role episode model depicts the
interpersonal process between the focal person and the role senders that
occurs over time, incorporating interpersonal and personal factors. These
factors may affect the role episode by influencing the focal person, the role
senders and the relationship between them. The model delineated in Figure
1 uses role theory as an anchor for theoretical development in the
work–family domain. The model can guide research to determine how
experiences and processes in the work and family domain are linked and how
the crossover effect is likely to vary for workers characterized by different
personality attributes and by the interaction between partners.
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 719
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Human Relations 54(6)
Table 1 Summary of characteristics and findings of crossover studies
Study/sample Crossover of
Direction of Mediating Design/data Findings
Barnett et al.(1995), Distress H to W and Commitment Longitudinal, three waves Increase in distress over
210 dual-earner W to H data collection time of one partner was
couples mirrored in the changes
in distress of the other.
Beehr et al.(1995), Coping strategies to Police officers Cross-sectional, vignettes Police officers’ coping
177 male and spouses’ well-being to spouses strategies were positively
female police officers related to spouses’ well-being.
and their spouses
Bolger et al.(1989), Job stress to H to W Longitudinal, daily diaries One spouse’s job stress
166 married couples home stress affected the other
spouse's stress at home.
Booth (1977), 560 W’s occupational W to H Tenure, division Cross-sectional W’s employment had no
couples status on H’s marital of labor impact on H’s well-being.
Burke et al.(1980), Occupational H to W W’s life demands Cross-sectional H’s job demands were
85 senior demands to strain W’s received related to W’s
administrators of social support dissatisfaction, decreased
correctional institutes social participation and
and spouses increased psychosomatic
02westman (ds) 26/4/01 9:23 am Page 720
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Westman Stress and strain crossover
Table 1 (cont.)
Study/sample Crossover of
Direction of Mediating Design/data Findings
Chan & Margolin Work mood and affect H to W Two consecutive work W’s negative work mood
(1994), 59 dual-earner at home W to H days and work fatigue on H’s
couples reactions at home.W’s
home affect on partner's
subsequent work mood.
Eckenrode & Gore Significant other’s Significant others Demographic Cross-sectional Frequency of significant
(1981), 356 women life event on women’s to women variables other’s life events affected
health women’s health status and
health behavior.
Fletcher (1983), Male occupational H to W Domestic Retrospective Job risks of H affect life
information on mortality on wives' psychological expectancy of H and W.
1,088,995 cases of mortality environment
Fletcher (1988), H’s occupation stress H to W Shared domestic Retrospective W’s life expectancy and
324,822 British men on W’s life environment. cause of death are
in 556 occupations expectancy Psychological associated with the
and 35,915 wives mechanisms, social occupational mortality
class risk of H.
Greenhaus & W’s employment status W to H Time commitment, Cross-sectional W’s employment was
Parasuraman (1987), on H’s job satisfaction H’s level of positively related to
425 couples and quality of life satisfaction with H’s job satisfaction
work and marriage & quality of life.
02westman (ds) 26/4/01 9:23 am Page 721
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Human Relations 54(6)
Table 1 (cont.)
Study/sample Crossover of
Direction of Mediating Design/data Findings
Hammer et al.(1997), Work–family conflict W to H and Career priority, Cross-sectional Crossover effects of
399 couples (WFC) H to W involvement in WFC from Hs to Ws and
work and family vice versa.
Haynes et al.(1983), W’s profession – H’s W to H Longitudinal (10 years) Hs of employed Ws in
269 couples heart disease white-collar jobs were
over 3 times more likely to
develop CHD than those
married to blue-collar workers
or housewives.
Jackson & Maslach H’s job stress on W’s H to W Family Cross-sectional H’s stress affected their
(1982), 142 police dissatisfaction and interactions, anger, coping and the distress
officers and spouses distress withdrawal and dissatisfaction of W.
Jackson et al.(1985), H’s job experiences H to W Cross-sectional Emotional interference
95 plant operators to W’s physical, caused by job
and spouses affective, and behavioral experiences was related
symptoms to spouses’ quality of life and
dissatisfaction with employee’s
02westman (ds) 26/4/01 9:23 am Page 722
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Westman Stress and strain crossover
Table 1 (cont.)
Study/sample Crossover of
Direction of Mediating Design/data Findings
Jones & Fletcher Job and domestic H to W and Accuracy of Cross-sectional Spouses have accurate
(1993a), 60 working stresses to evening W to H perception of each perceptions of their
couples mood:mental well- other’s job stress partners’ jobs.
being Transmission of stress from
H to W especially H in high-
stress jobs.
Jones & Fletcher Work stress to H to W Marital Longitudinal.Diaries Daily fluctuations in
(1993b) psychological and W to H communication every evening for 3 work stresses affect both
physical health about work weeks partners.
Long & Voges (1987), H’s job stress to H to W Communication Cross-sectional H’s stress is related to
301 prison officers well-being of W W’s stress and strain.
and their wives
Mitchell et al.(1983), Negative events and H to W Cross-sectional Partners’ events and
157 couples ongoing strain to strain were related to
partners’ depression partners’ depression.
Morrison & Job stressors on H to W Longitudinal The well-being of sea-
Clements (1997), partners’ going partners fluctuates
82 naval couples well-being as a function of the
deployment and the mariners’
job characteristics.
02westman (ds) 26/4/01 9:23 am Page 723
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Human Relations 54(6)
Table 1 (cont.)
Study/sample Crossover of
Direction of Mediating Design/data Findings
Parasuraman et al. Role stressors and W to H Cross-sectional. Females’ family role
(1992), 119 couples WFC to partners’ Moderated regression stressors were negatively
family satisfaction analysis related to their spouses’ family
satisfaction. Males’ work and
family stressors and WFC did
not affect their spouses’ family
Pavett (1986), Job stress to burnout Spouse to Type A, Cross-sectional CPA’s stress affected
149 spouses of CPAs spouse social support spouses’ physical and
psychological symptoms of
Riley & Eckenrode H’s life events to W’s H to W Personal resources Cross-sectional Women with lower
(1986), 314 women distress levels of personal resources
were distressed by the life
events of their significant
Roberts & O’Keefe W’s occupational W to H Income, education Cross-sectional W’s employment status
(1981), 752 couples status on H’s stress and occupation had little or no effect on
levels H’s stress experience.
02westman (ds) 26/4/01 9:23 am Page 724
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Westman Stress and strain crossover
Table 1 (cont.)
Study/sample Crossover of
Direction of Mediating Design/data Findings
Rook et al.(1991), H’s stress to W’s H to W Marital tension, Longitudinal, telephone H’s job stress was
1383 married females psychological social support, interviews associated with W’s
(a second interview distress W’s life events distress.
with a subset of 92)
Rosenfield (1980), W’s occupational W to H Cross-sectional W’s employment was
60 married couples status to H’s distress positively related to H’s
Rosenfield (1992), W’s occupational W to H Income, share in Cross-sectional W’s employment was
60 married couples status to H’s domestic labor negatively related to H’s
well-being well-being only when W’s
employment decreased H’s
relative income and increased
their share of domestic labor.
Staines et al.(1986), W’s employment W to H H’s Cross-sectional Hs of employed Ws
408 husbands to status on H’s job and adequacy as registered lower job and
housewives, 208 life satisfaction breadwinner life satisfaction than Hs
husbands to wives of housewives.
working for pay
02westman (ds) 26/4/01 9:23 am Page 725
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Human Relations 54(6)
Table 1 (cont.)
Study/sample Crossover of
Direction of Mediating Design/data Findings
Vinokur et al.(1996), Depression H to W and Economic Longitudinal, SEM Partners' depression is
815 job-seekers and W to H hardship, affected by common
their spouses transactions stressors and non-
between couples supportive and undermining
Westman & Etzion Burnout W to H and Partner’s job Cross-sectional, SEM Symmetric crossover
(1995), 101 career H to W stress, sense of effects of burnout from
officers and their control, social Hs to Ws and vice
working wives support versa.
Westman & Vinokur Depression W to H and Social Longitudinal, SEM Direct crossover of
(1998), 232 couples H to W undermining depression.Life events had an
impact on the crossover
process.Social undermining
mediated the crossover
H = husbands;W = wives.
02westman (ds) 26/4/01 9:23 am Page 726
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Definitions of core constructs and the relationships among them
The conceptual model classifies a selected array of perceptions of stress and
strains as antecedent influences of the crossover process. Furthermore, the
model posits interpersonal variables as possible mediators of the crossover
process and personal attributes as possible moderators of this process.
The core assumption is that one’s stress or strain has an impact on
others in different settings, indicating a complex causal relationship between
stress and strain in the individual arena and between stress and strain of the
dyads. Most researchers have defined job stress in terms of negative charac-
teristics of the individual–organizational interface using stressors such as
overload, role conflict and role ambiguity (e.g. House, 1981; Kahn et al.,
1964). Job-related strains, according to Jex and Beehr (1991), are behav-
ioral, psychological or physical outcomes resulting from the experience of
stress. Most of the crossover research focuses on psychological outcomes of
Figure 1 distinguishes between five dimensions of the crossover process:
job and family stress, strain, life events, the interaction process and personal
attributes. Starting at the left, the model shows that the received role of the
individual may lead to responses – strain (boxes A & B, arrows 1a & 1b).
The experienced stress and strain of one individual may create stress and
strain directly in the other individual (arrows 2a & 2b). The model also sug-
gests that common stressors, affecting both partners (e.g. divorce, job trans-
fers), also impact on the individual’s stress and strain.
In the following paragraphs we review the literature that analyzes the
stress and strain investigated in the crossover process and that discusses the
main mechanisms operating in the process. We then suggest a few possible
mechanisms for the crossover process. Some of the linkages in the proposed
model have been very well supported empirically, others require further
support, and several of the variables have not been investigated in crossover
research. Therefore, each section closes with a proposition that was formu-
lated with the idea of promoting and directing future research. The model
represents direct effects in boxes A, B and C, and mediating and moderating
variables in boxes D and E. The main focus of the article is to suggest hypoth-
eses whereby these operate, based on the reviewed literature in the succeed-
ing chapters.
Stress and strain (boxes A & B)
The current article is based on empirical studies that investigated the
crossover process from different angles. Some have focused on the crossover
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 727
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Human Relations 54(6)
Figure 1 A crossover model
Note: Arrows 2a & 2b relate to all aspects of Individual A’s stressors and strains affecting all of Individual B.
02westman (ds) 26/4/01 9:23 am Page 728
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
of job stress from the individual to the spouse (e.g. Burke et al., 1980), some
have examined the process whereby job stress of the individual affects the
strain of the spouse (Jones & Fletcher, 1993a; Long & Voges, 1987) and
others have studied how psychological strain of one partner affects the strain
of the other (Mitchell et al., 1983; Westman & Etzion, 1995). Figure 1 pre-
sents six major outcomes that have been investigated in crossover research:
mood, burnout, depression, dissatisfaction, work–family conflict (WFC) and
physical illness.
Mediating and moderating variables (box D)
Whereas research has identified several antecedent influences on the
crossover process, the psychological process mediating these linkages needs
to be explored. Detecting the main underlying mechanisms of the crossover
process is a basis for developing a systematic theoretical and empirical
approach in this domain. Therefore, the suggested model includes a group of
mediators. This means that the job stress and accompanying strain of Person
A may have an effect on the stress and strain of Person B through their effects
on interpersonal factors such as coping strategies and social undermining.
For reasons of parsimony, some of the model’s recursive processes involving
feedback loops have been omitted from Figure 1.
Hypothetical mechanisms for the crossover process
The crossover literature does not reveal a systematic theoretical and empiri-
cal approach that distinguishes between the possible explanations of crossover
effects. Few researchers consider how employees’ job demands may be trans-
lated into poor well-being of the spouses and how one partner’s strain affects
the other partner’s strain (arrows 2a & 2b in the model). Without an under-
standing of these mechanisms, a model that accurately captures the complex-
ities of the work–family nexus cannot be specified and tested. A better
understanding of these processes will enable identifying effective strategies for
coping with stress crossover. The literature suggests three mechanisms which
have been incorporated into the model: a direct process (arrows 2a & 2b)
which may for example operate via empathy, a spurious effect caused by
common stressors (arrows 3a & 3b) or an indirect effect (arrows 4 & 5).
Direct empathetic crossover (arrows 2a & 2b)
The first view expressed in the model is that direct crossover of stress occurs
from one partner to the other. The basis for this view is the finding that
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 729
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
crossover effects appear between closely related partners who care for each
other and share a great part of their lives together. Accordingly, strain in one
partner produces an empathetic reaction in the other that increases his or her
level of strain. This view is supported by social learning theorists (e.g.
Bandura, 1969; Stotland, 1969) who have explained the transmission of
emotions as a conscious processing of information. They suggest that indi-
viduals imagine how they would feel in the position of another and thus come
to experience and share their feelings.
Eckenrode and Gore (1981) demonstrated that husbands’ life events
increased the amount of stress their wives experienced. These findings show
that whereas life events research considers stress as an individual level vari-
able, one partner’s stressful events can affect the other. On the basis of their
findings they argued that stressful life events influence the entire network of
close ties. Similarly, Riley and Eckenrode (1986) suggested that the effect of
the undesirable events one experiences on the significant other’s distress may
be the result of empathy expressed in reports like ‘We feel their pain is our
own’ (p. 771).
However, in order to establish and support the ‘direct-empathy’ expla-
nation for the crossover process, we suggest adding measures of empathy to
crossover research. This will support the attribution of a direct crossover
effect to empathy and rule out the possibility that some other processes are
The literature shows that crossover may be unidirectional (arrows 2a
& 2b) or bi-directional (arrows 2a & 2b). Most stress crossover studies have
been unidirectional and found effects of husbands’ job stress on the well-
being of their wives (Burke et al., 1980; Jackson & Maslach, 1982; Long &
Voges, 1987; Pavett, 1986; Rook et al., 1991). These studies related to the
wives as the passive recipients of stress and strain from their husbands,
neither assessing nor controlling wives’ job and life stress, and in some cases
had mixed samples of working and non-working wives. Therefore, we cannot
rule out the possibility that what appears to be direct crossover of stress from
husbands to wives is an outcome of the wife’s job.
Despite the significant increase in the number of women who have
joined the work force at all job levels, no published research has focused exclu-
sively on the crossover of stress from wives to husbands. Some researchers
have investigated the impact of women’s employment, but not their job stress,
and found negative effects of wives’ employment status on husbands’ strain
(e.g. Booth, 1977; Burke & Weir, 1976; Greenhaus & Parasuraman, 1987;
Haynes et al., 1983; Roberts & O’Keefe, 1981; Rosenfield, 1980, 1992;
Staines et al., 1986). These studies did not specify which element of the wives’
Human Relations 54(6)
02westman (ds) 26/4/01 9:23 am Page 730
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
employment caused crossover, nor did they eliminate the possibility that it is
their own job stress that causes their strain.
Only five studies have investigated both spouses and examined bi-
directional crossover of stress or strain. Jones and Fletcher (1993a) found
transmission of husbands’ job demands on wives’ anxiety and depression
after controlling wives’ job stress. However, they did not find such an effect
from wives to husbands, perhaps because the women in their sample did not
experience high levels of stress. Yet all other studies that looked into bi-
directional crossover found symmetric crossover effects between partners.
Westman and Etzion (1995) demonstrated a crossover of burnout from
career officers to their spouses and vice versa, after controlling husband’s and
wife’s own job stress. Hammer et al. (1997) also found a bi-directional
crossover of work family conflict from husbands to wives and vice versa.
Whereas these were cross-sectional studies, the bi-directional nature of the
crossover effect has also been demonstrated in studies using longitudinal
designs. Barnett et al. (1995) and Westman and Vinokur (1998) found bi-
directional crossover of distress from husbands to wives and from wives to
husbands. The first proposition is based on these findings.
Proposition 1: There is a bi-directional direct crossover of stress and
strain between couples (arrows 2a & 2b).
Spuriousness – common stressors (circle C)
The common stressors mechanism suggested by Westman and Vinokur
(1998) is that the relationship between spouses’ strain is spurious. What
appears to be a crossover effect is the result of common stressors in a shared
environment increasing the strain in both partners (arrows 3a & 3b). Hobfoll
and London (1986) suggested that many stressors make simultaneous
demands on both individuals in a dyad. Burke et al. (1980) and Westman
and Etzion (1995) suggested that crossover effects might be the result of
common stressors, though they did not incorporate a measure of common
stressor in their studies.
One common stressor that may affect both partners is stressful life
events. Unfortunately, most crossover studies focus on job stress and do not
include life events as possible common stressors that might affect both
spouses. A near exception is Rook et al. (1991), who used husbands’ life
events as their measure of stress, though these stressful events were reported
by their wives. The validity of such a measure is questionable, though it is
an indication of the way wives perceive their husbands’ life events and how
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 731
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
it affects them. Morrison and Clements (1997) used life events as a control
for non-job-related stress but did not investigate its role in the crossover
process. Only two crossover studies measured common stressors. Vinokur
et al. (1996) found that family financial strain had an impact on the
crossover process and on the supportive and undermining transactions
between the partners, and Westman and Vinokur (1998) found that
common life events affected the crossover process by increasing each
partner’s depression (arrows 3a & 3b). The common stressors may also
influence the coping strategies and interaction process between the partners
(arrow 9).
Proposition 2: Common stressors increase the strain of both partners
(arrows 3a & 3b).
Proposition 2a: Common stressors have an impact on interpersonal
factors (arrow 9). The higher the intensity of common stressors the
greater the effect of social support and effective coping strategies.
Crossover as an indirect process (circle D)
The explanation of crossover as an indirect process posits a mediating and
moderating effect. We focus on coping mechanisms, social support and
undermining, which have been widely identified as mediators of stress and
strain responses (Coyne & Downey, 1991; Pearlin, 1989).
Coping strategies represent efforts to prevent or reduce the negative
effects of stress. Numerous studies suggest that problem-focused strategies
are positively related to well-being, whereas emotion-focused ones are nega-
tively related to well-being (Westman & Shirom, 1995). Most of these studies
examined the direct effect of partners’ coping on their partners’ well-being
or the moderating effects of coping on the relationship between stress and
strain (Edwards et al., 1990). However, a few studies have examined the
relationship between one partner’s coping and the other partner’s well-being.
Partners’ strain can affect their own as well as their partners’ coping strat-
egies. Therefore, coping can be viewed as a predictor of both partners’ strain
(arrows 5a & 5b) (Beehr et al., 1995) or as a mediator (arrows 4a/5b &
4b/5a) of the relationship between one partner’s stress and the spouse’s stress
or strain (Jackson & Maslach, 1982).
Researchers have provided two main directions of the relationships
between one’s coping and the partner’s well-being. Monnier and Hobfoll
(1997) demonstrated the impact of partners’ coping on their spouses’ well-
being. They found that active coping by respondents was negatively related
to their partners’ depression and that active, prosocial coping was inversely
Human Relations 54(6)
02westman (ds) 26/4/01 9:23 am Page 732
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
related to future depression, demonstrating that it may have some long-term
ameliorative crossover effects. The other view, proposed by Burke et al.
(1980) and Kahn et al. (1985), is that partners’ strain may determine their
spouses’ coping strategies (arrows 4a & 4b). Burke et al. (1980) found that
wives whose husbands reported high job demands used more emotion-
focused coping strategies, such as distraction, explosive outbursts and talking
to others. However, they used problem-focused coping strategies when their
husbands reported lower job demands. Kahn et al.’s (1985) findings that
spouses of depressed persons used more aggressive strategies support this
direction. The results supporting the second perspective may imply that
women who use less constructive coping experience greater crossover of
stress from their husbands than women using problem-focused strategies.
The possibility that one partner’s stress may exhaust his or her partner’s
coping capacity, thereby increasing the partner’s vulnerability to stress, has
not been investigated.
The two main views of whether depression affects the partner’s
coping or whether coping affects his or her depression are not contradic-
tory and may even be reciprocal, suggesting a feedback loop. The process
may commence at either point, with one partner’s coping or depression.
The important issue is the spiral that starts where one role sender’s state
and actions affect the partner and vice versa. Only experimentation and
longitudinal studies can indicate the initiation of the process. Another area
that merits further research is the impact of coping strategy on the focal
respondent’s partner, that is, whether certain patterns of the partners’
coping strategies attenuate or strengthen the crossover process. Certain
coping strategies may simultaneously enhance the well-being of the coper
and result in worse outcomes for partners. To illustrate this point, with-
drawal from stressful episodes may be functional for the individual adopt-
ing such a strategy, but it may involve delegating communal responsibilities
to a partner who may not be prepared to assume the added burdens. In the
long run, such coping strategies may also result in negative psychological
outcomes for the coper. The possibility that problem-focused coping may
negatively affect those close to the focal respondent has been considered
comparatively recently (Hobfoll et al., 1994). Therefore, it is important to
investigate how the interaction process affects both of the partners. It will
require longitudinal designs to detect changes in coping and their effects
over time.
Proposition 3: Coping strategies mediate the crossover process. One’s
stress and strain affect one’s own coping strategies and the coping
strategies of the partner (arrows 4a & 4b), and consequently, the
partner’s stress and strain (5a & 5b). Furthermore, one partner’s
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 733
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
coping strategies affect one’s stress and strain and the stress and strain
of the partner.
Communication characteristics are what people say to each other or
how they react to events they experience. The literature supporting this expla-
nation implies the need to focus on the couple’s communication pattern in
terms of the kinds of interaction likely to enhance the partners’ experience
of stress or strain. Several investigators have addressed the nature of infor-
mation communicated between the partners as mediating the crossover
process. Jones and Fletcher (1993b) suggested that communication may
mediate the relationship between partners’ mood. They found that the
woman’s mood was affected by her partner’s communication pattern: it was
more positive when her husband offloaded worries and frustrations, but
more negative when he became withdrawn or distracted.
Accurate knowledge of the spouse’s job stress may be an important
mechanism in the crossover process. The amount and kind of discussion are
not a necessary condition for crossover, as partners can learn about their
spouses’ stress and strain from other venues. Pearlin and Turner (1987) found
that many interviewees tried to segregate stress arising in the workplace from
the family domain, explaining their distress would only anguish the spouse;
the spouse would blame them for the problems; the spouse would give them
unwanted or inappropriate advice. Whatever the reason, people rarely suc-
ceeded in barricading distress generated at work from the family domain.
Spouses indicated they could tell when their partners were stressed, regard-
less of whether the partner referred to it. Mood changes, shifts in activities
and other clues aroused awareness to the distress of their partners, even when
they did not know the reason for it. Attempts to screen stresses from the
family and the uncertainty created by such attempts were additional sources
of strain. These findings show that attempts at segmentation are rarely suc-
cessful and may even intensify the crossover process.
The social interaction process
Social support refers to transactions with others that provide emotional and
instrumental support, appraisal and information (House, 1981). Antonovsky
(1979) found that inadequate social support in stressful situations may
increase vulnerability to distress. Therefore, an interaction style that does not
provide enough support, or that demands support from a spouse unable to
provide such support, can produce a crossover effect and also affect the
crossover process. Jones and Fletcher (1993a) suggested that lack of social
support might lead to a greater tendency to transmit stress to the other,
Human Relations 54(6)
02westman (ds) 26/4/01 9:23 am Page 734
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
though they did not investigate social support in this context. Repetti (1989)
found that spouse’s support moderated the association between workload
and marital withdrawal. Higher levels of withdrawal were followed by an
increase in social withdrawal and less expression of anger in evenings when
wives were supportive. These associations were not observed on relatively
low supportive evenings.
According to Hobfoll et al. (1994), many researchers have demon-
strated the beneficial effects of social support, focusing on the recipient and
overlooking its impact on the donor. The role of the donor in the interaction
may be one of the keys to the crossover process. Kessler and McLeod (1984)
found that caring for a wide network of people was an additional burden for
women, and this support translated into transmission of stress from their
network to themselves. Similarly, Riley and Ekenrode (1986) noted that
couples are influenced by each other’s distress indirectly, via the other’s
reduced social support, noting that demand for social support caused a drain
in others in the dyad or in the social group. They underscored two processes:
the transaction of support whereby individuals share their resources with the
needy; a diminishing of resources experienced by the providers of social
support both by sharing them and by empathetically experiencing the
demands of the needy. These findings indicate that stress experienced by one
partner creates demands for support (arrow 4a & 4b) and, when unable to
meet these expectations, the other is apt to feel anxious or guilty. Conversely,
a crisis experienced by a close partner may diminish the social support avail-
able to the individual. Similarly, Hobfoll and London (1986), studying Israeli
women whose relatives were mobilized to serve during wartime, found that
social support aided the recipients but depleted the resources of the donors
at a time when they too needed their stress resistance resources. The indi-
cations are that social support is a finite resource and that people compete
for it in a zero-sum game. Therefore, the expected donor experiences strain
either because of his or her inability to provide support or because of his or
her depleted resources.
Proposition 4: Job stress and strain influence social support (arrows 4a
& 4b) by requiring social support from the donors, thus depleting their
resources and enhancing their stress and strain (arrows 5a & 5b).
Social undermining is referred to in the literature as social hindrance,
social conflict and negative social support. According to Vinokur and van
Ryn (1993), social undermining consists of behaviors directed toward the
target person that express negative affect, convey negative evaluation or criti-
cism, or hinder the attainment of instrumental goals. Researchers have shown
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 735
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
that social conflict is symptomatic of the stress and strain of cohabiting part-
ners (Abbey et al., 1985; Kahn et al., 1985; Vinokur & van Ryn, 1993). The
explanation that the crossover process is mediated by negative social inter-
actions is supported by empirical findings from two lines of research. First,
research documents that frustration is often an outcome of stressful con-
ditions that trigger aggression (Berkowitz, 1989). Second, the literature on
family processes also reports that stressed couples exhibit high levels of nega-
tive conflictual interactions (Schaefer et al., 1981).
There are two main views on the relationship between social under-
mining and depression concerning the initiation of the crossover process.
Some researchers (Coyne & Downey, 1991; Russell & Cutrona, 1991;
Vinokur & van Ryn, 1993) have shown that social undermining increases the
stress and strain of partners (arrows 5a & 5b). Vinokur and van Ryn (1993)
found, in a longitudinal study, that increase in social undermining produced
a change for the worse in respondents’ mental health. In the same vein, Shinn
et al. (1984) suggested conceptualizing negative interpersonal relationships
as a stressor leading to strain.
There are also advocates, however, of the inverse view (arrows 4a &
4b) that depression precedes conflicting interactions (Nelson & Beach, 1990;
Schmaling & Jacobson, 1990). For example, MacEwen et al. (1992) found
that overload resulted in anxiety and depression, which in turn resulted in
more negative marital interactions. Similarly, Crouter et al. (1989) found that
high levels of stress at work were related to increased negative marital inter-
actions at home. The cross-sectional design of most of the studies precludes
concluding whether social conflict is an antecedent or consequence of
changes in well-being. To illustrate, Abbey et al. (1985) found positive cor-
relations between social conflict and anxiety and depression among students.
They concluded: ‘The greater the amount of social conflict respondents
reported receiving from the person closest to them, the greater their depres-
sion and anxiety’ (p. 119). However, they noted in their discussion that,
though for theoretical reasons they interpret these relations in terms of causal
pathways, neither their research design nor research evidence preclude the
rival hypothesis: the greater the depression and anxiety, the higher the degree
of social conflict. This causal inconclusiveness is typical of crossover research.
These seemingly contradictory findings lead us to propose that the
crossover of stress and strain is mediated by the negative interaction between
partners. Westman and Vinokur (1998) supported the mediating role of
undermining in the crossover process. They showed that the correlation in
depression symptoms in couples was due primarily to crossover via negative
social interaction. Women’s depression at both waves of their longitudinal
study increased their undermining behaviors toward their husbands (arrow
Human Relations 54(6)
02westman (ds) 26/4/01 9:23 am Page 736
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
4a), which in turn increased the husbands’ depressive symptoms (arrow 5b).
In sum, some studies have found that depression precedes conflictual inter-
actions and others have found that conflictual interactions precede depres-
sion. Though we find more theoretical justification for the hypothesis that
undermining leads to the other person’s depression, there is also theoretical
justification for the rival hypothesis. It is our conclusion that social under-
mining increases the partner’s depression, which in turn, tends to foster an
environment of social undermining. Thus, we suggest both directions of influ-
ence between depression and undermining, i.e. recursive relationships of
social undermining acting as an antecedent of strain for one person and as a
consequence of depression for the other. Feedback loops may occur between
undermining and depression, triggering a vicious circle in which one ampli-
fies the other.
Proposition 5: Social undermining mediates the process of crossover
from one spouse to another.
Proposition 5a: The strain of one spouse increases the process of social
undermining between the partners. As one (Individual A) becomes
more depressed he or she is more inclined to display undermining
behaviors toward the partner (arrows 4a).
Proposition 5b: In turn, this undermining behavior of Individual A ele-
vates the strain of the other partner (Individual B; arrow 5b). Conse-
quently, Individual B’s strain increases social undermining behaviors
toward Individual A (arrow 4a).
Personal attributes as mediating and moderating variables
(circle E)
There is no systematic empirical evidence on the contribution of personal
attributes to the crossover process. They can influence stress and strain
directly as exogenous variables (arrows 6a & 6b) and interactively (arrows
8a & 8b) as moderators that interact with Individual As job stress and strain
in explaining Individual B’s stress and strain. Personal attributes thus affect
susceptibility to both partners’ stress and strain. Some crossover researchers
have considered demographic characteristics. However, the conceptual links
to crossover are weak and the causality is ambiguous. The most frequently
investigated characteristics are gender, life stage and number of children, and
they have generally been weakly related to crossover.
One personality characteristic found to contribute to strain in the
crossover process is sense of control (Westman & Etzion, 1995; Westman
& Vinokur, 1998). Researchers have rarely looked into other personality
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 737
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
attributes that might help explain crossover. This lack is noteworthy,
considering the importance of personality in stress research (Spector &
O’Connell, 1994). Future crossover research should include personal attrib-
utes such as the Big Five trait dimensions (Digman, 1990), Type A Behav-
ior Pattern (Edwards & Baglioni, 1991) and workaholism, which would
help to fill an important gap in crossover research.
A personal trait relevant to crossover research is negative affectivity
(NA), defined by Watson and Clark (1984) as a stable tendency to experi-
ence negative emotions across time and situations. Though many studies have
concluded that NA may be associated with over-reporting of job stress and
strain (Burke et al., 1993), none has examined the impact of NA in explain-
ing the crossover process. Only Morrison and Clements (1997) used NA of
both spouses as a control variable. They found that the female’s NA was a
significant predictor of her reported well-being across each of the three
dependent measures they used (physical health, depression and GHQ).
However, NA affected only physical health amongst the males. The impact
of NA of one partner may affect his or her report on stress and strain, and
the ‘crossover’ detected may be spurious. Furthermore, couples may influ-
ence each other’s perceptions of their social environment. What we consider
stress crossover may be crossover of NA. By not controlling NA, crossover
research may have overestimated the effect of the individual’s job stress on
the spouse’s well-being.
Proposition 6: Personal attributes (e.g. gender, sense of control) affect
the crossover process by their direct effect on the stress and strain of
the partners (arrows 6a & 6b) and by moderating the intensity of the
crossover process (arrows 8a & 8b).
Several processes operate in conjunction. Another possibility is that
more than one explanation is appropriate to explain the crossover process
and that some of the proposed explanations operate in conjunction. Vinokur
et al. (1996), for example, found that financial strain, representing common
stressors, increased symptoms of depression in the job seeker and in the
partner. These depressive symptoms increased the partners’ undermining
behavior, which increased depressive symptoms in the job seeker, indicating
an indirect effect. Westman and Vinokur (1998) found direct crossover of
depression between spouses. In addition, social undermining was a strong
mediating factor, indicating an indirect effect of crossover. Finally, life events
representing common stressors were found to affect the depression of both
partners adversely, which increased social undermining and further increased
depression. These findings highlight the possibility that several mechanisms
may concurrently explain the crossover process.
Human Relations 54(6)
02westman (ds) 26/4/01 9:23 am Page 738
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Proposition 7: Crossover may be the result of several processes (such
as direct effect of one partner’s stress and strain on the other partner’s
stress and strain, common stressors experienced by both partners that
arouse strain in both, and indirect process) that are not mutually
The search should go on for other potential mediators that might shed
light on the mechanisms through which the direct crossover process works.
New topics for crossover research and propositions
Crossover in the workplace
As previous crossover research was based on the work–family interface,
researchers focused only on the family as the ‘victim’ of the job incumbent’s
stress. However, when the crossover framework is based on role theory we
can broaden the scope of research and investigate the crossover of stress from
home to the workplace and among role senders in the work setting. As the
simplest level of crossover is the dyadic process that transpires between two
individuals, the same principle can be applied to interactions in other
organizations. What happens to one member of a dyad, whether a family
member or a work-group member, affects the other. This approach is con-
sistent with Moos’s (1984) theory that people are part of a social system and
have to be understood as such. The family is one such system and the work
group is another. Each member is linked to other members, and change in
one will affect change in others. Whereas the conditions that people experi-
ence in the workplace can have pronounced effects on the stress transmission
within the family, the conditions experienced in the family may cross over to
others in the workplace. As Kanter (1977) argued, ‘if emotional climate at
work can affect families, so can the family’s emotional climate and demands
affect members as workers’ (p. 56). Furthermore, individuals in the work
team who share the same environment may start a crossover chain of stress
and strain among themselves, whether the source of stress is in the family or
at the workplace.
Bolger et al. (1989) suggested that conflicts with people in ongoing,
non-family relationships might be particularly distressing, because they have
continuity and usually lack sufficient intimacy and understanding to prevent
arguments from being perceived as a major threat. In an exploratory investi-
gation of the work domain, Westman and Etzion (1999) found crossover of
work-induced strain from school principals to teachers and vice versa, after
controlling their job stress. If these findings are replicated, and stress and
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 739
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
strain in the workplace are contagious, this might lead to burnout or a
depressive climate in organizations. Further studies should investigate the
impact of supervisors’ stress and strain on their subordinates’ strain, the
impact of colleagues’ stress and strain on one another, in an effort to detect
and identify distressed teams and workplaces. The new trends in structuring
work, including the increase in the use of team-based production and greater
interdependency, will only increase the possibility and frequency of crossover,
thus creating a ‘strain climate’. Researchers and management have to detect
the processes leading to this phenomenon and suggest ways to prevent and
manage it at individual, dyad and team levels.
The proposed model delineates the feedback process between focal
persons in various role sets. The reciprocal relationship between the work
and family settings can proceed recursively from the work domain to the
family domain or from the family domain to the work domain. After its initi-
ation, the crossover process unfolds as a spiral in various directions, dictated
by the setting and the other role players.
High-stress occupations and situations
Some crossover researchers have focused on husbands with stressful jobs
such as prison officers and career officers (Burke et al., 1980; Jackson &
Maslach, 1982; Long & Voges, 1987; Pavett, 1986; Westman & Etzion,
1995). Others have investigated highly stressed respondents. Jones and
Fletcher (1993b), for example, found a stronger bivariate relationship
between husbands’ job stress and wives’ well-being only in a subsample of
husbands reporting higher levels of stress. The small size of this subsample
and the similarity between spouses’ mood, which may be either a cause or a
result of crossover, call for replication. Furthermore, Hatfield et al. (1994)
indicate that the more intense emotions are, the greater their influence on the
emotions of the spouse. These findings indicate that stressful occupations and
stressful demands can potentially have a dual impact upon the individual and
his or her family, of which organizations should be aware because of the high
price it exacts from the individuals. We need comparison of crossover in
samples of stressful and less stressful occupations. Furthermore, we need to
investigate crossover in the same sample under stressful and non-stressful
conditions to determine whether fluctuations in stress affect the crossover
process and whether we can detect tranquility transmission. It is also import-
ant to compare couples having one partner reporting high stress to couples
having one partner reporting low stress, in order to investigate whether there
is a critical point at which stress crossover starts. Such research should assess
the job stress and resources of both partners, and have a design that enables
Human Relations 54(6)
02westman (ds) 26/4/01 9:23 am Page 740
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
investigation of the variation in spousal well-being as a function of differen-
tial stress. Identification of potentially stressful events or stressful occu-
pations that enhance the crossover process will help management in
formulating preventive measures. As there is some evidence that crossover
does not occur in low-stress situations we suggest a non-linear relationship
between stress of one spouse and the crossover process.
Proposition 8: There is a step function relationship between stress and
crossover, resulting in a threshold level of stress necessary to trigger
crossover. Crossover of stress and strain is stronger during high-stress
periods than during medium- and low-stress periods. It is also stronger
in high-stress occupations than in medium- and low-stress occu-
Gender differences in the crossover processes
Gender is a potential moderator of the impact of one partner’s stress on the
other partner’s strain because of differences in the traditional role demands
and expectations for men and women (Lambert, 1990). There is some indi-
cation that women are more susceptible than men to the impact of stressors
affecting their partners (Kessler, 1979). Kessler and McLeod (1984) showed
that events happening to spouses are more distressing for women than for
men. They suggested that because of their greater involvement in family
affairs, women become more sensitive not only to the stressful events that
they experience but also to those that affect other family members. Further-
more, research on social support has increasingly characterized support
seeking, giving and utilization as processes that involve men and women dif-
ferently (Etzion, 1984).
There is also evidence that people differ in the way they can read others.
Haviland and Malatesta (1981) found that women were more vulnerable to
emotional transmission than men. Conger et al. (1993), highlighting another
angle, found different responses to financial stress in the form of greater hos-
tility among men. This increasing hostility has the potential for undermining
the spouse, thereby increasing depressive symptoms. Although the notion
that women may be more vulnerable to stress than men is not well estab-
lished, it does appear that the relevance of the family as a direct source of
stress is stronger for women than for men.
Evidence concerning gender differences in crossover effects is mixed.
Six of the reviewed articles related to bi-directional crossover. Of these, four
studies found a bi-directional crossover effect of similar magnitude from hus-
bands to wives and from wives to husbands (Barnett et al., 1995; Hammer
et al., 1997; Westman & Etzion, 1995; Westman & Vinokur, 1998).
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 741
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
However, Parasuraman et al. (1992), investigating dual-career couples, found
that whereas men’s work and family stressors and WFC did not affect their
wives’ family satisfaction, women’s family role stressors had a significant
negative relationship with their husbands’ family satisfaction. Contrary to
these findings, Jones and Fletcher (1993a) reported a crossover effect of men’s
job demands on women’s psychological health but found no effect of
women’s job demands on men’s psychological health. One explanation for
the contradictory findings may be the wives’ occupations. The wives in Jones
and Fletcher’s original study did not report high levels of job stress. However,
when they analyzed a subsample characterized by high stress they did find a
bi-directional crossover effect.
Morrison and Clements (1997) investigated the impact of husbands’
job characteristics on spouse’s well-being. Though both spouses completed
questionnaires, the study investigated only the crossover from husbands to
wives, while controlling for wives’ stress. Interestingly, these researchers
found that wives’ stress had a lower impact on their health than did their
husbands’ job characteristic. A possible explanation is that the wives’ jobs
were less stressful, making this finding similar to that of Jones and Fletcher
(1993b), who found stronger relationships between male partners’ job stress
and female partners’ well-being when examining men in more stressful occu-
pations. Our review of the literature on gender differences in the crossover
of stress and strain leads us to the following proposition.
Proposition 9: There are gender differences in the direction and magni-
tude of the crossover process in dual-earner families. Females are more
susceptible to the crossover process as recipients of strain than males.
Positive crossover
All the reviewed studies investigated negative crossover such as when job
stress of one spouse affects the stress or strain of the other spouse. However,
just as stressful demands or a bad day at work have a negative impact on
partner’s well-being, positive job events may also cross over to the partner
and have a positive effect on his or her well-being. Whereas crossover is
usually defined as a transmission of stress, we can broaden the definition into
transmission of positive events or feelings as well. Positive experiences and
feelings are not merely the absence of stress but qualitatively different experi-
One possible reason for the neglect of the investigation of positive
crossover is that stress research relies heavily on medical models, with their
emphasis on negative effects just as negative affectivity was investigated for
Human Relations 54(6)
02westman (ds) 26/4/01 9:23 am Page 742
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
many years before researchers broadened their interest to positive affectivity.
One can think of many instances of positive crossover such as enjoyable
experiences at one’s job leading to job satisfaction crossover and eliciting a
good mood in the partner at home. Similarly, a person whose work group is
in a good mood because of getting a bonus or recognition for a job well done
may transmit this good mood to the spouse. Conversely, family life can
support, facilitate or enhance work life. Supportive family relationships and
attitudes can create positive crossover to the work set-up. Investigating posi-
tive crossover can enhance theoretical thinking and make practical contri-
butions to the crossover literature. In the work arena, management should
be aware that their positive actions can result in additional positive outcomes
that they had not originally planned. Just as crossover at the workplace can
cause a burnout climate in the organization, we can think of positive
crossover where positive experiences impact the team, the department and
the organization. Possible antecedents for such a climate may be frequent
positive events such as vacation bonuses, specific changes in work schedules,
Proposition 10: There is positive crossover between partners. Positive
job experiences are transmitted to the spouse at home causing positive
feelings in him or her. In the same vein, positive non-work experiences
may cross over to the partner and thus create positive crossover into
the work setting. Positive experiences may buffer the stress crossover
process between partners.
Research designs
The pioneering crossover studies were either qualitative or cross-sectional
and retrospective, demonstrating ambiguous findings so that there was a need
for further research that had been designed more rigorously. However, several
studies employed designs of daily repeated measures of stress and strain from
both husbands and wives (Bolger et al., 1989; Chan & Margolin, 1994; Jones
& Fletcher, 1993b, 1996; Repetti, 1989). These designs enabled researchers
to study the dynamics within situations that appear static in conventional
cross-sectional studies (Bolger et al., 1989).
Four studies have employed a longitudinal design with two measures
(Barnett et al., 1995; Morrison & Clements, 1997; Vinokur et al., 1996;
Westman & Vinokur, 1998). Barnett et al. (1995), in a longitudinal study of
dual-earner couples, found that changes in job experiences of one partner
affected the distress level of the other partner. Furthermore, they found that
change over time in one partner’s distress was a function of his or her own
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 743
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
job stress and of the partner’s job stress and distress. They conclude that the
correlation between marital partners’ mental health indicators is due to
shared circumstances or to reciprocal effects over time. Morrison and
Clements (1997) used a longitudinal design of navy couples with two
measurements, a month apart. They found that the well-being of partners
who remained at home fluctuated as a function of the deployment status and
the job characteristics of their partners.
Very little is known as to which variables are antecedents, consequences
or mediators in the crossover process because so little attention has been paid
to causal relationships. Because of their longitudinal design and the use of
SEM (structural equation modelling), the studies of Vinokur et al. (1996) and
Westman and Vinokur (1998) ruled out rival hypotheses concerning direc-
tion of causation and the findings are relatively internally valid. One of the
most pressing research needs is to illuminate more specifically how the
crossover process occurs. A longitudinal research strategy would seem appro-
priate to study crossover effects and causality more rigorously, by analyzing
covariation across time. Such designs are instrumental to examining how a
spouse’s well-being varies as a function of fluctuations in intensity of the job-
holder’s job stress. Examining the unfolding dynamics of stress over time will
help to complement and integrate the important insights that researchers
have already uncovered regarding crossover.
Future crossover studies should take into account investigation of the
amount of variance in partners’ distress explained by their own job demands,
by their partners’ demands, and by their dispositions. Yet another important
problem is the multiple levels of analysis that must be employed when investi-
gating the work–family interface: individuals, partners, families, teams and
organizations. The clear advantage of crossover research is that it is based
on observation of two partners. Collecting data from both partners avoids
confounding – and enables controlling – each partner’s stress. Information
about the dyad adds to our understanding of well-being in husbands and
wives above and beyond that provided by information about the individual.
Thompson and Walker (1982) point out that for research to be dyadic, the
problem must be conceptualized at the level of the relationship and the analy-
sis must be interpersonal, focusing on the responses between the two indi-
viduals. Bi-directionality is a crucial issue in the study of crossover as it
enables investigation of symmetric models. Some of the crossover researchers
collected data from both partners concerning one partner’s job stress; some
collected data from one spouse about engaging in social undermining while
the spouse related to received undermining, enabling measurement of the
amount of undermining. However, we need more measures at a higher level
of analysis such as the family level (e.g. the rate at which arguments between
Human Relations 54(6)
02westman (ds) 26/4/01 9:23 am Page 744
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
partners escalate) and at the organizational level (conflict in teams). Finally,
there is a need for triangulation of data sources concerning indicators of
Theoretical implications
The proposed model has interesting theoretical implications. A stressful event
that starts at work is transferred from the employee to the spouse and to
other employees, then from the spouse to others in the family. The literature
reviewed supports the theoretical and practical relevance of the proposed
model. Overall, the proposed model seems feasible based on the existing evi-
dence, though additional research is needed to test it more directly, particu-
larly some moderators. To investigate the crossover process thoroughly we
have to investigate three phases of relationships in the causal chain: the
relationships between the individual’s stress and strain; the work–family links
– the level of spillover for the same individual; and the crossover of one
partner’s stress to his or her partner or to groups. Each phase is a necessary
but not sufficient condition for the next, because of the mediating variables
that may intervene in each relationship. As our model and findings show,
individual differences as moderators and the social interactions as mediators
are important factors of the crossover process. Though investigating all these
relationships sequentially within one study is an ambitious undertaking, it
will shed light on the crossover process. To date, no study has employed a
longitudinal design to assess all the direct and indirect relationships among
the constructs specified by the model. A comprehensive understanding of the
crossover process will elicit ample opportunities to buffer this chain of influ-
ence, whether in the two intra-individual phases, or at the inter-individual
Though crossover is defined as movement of demands between indi-
viduals across borders, we have suggested that it can occur between indi-
viduals in the same domain (Figure 2, arrow 2); for example, when stress
generated in the workplace affects other job incumbents through a crossover
process (Westman & Etzion, 1999). There is some research on the crossover
of strain in the family domain from parents to children (Katz & Gottman,
1995), though the source of this strain, whether initiated in the work or
family domain, is not very clear. Crossover research has to focus on its effect
on the individual, the family, the work team and the organization, as illus-
trated in Figure 2.
Crossover research suffers from a lack of systematic study of the indi-
vidual and societal conditions under which one or another form of crossover
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 745
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
is more likely to emerge. A contingency theory of crossover can be developed
once a typology of mediating and moderating variables has been specified.
Meaningful prevention and intervention strategies must rest on sound theory
and evidence.
Five broad themes emerge from this article: (1) links between the work-
place and the family are exceedingly complex, requiring specification of the
multiple underlying pathways of influence; (2) identifying and testing theor-
etically based mediators and moderators of these processes remain import-
ant tasks; (3) in particular, exploring the psychological process mediating the
linkages between the antecedent’s influences and the crossover process is
crucial to further developing crossover theory and for practical implications;
(4) distinguishing between possible explanations of crossover effects is a
promising direction for theory development; and (5) resolving the two con-
flicting views on the precedence of coping and depression will advance our
understanding of crossover processes. A better understanding of the process
may lead to ways of solving this issue. It will also extend role theory, which
postulates relationship between role stress and anxiety.
Human Relations 54(6)
Figure 2 Future directions for crossover research
Note: The full solid line (1) has been investigated frequently and represents crossover from one
partner’s job to his or her partner at home.The narrow line (2) has rarely been investigated and
represents crossover from one partner’s job to others in the same job setting.The dotted lines
(3 & 4) represent crossover from one partner’s family to his or her partner at work (3) and from
one partner’s family to other members in the job setting (4). None of these kinds of crossover
has been investigated yet.
02westman (ds) 26/4/01 9:23 am Page 746
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Pursuing the challenges outlined here will go a long way toward clari-
fying stress and strain crossover. Furthermore, it will also make important
contributions to role theory as a whole, particularly our understanding of the
role episode. By understanding the crossover of stress and strain, it becomes
possible to revisit some of the basic assumptions in role theory pertaining to
relations between the various role sets of which the focal person is a member,
and how the expectations emanating from those sets interactively create the
role episode that is at the center of this process.
A question raised for role theory by the present research is whether
crossover might lead people to redefine their roles at work and at home,
thereby altering their perceived role process.
1 Table 1 presents empirical crossover research published between 1977 and 1999,
screened out of a computer search using the keywords crossover, carry-over, contagion,
transfer and transmission.
2 I am grateful to one of the reviewers who raised this issue.
Abbey, A., Abramis, D.J. & Caplan, R.D. Effects of different sources of social support and
social conflict on well-being. Basic and Applied Social Psychology, 1985, 6, 111–29.
Antonovsky, A. Health, stress and coping. San Francisco: Jossey-Bass, 1979.
Bandura, A. Principles of behavior modification. New York: Holt, Rinehart & Winston,
Barnett, R.C. Toward a review and reconceptualization of the work/family literature.
Genetic, Social, and General Psychology Monographs, 1998, 124, 125–82.
Barnett, R.C., Raudenbush, S.W., Brennan, R.T., Pleck, J.H. & Marshall, N.L. Changes in
job and marital experience and change in psychological distress: A longitudinal study
of dual-earner couples. Journal of Personality and Social Psychology, 1995, 69, 839–50.
Beehr, T.A., Johnson, L.B. & Nieva, R. Occupational stress: Coping of police and their
spouses. Journal of Organizational Psychology, 1995, 16, 3–25.
Berkowitz, L. Frustration–aggression hypothesis: Examination and reformulation. Psycho-
logical Bulletin, 1989, 106, 59–73.
Bolger, N., DeLongis, A., Kessler, R. & Wethington, E. The contagion of stress across mul-
tiple roles. Journal of Marriage and the Family, 1989, 51, 175–83.
Booth, A. Wife’s employment and husband’s stress: A replication and refutation. Journal of
Marriage and the Family, 1977, 39, 645–50.
Bronfenbrenner, U. Toward an experimental ecology of human development. American
Psychologist, 1977, 32, 513–31.
Burke, R.J. & Weir, T. Relationship of wives’ employment status to husband, wife, pair
satisfaction and performance. Journal of Marriage and the Family, 1976, 38, 279–87.
Burke, R.J., Brief, A.P. & George, J.M. The role of negative affectivity in understanding
relations between self-report of stressors and strains: A comment on the applied psy-
chology literature. Journal of Applied Psychology, 1993, 78, 402–12.
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 747
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Burke, R.J., Weir, T. & DuWors, R.E. Work demands on administrators and spouse well-
being. Human Relations, 1980, 33, 253–78.
Chan, C.J. & Margolin, G. The relationship between dual-earner couples’ daily work mood
and home affect. Journal of Social and Personal Relationships, 1994, 11, 573–86.
Conger, R.D., Lorenz, R.O., Edler, G.H., Simons, R.L. & Xiaojia, G.E. Husband and wife
differences in response to undesirable life events. Journal of Health and Social Behav-
ior, 1993, 34, 71–88.
Coyne, J.C. & Downey, G. Social factors and psychopathology: Stress, social support, and
coping processes. Annual Review of Psychology, 1991, 42, 401–25.
Crouter, A.C., Perry-Jenkins, M., Huston, T.L. & Crawford, D.W. The influence of work-
induced psychological states on behavior at home. Basic and Applied Social Psychol-
ogy, 1989, 10, 273–92.
Digman, J.M. Personality structure: Emergence of the five-factor model. Psychological
Assessment, 1990, 41, 417–40.
Eckenrode, J. & Gore, S. Stressful events and social support: The significance of context.
In B. Gottlieb (Ed.), Social networks and social support. Beverly Hills, CA: Sage, 1981,
pp. 43–68.
Edwards, J.R. & Baglioni, A.J. Relationship between Type A behavior pattern and mental
and physical symptoms: A comparison of global and component measures. Journal of
Applied Psychology, 1991, 76, 276–90.
Edwards, J.R., Baglioni, A.J. & Cooper, C.L. Stress, Type-A, coping, and psychological and
physical symptoms: A multi-sample test of alternative models. Human Relations, 1990,
43, 919–56.
Etzion, D. The moderating effect of social support on the relationship of stress and burnout.
Journal of Applied Psychology, 1984, 69, 615–22.
Evans, P. & Bartolome, F. Must success cost so much? New York: Basic Books, 1980.
Fletcher, B. Marital relationships as a cause of death: An analysis of occupational mortal-
ity and the hidden consequences of marriage – some U.K. data. Human Relations, 1983,
36, 123–34.
Fletcher, B. Occupation, marriage, and disease specific mortality concordance. Social
Science & Medicine, 1988, 27, 615–22.
Greenhaus, J.H. & Parasuraman, S. A work–nonwork interactive prospective of stress con-
sequences. Journal of Organizational Behavior Management, 1987, 8, 37–60.
Hammer, L.B., Allen, E. & Grigsby, T.D. Work–family conflict in dual-earner couples:
Within individual and crossover effects of work and family. Journal of Vocational
Behavior, 1997, 50, 185–203.
Hatfield, E., Cacioppo, J. & Rapson, R. Emotional contagion. New York: Cambridge Uni-
versity Press, 1994.
Haviland, J.M. & Malatesta, C.Z. The development of sex differences in nonverbal signals:
Fallacies, facts and fantasies. In C. Mayo & N.M. Henley (Eds), Gender and nonver-
bal behavior. New York: Springer-Verlag, 1981, pp. 183–208.
Haynes, S.G., Eaker, E.D. & Feinleib, M. Spouse behavior and coronary heart disease in
men: Prospective results from the Framingham heart study. American Journal of Epi-
demiology, 1983, 118, 1–22.
Hobfoll, S.E. & London, M. The relationship of self concept and social support to
emotional distress among women during war. Journal of Social Clinical Psychology,
1986, 12, 87–100.
Hobfoll, S.E., Dunahoo, C.L., Ben-Porat, Y. & Monnier, J. Gender and coping: The dual-
axis model of coping. American Journal of Community Psychology, 1994, 22, 49–82.
House, J.S. Job stress and social support. Reading, MA: Addison Wesley, 1981.
Jackson, S.E. & Maslach, C. After-effects of job-related stress: Families as victims. Journal
of Occupational Behavior, 1982, 3, 63–77.
Jackson, S.E., Zedeck, S. & Summers, E. Family life disruptions: Effects of job-induced
Human Relations 54(6)
02westman (ds) 26/4/01 9:23 am Page 748
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
structural and emotional interference. Academy of Management Journal, 1985, 28,
Jex, S.M. & Beehr, T.A. Emerging theoretical and methodological issues in the study of
work-related stress. Research in Personnel and Human Resources Management, 1991,
9, 311–65.
Jones, F. & Fletcher, B. An empirical study of occupational stress transmission in working
couples. Human Relations, 1993a, 46, 881–902.
Jones, F. & Fletcher, B. Transmission of occupational stress: A study of daily fluctuations
in work stress and strain and their impact on marital partners. In H. Schroder, K.
Rescke, M. Johnston & S. Maes (Eds), Health psychology: Potential in diversity.
Regensburg: Roderer Verlag, 1993b, pp. 328–38.
Jones, F. & Fletcher, B. Taking work home: A study of daily fluctuations in work stressors,
effects on mood and impact on marital partners. Journal of Occupational and Organiz-
ational Psychology, 1996, 69, 89–106.
Kahn, J.P., Coyne, J.C. & Margolin, G. Depression and marital disagreement: The social
construction of despair. Journal of Social and Personal Relationships, 1985, 447–61.
Kahn, R.L. & Byosiere, P. Stress in organizations. In D. Dunnette & L. M. Hough (Eds),
Handbook of industrial and organizational psychology. Palo Alto, CA: Consulting Psy-
chology Press, 1992, pp. 571–651.
Kahn, R.L., Wolfe, D.M., Quinn, R.P., Snoek, J.D. & Rosenthal, R.A. Organizational
stress. New York: Wiley, 1964.
Kanter, R.M. Work and family in the United States. New York: Russell Sage Foundation,
Katz, L. & Gottman, J. Marital interaction and child outcomes: A longitudinal study of
mediating and moderating processes. In D. Cicchetti & S. Toth (Eds), Emotions, cog-
nition, and representation. Rochester symposium on developmental psychopathology,
Vol. 6. Rochester, NY: University of Rochester Press, 1995, pp. 30142.
Kessler, R.C. A strategy for studying differential vulnerability to the psychological conse-
quences of stress. Journal of Health and Social Behavior, 1979, 20, 100–8.
Kessler, R.C. & McLeod, J.D. Sex differences in vulnerability to undesirable life events.
American Sociological Review, 1984, 49, 620–31.
Lambert, S.J. Processing linking work and family: A critical review and research agenda.
Human Relations, 1990, 43, 239–57.
Long, N.R. & Voges, K.E. Can wives perceive the source of their husbands’ occupational
stress? Journal of Occupational Psychology, 1987, 60, 235–42.
MacEwen, K., Barling, J. & Kelloway, K. Effects of short-term role overload on marital
interactions. Work and Stress, 1992, 6, 117–26.
Mitchell, R., Cronkite, R. & Moos, R. Stress, coping and depression among married
couples. Journal of Abnormal Psychology, 1983, 92, 433–48.
Monnier, J. & Hobfoll, S. (1997) Crossover effects of communal coping. Journal of Social
and Personal Relationships, 1997, 14, 263–70.
Moos, R. Context and coping: Toward a unifying conceptual framework. American Journal
of Community Psychology, 1984, 12, 5–25.
Morrison, D. & Clements, R. The effects of one partner’s job characteristics on the other
partner’s distress: A serendipitous, but naturalistic, experiment. Journal of Occupational
and Organizational Psychology, 1997, 70, 307–24.
Nelson, G. & Beach, S. Sequential interaction in depression: Effects of depressive behavior
on spousal aggression. Behavior Therapy, 1990, 21, 167–82.
Parasuraman, S., Greenhaus, J.H. & Granrose, C.S. Role stressors, social support and well-
being among two-career couples. Journal of Organizational Behavior, 1992, 13,
Pavett, C.M. High-stress professions: Satisfaction, stress, and well-being of spouses of pro-
fessionals. Human Relations, 1986, 39, 1141–54.
Westman Stress and strain crossover
02westman (ds) 26/4/01 9:23 am Page 749
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Pearlin, L. The sociological study of stress. Journal of Health and Social Behavior, 1989,
30, 241–56.
Pearlin, L. & Turner, H.A. The family as a context of the stress process. In S. Kasl & C.
Cooper (Eds), Stress and health: Issues in research methodology. New York: Wiley,
Piotrkowski, C. Work and the family system: A naturalistic study of the working-class and
lower-middle-class families. New York: The Free Press, 1979.
Repetti, R. Effects of daily workload on subsequent behavior during marital interaction:
The roles of social withdrawal and spouse support. Journal of Personality and Social
Psychology, 1989, 57, 651–9.
Riley, D. & Eckenrode, J. Social ties: Costs and benefits within different subgroups. Journal
of Personality and Social Psychology, 1986, 51, 770–8.
Roberts, R. & O’Keefe, S.J. Sex differences in depression reexamined. Journal of Health
and Social Behavior, 1981, 22, 394400.
Rook, S.K., Dooley, D. & Catalano, R. Stress transmission: The effects of husbands’ job
stressors on emotional health of their wives. Journal of Marriage and the Family, 1991,
53, 165–77.
Rosenfield, S. Sex differences in depression. Do women always have higher rates? Journal
of Health and Social Behavior, 1980, 22, 394400.
Rosenfield, S. The cost of sharing: Wives’ employment and husbands’ mental health.
Journal of Health and Social Behavior, 1992, 33, 213–25.
Russell, D.W. & Cutrona, C.E. Social support, stress, and depressive symptoms among the
elderly: Test of a process model. Psychology and Aging, 1991, 6, 190–201.
Schaefer, C., Coyne, J.C. & Lazarus, R.S. The health-related functions of social support.
Journal of Behavioral Medicine, 1981, 4, 381406.
Schmaling, K. & Jacobson, N. Marital interaction and depression. Journal of Abnormal
Psychology, 1990, 99, 229–36.
Shinn, M., Lehmann, S. & Wong, N.W. Social interaction and social support. Journal of
Social Issues, 1984, 40, 55–76.
Spector, P.E. & O’Connell, B.J. The contribution of personality traits, negative affectivity,
locus of control and Type A to the subsequent reports of job stressors and job strains.
Journal of Occupational and Organizational Psychology, 1994, 67, 1–12.
Staines, G.L., Pottic, K.G. & Fudge, D.A. Wives’ employment and husbands’ attitudes
toward work and life satisfaction. Journal of Applied Psychology, 1986, 71, 118–28.
Stotland, E. Exploratory investigations of empathy. In L. Berkowitz (Ed.), Advances in
experimental social psychology, Vol. 4. New York: Academic Press, 1969, pp. 271–314.
Thompson, L. & Walker, A.J. The dyad as a unit of analysis: Conceptual and methodo-
logical issues. Journal of Marriage and the Family, 1982, 44, 889–900.
Vinokur, A., Price, R.H. & Caplan, R.D. Hard times and hurtful partners: How financial
strain affects depression and relationship satisfaction of unemployed persons and their
spouses. Journal of Personality and Social Psychology, 1996, 71, 166–79.
Vinokur, A. & van Ryn, M. Social support and undermining in close relationships: Their
independent effects on mental health of unemployed persons. Journal of Personality and
Social Psychology, 1993, 65, 350–9.
Watson, D. & Clark, L.E. Negative affectivity: The disposition to experience aversive
emotional states. Psychological Bulletin, 1984, 96, 465–98.
Westman, M. & Etzion, D. Crossover of stress, strain and resources from one spouse to
another. Journal of Organizational Behavior, 1995, 16, 169–81.
Westman, M. & Etzion, D. The crossover of strain from school principals to teachers and
vice versa. Journal of Occupational Health Psychology, 1999, 4, 269–78.
Westman, M. & Shirom, A. Dimensions of coping behavior: A proposed conceptual frame-
work. Anxiety, Stress, and Coping, 1995, 8, 87–100.
Westman, M. & Vinokur, A. Unraveling the relationship of distress levels within couples:
Human Relations 54(6)
02westman (ds) 26/4/01 9:23 am Page 750
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
Common stressors, emphatic reactions, or crossover via social interactions? Human
Relations, 1998, 51, 137–56.
Wilensky, H.L. Work, careers and social integration. International Social Science Journal,
1960, 12, 543–60.
Zedeck, S. Introduction: Exploring the domain of work and family concerns. In S. Zedeck
(Ed.), Work, families and organizations. San Francisco: Jossey-Bass, 1992, pp. 1–32.
Westman Stress and strain crossover
Mina Westman received her PhD in Organizational Behavior from the
Faculty of Management at Tel Aviv University. She is a senior lecturer at
the Organizational Behavior Program of the Graduate School of Business
Administration, Tel Aviv. Her current research interests are work and
family, job insecurity, respite and crossover of stress and strain.
02westman (ds) 26/4/01 9:23 am Page 751
at Tel Aviv University on March 11, 2015hum.sagepub.comDownloaded from
... The family-to-work conflict might also have crossover effects (Westman, 2001), that is, this conflict and its outcomes may also be experienced by others with whom the worker has a close relationship, including family members (Bakker & Demerouti, 2013). Evidence suggests that work-to-family conflict and family-to-work conflict are indeed involved in spillover and crossover processes in dual-earner couples (Li et al., 2021;Matei et al., 2021;Matei & Vîrgă, 2022) and in parent-child dyads (Orellana et al., 2021b;Schnettler et al., 2021c). ...
... This study also tested crossover associations between these dual-earner parents (hypotheses 2a to 2f). It was proposed that parents in dual-earner couples experiencing higher family-to-work conflict would mutually influence their satisfaction in the work and family domains (Westman, 2001). We found that one parent's family-towork conflict and job satisfaction were indeed associated with the job satisfaction and family satisfaction of the other parent. ...
... In this study there was a unidirectional crossover from a man's job satisfaction to his partner's life satisfaction, whereas his life satisfaction was not associated with his partners' job satisfaction. This may also reflect the traditional gender-based demands and expectations (Westman, 2001), given the traditional role of men as the family's main "breadwinner" it is likely that the family's financial situation depends more on the father's than on the mother's job. ...
Family-to-work conflict has received less attention in the literature compared to work-to-family conflict. This gap in knowledge is more pronounced during the COVID-19 pandemic, despite the documented increase in family responsibilities in detriment of work performance, particularly for women. Job satisfaction has been identified as a mediator between the family and work domains for the individual, but these family-to-work dynamics remain unexplored at a dyadic level during the pandemic. Therefore, this study tested the relationship between family-to-work conflict and job and family satisfaction, and the mediating role of job satisfaction between family-to-work conflict and family satisfaction, in dual-earner parents. A non-probability sample of 430 dual-earner parents with adolescent children were recruited in Rancagua, Chile. Mothers and fathers answered an online questionnaire with a measure of family-to-work conflict, the Job Satisfaction Scale and Satisfaction with Family Life Scale. Data was analysed using the Actor-Partner Interdependence Model with structural equation modelling. Results showed that, for individuals, a higher family-to-work conflict is linked to lower satisfaction with both their job and family life, and these two types of satisfaction are positively associated with one another. Both parents experience a double negative effect on their family life satisfaction, due to their own, and to their partner's family-to-work conflict; however, for fathers, this effect from their partner occurs via their own job satisfaction. Limitations and implications of this study are discussed, indicating the need of family-oriented workplace policies with a gender perspective to increase satisfaction in the family domain for workers and their families.
... Considering the COR-based approach and the COVID-19 outbreak [51], this is especially crucial for healthcare workers dealing with the management and consequences of the pandemic. According to the most recent integration of the COR theory, namely the crossover model [73,74], psychological states are transmitted across individuals so that relationships at work allow for resource gains. Therefore, compassion at work may act as a valuable mechanism through which employees interact to exchange and build significant resources for their work and well-being. ...
... Finally, in relation to H3 (We expected to find a mediating effect of compassion at work in the relationship between burnout symptoms and general well-being), we showed that experiencing compassion at work may partially buffer the effects of burnout symptoms on general well-being. This is further confirmation of the crossover model [73,74], which states that individuals interact to exchange and build significant resources. Furthermore, our findings are consistent with the PERMA model of well-being [113], according to which positive emotions (E) and positive relationships (R) are two main aspects of wellbeing conditions. ...
... Regarding the ability of the organization to sustain compassionate processes, Atkins and Parker [74] showed that interindividual compassion in organizations does not occur in a void. On the contrary, it is embedded in several different contexts: the personal contexts of the actors directly involved in the interaction (e.g., individual differences, organizational roles), their relational contexts (e.g., their degree of perceived similarity or closeness, the social power occurring between them), and the general organizational context (e.g., the degree to which the organization supports compassionate care in the workforce, through norms, structures, practices, and leadership styles). ...
Full-text available
Managing the COVID-19 pandemic posed several challenges for healthcare professionals, which likely heightened their risk of burnout (Amanullah and Ramesh Shankar, 2020) and, consequently , their general physical and mental health. Although it may not be possible to address and eliminate the causes of burnout, current research informs healthcare organizations about protective strategies to reduce its detrimental consequences. The promotion of compassionate interactions among healthcare professionals may play such a role. Compassion within healthcare organizations positively affects individual performance and well-being. Building on these considerations and within the framework of the Conservation of Resources theory, this study explores the relationships among burnout dimensions, received compassion at work, and general health in 711 Italian healthcare professionals (68.5% female), aged between 21 and 73 years (Mage = 36.4, SD = 11.2). Analyses were conducted to investigate the association between burnout and general well-being (H1) and between burnout symptoms and perceived compassion at work (H2); and the mediational role of compassion in the relationship between burnout symptoms and general well-being. H1 and H2 were confirmed (r < 0.01 for both), and a SEM model showed the mediating role of compassion at work in the association between burnout symptoms and general well-being (RMSEA < 0.08, SRMR < 0.08, CFI and TLI > 0.90). Theoretical and practical implications of the findings are discussed in the paper.
... Crossover is defined as the process that occurs when psychological well-being experienced by a person affects the level of psychological well-being of another person in the same social environment [36]. Consequently, crossover is a dyadic inter-individual transmission of mental states or emotions among closely related individuals, which occurs within a particular domain of life such as family or workplace [37]. ...
... Although Westman [36] initially placed the emphasis on negative forms of well-being, such as job stress, strain, and burnout, it is possible that positive experiences may also cross over from one partner to the other via the same mechanisms as the negative aspects [36]. Work engagement is one of the most widely researched positive aspects of individual well-being [42]. ...
... Although Westman [36] initially placed the emphasis on negative forms of well-being, such as job stress, strain, and burnout, it is possible that positive experiences may also cross over from one partner to the other via the same mechanisms as the negative aspects [36]. Work engagement is one of the most widely researched positive aspects of individual well-being [42]. ...
Full-text available
Work engagement can cross over from one individual to another, and this process may depend on several factors, such as the work context or individual differences. With this study, we argue that agreeableness, one of the Big five personality measures that characterized empathetic, can be instrumental in the crossover process. Specifically, we hypothesize that agreeableness can facilitate this process so that engagement of an actor can more easily cross over to their partner when either of them or both have high agreeableness. To evaluate our hypotheses, we implemented an intervention to the working schedules of 74 participants for two weeks. The intervention involved pairing participants to work together so that to create dyads with varying levels of dissimilarity. The results from a multilevel regression model indicate that there is a crossover effect and partner’s work engagement can be transferred to actor after a two-week collaboration. This effect is further intensified if either one or both members in the dyad are characterized by high levels of agreeableness. These findings help to decode the mechanisms underlying the crossover process and illustrate how to ideally coordinate work dyads to take advantage of the crossover effect and maximize employee engagement.
... Another theory which has the potential to increase our understanding of compersion is crossover theory, which focuses on the emotional influence that close dyadic partners have on each other. The theory suggests that the experiences of one partner (such as, at their job) can affect the well-being of the other partner, for better or for worse (Steiner & Krings, 2016;Westman, 2001;Westman et al., 2009). Although, thus far, crossover theory has only been applied in the context of monogamous relationships, the concept has intuitive appeal in the context of polyamory. ...
... Longitudinal studies would be especially helpful to see how these processes unfold over time. In addition, crossover theory has primarily been used to explain the transmission of stress and other negative emotions between partners (Chen et al., 2020;Westman, 2001). One prediction that follows from this body of research is that the dissolution of a partner's relationship with a metamour may adversely affect the other partner(s), perhaps due to a mirroring of the partner's distress or potentially because the partner's grief may interfere with their full engagement in their other relationships. ...
Full-text available
Compersion refers to the positive feelings, such as joy, excitement and contentment, that one may experience in response to one’s partner’s other consensually non-monogamous (CNM) intimate relationship(s). In the study, we recruited 44 CNM participants who had experienced compersion to complete an open-ended online survey regarding the factors that facilitated and hindered their experiences of compersion. A thematic analysis identified three main themes: intrapersonal/individual factors, experiences in and characteristics of the relationship with one’s partner(s), and feelings/judgments about one’s metamour (one’s partner’s partner). The factors most commonly named by participants as facilitating compersion included: feelings of self-worth, feeling secure and that one’s needs were being met in the relationship with the partner, communication with one’s partner, and positive regard for one’s metamour. Participants shared conflicting experiences regarding the nature of the relationship between jealousy and compersion and whether the ability to feel compersion was innate or learned. Findings were generally consistent with the small body of literature on this phenomenon. Several theories, including Broaden-and-Build, Self-Expansion, and Crossover, may help us understand the underpinnings of compersion and the pathways through which the experience might strengthen and deepen relationships. The study’s results suggest multiple hypotheses ripe for future testing. Increasing our knowledge of this little known phenomenon carries the potential to help us identify strategies to manage jealousy and increase positive feelings across all relationship types.
... Another development that deserves special attention from work break researchers is the increased prevalence in working from home (e.g., Zhang et al., 2021). By drawing from the theories of work-life interface (e.g., crossover theories; Westman, 2001) or work-life integration (Williams et al., 2016), researchers can pose interesting new questions. For example, how does the interaction of family-related demands and work breaks influence well-being and performance? ...
... Working from home blurs the lines between work and nonwork domains. From a crossover perspective (Westman, 2001), these blurred lines imply that people working from home who accumulate strain in the work domain can experience recovery in the home domain (or vice versa). Moreover, work breaks can be interrupted by family demands. ...
Full-text available
Recovery from work is a critical component for employees’ proper functioning. While research has documented the beneficial effects of after-work recovery, it has focused far less on the recovery that happens while at work in the form of work breaks. In this review, we systematically review available empirical evidence on the relationship between work breaks and well-being and performance among knowledge workers. Doing so enables us to (1) integrate studies from multiple disciplines, (2) propose a conceptual framework for categorizing work breaks, and (3) provide a future research agenda for studying the role of work breaks in employee well-being and performance. Using Cochrane’s guidelines, we review observational and intervention studies (N = 83). Based on the extant research, we propose that work breaks can be described and classified in terms of five features: initiator, duration, frequency, activities, and experiences. The result of our review is an integrative model that comprehensively captures the relationship between work breaks and well-being and performance outcomes, as well as the mechanisms and boundary conditions of these relationships. We conclude by proposing avenues for the future study and practice of work breaks.
... Moreover, COR theory states that individuals' work engagement is more likely to be facilitated through a resource crossover process whereby organizational and social resources are transferred to individuals, thus increasing their personal resources and providing the nutrients necessary to enhance their level of engagement [7]. Crossover is achieved through direct transmission (i.e., via empathy), positive interactions between partners, and a work environment that supports all employees [32]. On the one hand, colleagues' positive state of work can directly affect other people's emotions and behavior [33]. ...
Full-text available
The sustainable development of education requires the continuous engagement of teachers, and the professional community has long been considered an important facilitator of teacher engagement. However, teachers’ professional community has often been analyzed as a unified construct, and thus the question of how teacher engagement is enhanced remains unanswered. Based on the conservation of resources theory, in this study, we investigated how teacher work engagement was affected by the crossover of job resources between the professional community (including shared norms, collective responsibility, collaboration, and reflective dialogue) and teachers (self-efficacy). The sample included 1123 primary and secondary school teachers in China. Covariance structural modeling was used to test our hypotheses. Shared norms and collective responsibility played a fundamental role and positively predicted collaboration, which in turn enhanced reflective dialogue. Teacher self-efficacy partially mediated the effect of the four dimensions of the professional community on teachers’ work engagement. The findings of this study indicate that the professional community offers valuable organizational and social resources that can be used by teachers to enhance their personal resources, such as self-efficacy, and thus become more engaged in their work. Shared norms and collective responsibility serve to shape a growth-oriented school culture that stimulates teachers’ willingness to collaborate and improves their confidence in teaching, and thus should be stressed by school leaders when introducing changes.
... Drawing upon basic aspects of polyvagal theory (Porges, 2001) and established findings on stress crossover (Bakker et al., 2009;Neff & Karney, 2007;Westman, 2001), the present study sought to investigate how partners influence each other during interactions based on stress cues in their voices (f o ) and how externally induced stress changes the dynamic process by which stress is transferred between partners. We summarize our key findings below before turning to address their limitations and implications for future modeling of dynamic processes in dyads. ...
It has been demonstrated that stress, experienced outside of a relationship, can spill into a relationship and cross over during interactions from one partner to the other. However, the mechanism of how stress cross over in real-time between partners is still unknown. To overcome this limitation, we invited 189 couples (N = 378 individuals) for two interactions and stressed either the man, the woman, or both partners between the interactions with a standardized stress-induction procedure. Vocally-encoded emotional arousal (i.e., fundamental frequency, indexed as fo) was extracted from both partners in 25,834 talk turns. Dynamical systems modeling revealed four patterns of dynamic influence prior to stress induction, which started to erode after the stress induction. This demonstrates that the initially unstressed mates become stressed during the behavioral exchange and stress crosses over from the unstressed partner to the stressed mates, interfering with their ability to down-regulate stress.
Full-text available
Family supportive supervisor behaviors (FSSB) have emerged as a powerful resource of informal support for the well‐being and development of employees. However, research to date offers limited insight into the antecedents and underlying processes that may trigger FSSBs. We investigate the association between family motivation of supervisors and FSSBs, and how the latter mediates the association between supervisors’ family motivation and subordinates’ work performance. Furthermore, we examine the role of supervisors’ satisfaction with their work–family balance as a contextual variable influencing our proposed associations. We draw on FSSB and perspective taking theory as over‐arching frameworks for our hypotheses. Using matched and multisource supervisor‐subordinate data collected from an organization in Chile (196 subordinates and 75 supervisors), our findings revealed that FSSBs are mechanisms linking supervisors’ family motivation to subordinates’ work performance. Interestingly, this positive association is moderated by supervisors’ satisfaction with their work–family balance, such that the mediation of FSSBs is stronger for supervisors who are not satisfied with their work–family balance.
Purpose Building on the spillover-crossover model (Bakker and Demerouti, 2013), this study aimed to examine the processes through which three forms of social support at work (i.e. from coworkers, from supervisor and organizational family-friendly) were positively associated with an individual's level of work-family balance (spillover effect), and through this latter, with one's partner's family life satisfaction (crossover effect), via the partner's perception of family social support as provided by the incumbent person. Design/methodology/approach The authors sampled 369 heterosexual couples using a time-lagged design, surveying forms of social support at work and work-family balance at t1 and family social support and partner's family life satisfaction at t2. Data were analyzed through structural equation modeling. Findings The results showed that coworkers’ support and organizational family-friendly support positively predicted work-family balance. Furthermore, work-family balance mediated the associations between organizational family-friendly support and coworkers’ support with instrumental family social support. Moreover, only emotional family social support positively predicted partner's family life satisfaction. Originality/value The authors simultaneously examined the direct and indirect associations of three concurrent forms of social support at work with one's work-family balance (spillover effect). Moreover, in line with the spillover-crossover model, the authors adopted a systemic approach and assessed how one's work-family balance is associated with emotional and instrumental family social support as perceived by one's partner and the latter's family life satisfaction (crossover effect).
Full-text available
Emergency responders (ERs), often termed First Responders, such as police, fire and paramedic roles are exposed to occupational stressors including high workload, and exposure to trauma from critical incidents, both of which can affect their mental health and wellbeing. Little is known about the impact of the ER occupation on the mental health and wellbeing of their families. The aim of the current study was to investigate what mental health and wellbeing outcomes and experiences have been researched internationally in ER families, and to examine the prevalence and associated risk and protective factors of these outcomes. We conducted a systematic review in accordance with an a priori PROSPERO approved protocol (PROSPERO 2019 CRD42019134974). Forty-three studies were identified for inclusion. The majority of studies used a quantitative, cross-sectional design and were conducted in the United States; just over half assessed police/law enforcement families. Themes of topics investigated included: 1) Spousal/partner mental health and wellbeing; 2) Couple relationships; 3) Child mental health and wellbeing; 4) Family support and coping strategies; and 5) Positive outcomes. The review identified limited evidence regarding the prevalence of mental health and wellbeing outcomes. Family experiences and risk factors described were ER work-stress spillover negatively impacting spousal/partner wellbeing, couple relationships, and domestic violence. Traumatic exposure risk factors included concerns family had for the safety of their ER partner, the negative impact of an ER partners' mental health problem on the couples' communication and on family mental health outcomes. Protective factors included social support; however, a lack of organisational support for families was reported in some studies. Study limitations and future research needs are discussed. Progressing this area of research is important to improve knowledge of baseline needs of ER families to be able to target interventions, improve public health, and support ER's operational effectiveness.
Full-text available
Research on work/family issues is currently being done by investigators from a variety of disciplines, including psychology, occupational health, sociology, and, less centrally, organizational behavior. Such energy and diversity might be expected to yield significant advances; however, for the most part, this promise has not been realized. Progress has been hampered by the lack of an inclusive model for understanding the processes by which work and family variables influence one another, a model that is theoretically grounded and integrates the major paradigms from these several disciplines. In an effort to develop a more inclusive and theoretically-grounded research model, I have organized this review around three critical theoretical issues that strongly shape the research literature and that need to be addressed in any proposed model.
A nursery school teacher brought a little girl to a psychologist’s attention when the girl was being made the focus of contemptuous looks by her peers and playmates. “Why do they look at her that way?” asked the teacher. “She’s attractive and bright and she wants to play with the other children.” The psychologist looked over at the child. “What little girl?” she asked. “That’s a boy.” The psychologist was wrong and that was the problem. This 4-year-old girl had the lower brow ridge of a boy, the smaller mouth and the gestures of a boy. She did not often smile; she stood obliquely and she gestured expansively. She engaged in rough-and-tumble play. She also nurtured dolls and was learning to read early. The children in playschool did not know what to make of her. The adults said, “There’s something strange about the nice child over there, but I can’t quite put my finger on it.” That last sentence sums up what our research in nonverbal behavior is all about. We are putting our fingers on the important things that people have few words for—the barely conscious gestures that separate and identify the sexes.