Subjective well-being (SWB) refers to satisfaction with one's
life and experience of more frequent pleasant emotions as compared to
unpleasant emotions (Diener et al., 1999). In the workplace, SWB affects
the productivity of employees, their ability to make decisions, and
attendance (Danna and Griffin, 1999). Because employees spend a
substantial part of their lives at work, and are dependent on their job
to meet several personal needs, their work and personal lives are
intertwined. As a result, stressors may originate from the conflict be
tween these roles and that conflict may affect the overall well-being of
an employee (Danna and Griffin, 1999). Yet despite the causes and
effects of SWB related to the workplace, SWB remains an under-explored
subject in the work domain. As Danna and Griffin noted in their review
of past research, "Indeed, for a variety of reasons these [health
and well-being] issues should occupy a much more prominent niche in
mainstream organizational research" (1999: 357; words in brackets
added).
While cross-cultural studies on SWB have found differences in
average SWB scores of respondents across countries and have attributed
it to various factors including individualism-collectivism, status of
human rights, and wealth (Diener et al., 1995), not much research has
been conducted to examine the mechanisms that link nationality to SWB.
Accordingly, our study aims to extend past research by identifying the
role of two work domain factors--work locus of control and family-work
conflict--in explaining cross-cultural differences in SWB. We examine
these linkages in the case of managers in the United States and the
People's Republic of China.
With a population of nearly 1.3 billion and a gross domestic
product of nearly $8.86 trillion (Central Intelligence Agency, 2006),
the People's Republic of China (referred to as "China" in
the subsequent text) has emerged as an important player in the world
economy. The international marketplace recognizes significant business
opportunities in China: joint ventures, outsourcing partnerships,
low-cost suppliers of a wide variety of goods from toys to high-tech
electronic products, and significant and largely untapped markets (e.g.,
Erickson, 2001). Cross-cultural researchers have highlighted the
difficulties managers face when seeking to transfer management
techniques such as human resource management practices and policies into
the Chinese context (Teagarden and Von Glinow, 1990) and have documented
the myriad ways that the eastern culture in China differs dramatically
from western cultures such as the U.S.
In addition to differing economic, legal, political, and
educational systems in the U.S. and China-such as the smaller proportion
of private sector jobs (Lu et al., 2002) and the intensely competitive
educational system (Tang, 1999) in China-research over the past 20 years
demonstrates that Chinese and U.S. managers differ considerably on a
number of cultural dimensions such as individualism-collectivism, power
distance, and long-term orientation (e.g., Chen, 1995; Earley, 1989,
1993; Hofstede, 1991, 1993; Hofstede and Bond, 1988; Ralston et al.,
1993; Schwartz, 1994; Shenkar and Ronen, 1987; Smith et al., 1996). Of
these dimensions, a greater volume of research has focused on the
differences between the U.S. and China related to
individualism-collectivism, which refers to the extent to which
individuals are connected to their society (Earley and Gibson, 1998).
Pursuing individual goals is more important than pursuing group goals in
an individualist society. A meta-analysis of individualism-collectivism
research (Oyserman et al., 2002) supported Hofstede's (1991)
assertion that people in the U.S. were higher in individualism and lower
in collectivism compared to the Chinese.
Another cultural dimension with significant differences between the
two nations is power distance. In high-power distance societies,
employees are thought to accept hierarchy and power differences and
comply quickly and automatically with the decisions of the powerful
(Hofstede, 1980, 1986). Studies have supported the idea of greater
acceptance of hierarchy in East Asia than in low-power distance, Western
countries (Bond et al., 1985; Schwartz, 1994; Westwood and Everett,
1987). Similarly, China is characterized by a long-term orientation,
while people in the U.S. tend to be more focused on the short term
(Bond, 1987).
These cultural differences are also manifested in employee
well-being. Spector's results indicated that employees in China
have lower psychological and physical well-being and reduced job
satisfaction relative to their U.S. counterparts (Spector et al., 2001,
2004). Similarly, Diener and his colleagues found lower SWB among people
in China compared to people in the U.S. (e.g., Diener and Suh, 1995).
However, there is a growing recognition that research needs to focus
also on the mechanisms that explain how nationality affects SWB (Diener
et al., 2003; Hong et al., 2000; Joplin et al., 2003). The purpose of
our research is to fill this gap by uncovering additional mediating
variables between nationality and employee well-being in the U.S. and
China. In particular, we focus on work locus of control and family-work
conflict as the primary factors that differ between the U.S. and China
and are instrumental in explaining differences in SWB. We argue that the
effects of work locus of control and family-work conflict on SWB will be
further mediated by social support and active coping. The relationships
are depicted in Figure I, and are explained in the following sections.
NATIONAL CULTURE AS PREDICTOR OF WORK LOCUS OF CONTROL, FAMILY-WORK
CONFLICT, AND SUBJECTIVE WELL-BEING
Work locus of control (WLOC) is an extension of Rotter's
(1966) concept of locus of control which asserts that individuals differ
in terms of their beliefs about whether they control the outcomes in
their lives (i.e., internal locus of control) or the outcomes are
controlled by factors such as luck and other people (i.e., external
locus of control). Building on the argument of Paulhus and Christie
(1981) that there might be a generalized perception of control for
various spheres of an individual's life, Spector (1988) formulated
the work locus of control scale. The notion of WLOC has frequently been
linked with increased job satisfaction and psychological well-being
(Karasek, 1979; Spector, 1986; Spector et al., 2002).
While WLOC differs greatly among individuals, cross-cultural
research suggests national culture can have an impact on the level and
the ways in which WLOC is experienced (e.g., Lu et al., 2003).
Individualism-Collectivism is one dimension of national culture that can
influence WLOC. Individualism is an expression of the need for
independence and self-sufficiency whereas collectivism is an expression
of the need for affiliation (Hofstede, 1980). For example, members of
individualistic societies perceive themselves as having primary control
over events in their lives and place value on autonomy and achievement
as the result of one's actions. On the other hand, in collectivist
societies individuals give primacy to group goals over personal goals.
Consequently, in a collectivist society, it may be considered
appropriate to grant power to the group (or social institutions) for
actions and outcomes in one's personal life. Thus, it has been
suggested that members of collectivist societies experience secondary
control through their emphasis on interpersonal relationships and the
social environment (Triandis, 1994). In a study comparing the U.S. and
Japan, another collectivist society, researchers found that individuals
in Japan were more likely to form alliances with powerful people so as
to be better assured of outcomes. In such a society, a feeling might
generate among individuals that their fate rests in the hands of
powerful others rather than in their own hands (Weisz et al., 1984).
Given the emphasis on primary control in individualist cultures such as
the United States, and the emphasis on secondary control in collectivist
cultures such as China, our hypothesis is:
Hypothesis 1a: Employees in the U.S. are likely to report a higher
internal work locus of control than employees in China.
[FIGURE I OMITTED]
Collectivist societies tend to be comparatively more homogeneous
because people have a common heritage going back several centuries
causing the perception of the society to be that of a big family (Earley
and Gibson, 1998). Triandis et al. (1988) used the terms idiocentric and
allocentric to refer to people in individualist and collectivist
cultures, respectively. Idiocentric people have a self-concept
independent of others whereas allocentric people have an inter-dependent
self-concept. In individualist societies the concept of kinship refers
to immediate family (spouse and children), while people in collectivist
societies extend the concept of kinship to distant relations,
neighborhood, and other institutions of the society. In the United
States nuclear families are often located a great distance from parents
and siblings, while in China it is more likely that families will live
near the extended family. As Hofstede argued, "individualism
implies a loosely knit social framework in which people are supposed to
take care of themselves and of their immediate families only"
(1980: 45). Thus, one would expect higher importance being accorded to
the family in an individualist society compared to a collectivist
society where the family is on level ground with other institutions in
the society, including the workplace. As Gannon (2001) noted in his work
on global cultures, the U.S. culture depicts a profound belief in
family. On the other hand, as Yang et al. (2000) argued, work in China
might be seen as something for which the family must be ready to
sacrifice because it is for the collective good of the society.
In a pioneering work, Kahn et al. defined role conflict as the
"simultaneous occurrence of two (or more) sets of pressures such
that compliance with one would make more difficult compliance with the
other" (1964: 19). Family-work conflict refers to lack of
compatibility between the roles one plays at work and in the family such
that family-related role pressures obstruct work performance. The
conflict could be of three kinds: time-based, behavior-based, and
role-based (Greenhaus and Beutell, 1985). For example, an individual may
not be able to find the required time to meet work requirements due to
more time demanded by his or her family role. However, in China there is
less separation of the two domains (IshiiKuntz, 1994). The Chinese
employees regard work as contributing to the welfare of the family and
consider it to be more important than leisure (Redding, 1993). Thus, it
is less likely that the Chinese employees will view the family as
interfering with their work.
Hypothesis 1b: Employees in the U.S. will report a greater level of
family-work conflict than employees in China.
Subjective well-being refers to a state of being in which
individuals experience general happiness and are relatively free of
worry and depression. Both internal and external factors have been found
to influence SWB. For example, extraversion (Lucas and Fujita, 2000),
neuroticism (Fujita et al., 1991), and the pursuit of goals (Emmons and
Diener, 1986) have been found to be correlated with SWB.
The influence of national culture may provide a better explanation
for the differences in SWB. Using multiple samples of college students
from various nations, Diener and Oishi (in press) compared Asian and
non-Asian countries on life satisfaction, emotions, and happiness.
Students in the U.S. consistently reported higher levels than students
in China on these dimensions. Specifically, in the experience of
pleasant emotions and happiness, on a ten-point scale, the U.S.
respondents averaged 6.1 and 8.1, while Chinese respondents averaged 4.6
and 7.3, respectively. Additional research using adults has shown
similar findings. For example, Veenhoven (1993) found the average life
satisfaction score for Denmark, a country closer to the U.S. on
individualism, was close to 8 whereas the Japanese life satisfaction
score was closer to 6 on a scale of 1 to 10. Similarly, Oishi (2002)
found that Asian Americans reported lower SWB than European Americans in
the U.S. Consistent with past research, we predict:
Hypothesis 1c: Employees in the U.S. will report a greater level of
subjective well-being than employees in China.
WLOC, Family-Work Conflict, Social Support, Active Coping, and
Subjective Well-being
Social support includes emotional concern, care, sympathetic
listening, and advice one receives from others (Fenlason and Beehr,
1994). It plays a crucial role in linking the appraisal of stress to the
outcome of individual well-being. AS House stated," ... the
quantity and quality of people's social relationships with spouses,
friends, co-workers, and supervisors appear to have an important bearing
on the amount of stress they experience, their overall well-being, and
on the likelihood that stress will adversely affect their overall
well-being" (1981: 7).
The extent to which individuals seek social support can be
determined by both individual and environmental characteristics.
Individual characteristics include extraversion (Zellars and Perrewe,
2001) and optimism (Scheier et al., 1986). We argue that people with a
higher internal WLOC will be able to gain greater social support from
co-workers for the following reasons. Firstly, as a means to control the
work environment, people with an internal WLOC will proactively build
relations with their friends, relatives, and co-workers. For example, in
one of the early experiments, Goodstadt and Hjelle (1973) found that
internals attempted to influence others through personal persuasion to a
greater extent compared to externals. On similar lines, Ringer and Boss
(2000) found that internals used reason as an effective tool for
inter-personal influence. More recently, among a sample of 600 employees
of the Canadian Civil Service, Ito and Brotheridge (2003) found those
with an internal locus of control were more likely to seek advice and
assistance. Further, internals are more likely to engage in pro-social
behavior, thereby receiving something in return from co-workers (Taylor
and Brown, 1988). The initiative shown by people with a higher internal
WLOC (Spector, 1988) will make them more likely to receive social
support from their supervisors and co-workers. Thus,
Hypothesis 2a: WLOC will be related to social support such that
internals will report higher levels of social support than externals.
Active coping strategies, as the name indicates, involve actively
controlling the stressor. It is the attempt to tackle the problems by
"cognitively analyzing the situation and/or by concrete
action" (de Rijk et al., 1998: 1). Individuals with a high internal
locus of control not only perceive greater control but also seek
situations in which control is possible (Julian and Katz, 1968; Kahle,
1980), exert more effort to control their environment in any given
situation, seek new information and make better use of it (Phares,
1976). Thus, such individuals are likely to be more naturally inclined
and better equipped to directly confront the challenges rather than
avoid them (Latack, 1986). Individuals with an internal locus of control
are more likely to engage in the active coping strategies of working
harder, longer, and more efficiently than those with an external locus
of control (Ito and Brotheridge, 2003). We believe work locus of control
will operate similarly to the broader construct of locus of control in
the prediction of active coping. Consequently,
Hypothesis 2b: WLOC will be related to active coping such that
internals will engage in more active coping than externals.
Internals believe they control the outcomes and so, in order to be
consistent with these beliefs, they will make an attempt to deal with
work-related challenges (including stress) to achieve a higher state of
well-being. Externals, on the other hand, tend to be more prone to
anxiety in the face of daily challenges (Spector, 1982). In research
exploring whether the relationship between work locus of control and
well-being is universal, Spector et al. (2002) analyzed data from 5,185
participants in 24 geopolitical entities. No support was found for the
universality of the relationship between work locus of control and
physical well-being as non-significant values were found for seven of
the samples (e.g., France, Hong Kong, South Africa, U.K.). However,
support was found for the universality of the relationship between work
locus of control and psychological well-being in all but three of the
samples (i.e., U.K., Romania, and Bulgaria). The authors did not find
the moderating effect of individualism-collectivism on the relationship
between work locus of control and psychological well-being, thereby
indicating that the positive relationship between internal WLOC and
well-being can be generalized across nations. This allows us to argue
that the relationship between WLOC and SWB will hold in both the U.S.
and China, even though these two countries differ on the
individualism-collectivism dimension.
Hypothesis 2c: WLOC will be positively related to subjective
well-being such that internals will report higher levels of subjective
well-being than externals.
Numerous physical and psychological negative outcomes have been
associated with family-work conflict. These have included increased
depression (Googins, 1991), stress (Jex and Elacqua, 1999), increased
somatic complaints (Burke, 1988), physical ailments (Frone et al., 1997)
and lower energy levels (Googins, 1991). Previous research has also
acknowledged the detrimental effects of family-work conflict on SWB
(Judge and Colquitt, 2004; Martins et al., 2002). Consistent with past
research, we predict:
Hypothesis 3: Family-work conflict will be negatively related to
subjective well-being.
Social support bolsters a feeling of self-worth and positive
thinking (Williams and House, 1985). Thus, people receiving social
support might look upon problems as opportunities and think of
themselves as potential winners. The information support from co-workers
in terms of new strategies to overcome problems would enable an
individual to successfully confront the challenges. As Daniels and Guppy
(1994) argued, social support can be viewed as a means of control over
some aspects of the working environment, although through the actions of
others. Further, Marcelissen et al. (1988) argued that social support
could modify the cognitive interpretation and analysis of the problem on
hand. Social support in the workplace may prevent role overload and
provide new strategies to solve the problems. Thus, individuals who
receive social support will be able to use active coping strategies in
dealing with stress.
Hypothesis 4a: Social support will be positively related to active
coping.
Chay (1993) argued that irrespective of the stress level faced by a
person, supportive relationships generally enhance well-being. This
could be because social support is likely to fulfill the need for
affiliation of employees and raise one's self-evaluation (Caplan et
al., 1975). Payne's (1980) literature review found evidence for a
direct relationship between social support and well-being. Similar
findings were echoed by Henderson (1984) in a subsequent review. As
Cohen and Wills concluded in another review article, "This kind of
support could be related to overall well-being because it provides
positive affect, a sense of predictability and stability in one's
life situation, and a recognition of self-worth" (1985: 311).
Hypothesis 4b: Social support is positively related to subjective
well-being.
Koeske, Kirk, and Koeske (1993) found that active coping served as
a buffer against the effect of job stressors on negative job-related
outcomes such as burnout and dissatisfaction. Individuals following
active coping might be able to resolve their problems through the
effective use of information, material, and social resources. Leana and
Feldman (1988) argued that even if the problem cannot immediately be
resolved, individuals following active coping might be able to control
their stress through other forms of constructive behavior (e.g., seeking
counseling, physical exercise). On the other hand, a person engaging in
escapist or avoidance coping would continue to suffer from negative
emotions because he/she does not address the source of stress (Zellars
et al, 2004). Thus, our final hypothesis is as follows:
Hypothesis 5: Active coping will be positively related to
subjective well-being.
METHOD
Participants
For the U.S. sample, 153 surveys were distributed to Executive MBA
students at two large public universities. These students were working
professionals. We received 136 completed responses, thereby giving us a
response rate of 88.9%. Due to a clerical error, we received demographic
responses on only about 50% of the U.S. sample. Of those U.S. responses,
53 percent were male and the average age was 29.53 years with a standard
deviation of 6.92. For the Chinese sample, we distributed surveys to 232
middle- and senior-level managers working in state-owned companies in
the Tianjin and Shanghai areas of China who were attending executive
business education classes. We received 155 completed surveys for a
response rate of 67%. Sixty-nine percent of the Chinese respondents were
male and the average age was 37.2 years, with a standard deviation of
7.64. All the respondents were assured anonymity.
Measurement Equivalence
In cross-national research, it is important to ascertain whether a
survey item has the same meaning for respondents from different
countries (Cheung and Rensvold, 1999). As a first step, it is important
to maintain conceptual equivalence of the measures across two cultures
and two languages. We translated the English version to Chinese and
translated back into English with the help of two independent,
bi-lingual graduate students (Brislin, 1980) to ensure consistency (in
word and interpretation) across the two versions. The format and
response options of the surveys were identical for the English and
Chinese surveys.
One of the methods to verify measurement equivalence of scale items
is factorial invariance. Factorial invariance exists when the factor
loadings of scale items in relation to the underlying latent constructs
are not significantly different between the two groups (Cheung and
Rensvold, 1999). If certain items are not factorially invariant, then
one recommendation is to drop these items from subsequent analysis.
Accordingly, we dropped seven items from the work locus of control scale
and five items from the SWB scale. All the items on the other scales
demonstrated measurement invariance. The differences in factor loadings
between the two samples on work locus of control and SWB scales are
shown in Tables 1 and 2.
Measures
Nationality. In line with prior comparative research across nations
(e.g., Spector et al., 2004; Yang et al., 2000), we did not directly
measure the complex socio-cultural differences but instead relied on
past findings of country differences (Hofstede, 1984; Spector et al.,
2002) to capture the nationality differences through a dichotomous
measure. In the present study, the U.S. managers were coded as
"one" and the Chinese managers were coded as "zero."
The justification for using a dichotomous variable to capture
cultural differences is as follows. As Oyserman et al. noted, a major
contentious assumption in using the direct assessment approach (i.e.,
using scales of Individualism-Collectivism) is that "cultural frame
is a form of declarative knowledge (e.g., attitudes, values, beliefs)
that respondents can report on rather than some set of more subtle and
implicit practices and social structures that respondents cannot report
on because these practices are deeply woven into everyday life and are a
normal part of living" (2002: 7). On similar lines, Peng, Nisbett,
and Wong (1997) argued that even a tiny bit of individualism in the
Chinese context would stand as an outlier and might be given the highest
possible rating on a Likert-type scale by a Chinese respondent. Thus,
the comparison of scaled scores between the U.S. and China might he
misleading. Peng et al. found that the rating method (for a scale of
cultural values) of measuring differences between people in China and
the U.S. had low convergence with the cultural difference assessed by
independent experts. Accordingly, we relied on a large body of previous
research that has used diverse methods and measures to conclude that the
U.S. respondents would rank much higher on individualism and much lower
on power distance compared to the Chinese. Therefore, we used a
dichotomous variable to indicate the cultural contrast between the two
countries.
Work Locus of Control (WLOC). We used Spector's (1988) scale
consisting of 16 items to measure WLOC. A sample item from this scale is
"If you know what you want out of a job, you can find a job that
gives it to you." The items were coded such that higher scores
reflect an internal work locus of control while lower scores reflect an
external work locus of control. The reliability of the scale for all
respondents was .83.
Even though seven items that were factorially variant between the
two samples were dropped and one item dropped in order to increase the
magnitude of reliability coefficient (Table 2), the remaining eight
items captured equally the aspects of internal and external control,
thereby preserving the content validity of this scale.
Family-Work Conflict. This five-item scale was developed by
Netemeyer, Boles, and McMurrian (1996) with data collected from three
samples. A sample item from this scale is "I have to put off doing
things at work because of demands on my time at home." The
reliability of this scale for all respondents was .90.
Social Support. We measured the individual respondent's
self-reported social support from four sources: former co-workers,
spouse or significant other, friends, and relatives. There was a
four-item scale for each source of social support. A sample item is
"How much do each of these people go out of their way to make your
life easier for you?" There was also an item for social support
from any source other than the four mentioned earlier. This scale was
adapted from Caplan et al. (1975). The reliability of this scale for all
respondents was .84.
Active Coping. We measured active coping with a 17-item scale taken
from Latack (1986). A sample item is "When things go wrong at work,
I try to be very organized so that I can keep on top of things."
The reliability of this scale for all respondents was .86.
Subjective Well-being (SWB). Following the earlier work of Diener
and colleagues (Diener et al., 1995; Diener et al., 1995), SWB was
measured through the scales for affective well-being and life
satisfaction. Both scales were five-point Likert type and had the same
anchors. In the affective well-being scale, respondents indicated how
frequently over the past two weeks they had experienced twelve different
(positive and negative) emotions, such as pride, anger, and joy. The
life satisfaction scale had five items and a sample item is "In
most ways my life is close to my ideal." The reliability of the
scale for all respondents was .85.
Based on factorial invariance analysis, four items that were
variant between the two groups were dropped from affective well-being
and one item was dropped from the life satisfaction scale, as shown in
Table 2.
RESULTS
Descriptive statistics and correlations between variables are
reported in Table 3. As can be seen, there is a positive correlation
between nationality and SWB such that the U.S. managers reported higher
SWB than Chinese managers.
We tested the hypothesized model in Figure I through structural
equation modeling. Single indicators of latent variables (not shown in
Figures I and II) were used for WLOC, family-work conflict, social
support, active coping, and SWB to account for the measurement error in
respective scales (Joreskog and Sorbom, 1988). Error variances of these
variables were fixed at (1-[alpha]) times variance.
First, we tested the conceptual model shown in Figure I. The fit
indices were: [X.sup.2] = 15.07, df = 4, p < .05; NFI = .99; CFI =
1.00; RMSEA = .10. Chi square is significant and RMSEA is greater than
.08, and this indicates that the hypothesized model does not fit the
data and there is scope for improvement. It was possible to improve the
model further by dropping two non-significant paths (from country to
SWB, and from family-work conflict to SWB) and adding two new paths
(from WLOC to family-work conflict, and from family-work conflict to
active coping). Accordingly, the significant paths in the case of the
best fitting model are shown in Figure II. The fit indices for this
model were: [X.sup.2] = 8.67, df = 4, p > .05; NFI = 1.00; CFI =
1.00; RMSEA = .06. Since chi square is non-significant, it implies that
the revised model is consistent with the data.
Results revealed strong support for Hypotheses la and lb. As
predicted, U.S. managers reported a higher internal work locus of
control and a higher level of family-work conflict. Though there was a
significant zero order correlation between country and well-being,
indicating that the U.S. managers reported higher well-being than the
Chinese managers, there was no residual effect of country on SWB in
Figure II. This implies that the effect of country on SWB (Hypothesis
1c) is completely mediated by other variables in the model. Though
family-work conflict and SWB were negatively correlated, as shown in
Table 3, there was no direct relationship between family-work conflict
and SWB in Figure II. In a separate model (not shown) with country,
family-work conflict, and SWB as the variables, we found a significant,
negative relationship between family-work conflict and SWB. A new path
showing negative relationship between family-work conflict and active
coping also emerged in the best fitting model in Figure II. Taken
together, this implies that the effect of family-work conflict on SWB
was negative, thereby supporting Hypothesis 3 but the effect was
completely mediated through active coping. Hypotheses 2a and 2c were
supported: internals reported greater social support than externals and
displayed higher well-being. However, work locus of control was not
related to active coping and, so, Hypothesis 2b was not supported.
However, a new path emerged in the best fitting model indicating a
negative relationship between internal work locus of control and
family-work conflict. Thus, work locus of control has a direct effect on
SWB as well as an indirect effect mediated through family-work conflict,
active coping, and social support. Social support was positively related
to active coping and SWB and, so, Hypotheses 4a and 4b were supported.
Finally, active coping had a positive relationship with SWB, thereby
supporting Hypothesis 5.
[FIGURE II OMITTED]
DISCUSSION
Our purpose was to examine the mechanisms that link the nationality
of employees with their SWB. Though SWB has relevance for the workplace
(Danna and Griffin, 1999), there has been little research on the role of
work domain factors as the explanation of differences in SWB across
cultures. Our focus was on the United States and China and we used
mainly the cultural differences as the underlying dimension to predict
family-work conflict as a stressor and WLOC as a personality-based
resource to manage stress effectively. As one of the few studies that
utilized data from public sector managers in China--a group that has so
far been relatively difficult to access by U.S. researchers--we believe
the findings from our comparative study will be useful for researchers
as well as practitioners in the following ways.
Theoretical Implications
Consistent with prior research of Diener and colleagues, we found
U.S. respondents to report higher SWB than the Chinese respondents, as
indicated by the correlation between nationality and SWB. The effect of
nationality on SWB was completely mediated by WLOC, family-work
conflict, social support, and active coping. Consistent with the work of
Spector and colleagues, the U.S. employees reported higher internal WLOC
than the Chinese employees. Similarly, as suggested by Yang et al.
(2000), we found that U.S. employees experienced greater family-work
conflict compared to their Chinese counterparts.
Consistent with earlier research (Spector, 1989, 1988), work locus
of control had an independent positive relationship with SWB. In
addition, WLOC also had a relationship with SWB mediated through social
support, family-work conflict, and active coping. Social support had a
positive relationship with SWB, consistent with the direct effect theory
of Cohen and Wills (1985). Though the path between family-work conflict
and SWB was not significant, there was an indirect negative relationship
between family-work conflict and SWB mediated through active coping.
Thus, although not hypothesized, there was a direct negative
relationship between family-work conflict and active coping.
Taken together, the findings imply that between-culture differences
in SWB in the case of U.S. and China could be explained primarily by
family-work conflict and WLOC, which, in turn, have direct and indirect
effects on SWB. The higher family-work conflict in the case of U.S.
managers works towards lowering their SWB compared to Chinese managers.
However, internal WLOC, being higher in the U.S. managers compared to
the Chinese, has the balancing effect of improving their SWB. It may be
noted that there were no differences between the two country samples in
the extent of social support and active coping. Thus, our findings
highlight that cultural differences are important only in affecting the
nature of stressors and individual dispositions--family-work conflict
and work locus of control, respectively, in our study. In other words,
cultural differences might affect only the distal variables in the
pathways to SWB.
Our attempt to improve the model fit also indicated that internal
WLOC helped in reducing family-work conflict. Thus, even though U.S.
managers experienced higher levels of family-work conflict due to
differences in the socio-economic setup, the personal disposition to
feel more strongly in control of events does mitigate the family-work
conflict. While our focus was on family-work conflict, it is quite
possible that factors related to a person's work may adversely
affect how well he/she is able to perform family-related roles thereby
lowering one's subjective well-being. Examining work-family
conflict (Boles et al., 2001) in the context of country differences in
subjective wellbeing is an important area of future research.
Finally, it is important to note that the variables in our model
accounted only for partial variance in SWB. There is ample scope for
future research to examine other personality traits (e.g., self-esteem,
emotional stability), work stressors (e.g., job insecurity), and coping
mechanisms (e.g., avoidance coping) while examining the cultural
differences in SWB. In addition, Diener et al. (1995) identified several
factors not related to workplace that could explain differences in SWB
across nations.
Practical Implications
First, our study highlights that work locus of control and
family-work conflict are two important work-related concepts that differ
across the U.S. and Chinese cultures and that they mediate the
relationship between nationality and SWB. People with higher internal
work locus of control are more likely to take responsibility for the
outcomes in the workplace and work without close supervision (Spector,
1982, 1988). In our study we found a significant difference in the
values of work locus of control between the employees in the U.S. and
China. The U.S. employees reported higher internal work locus of
control. Therefore, for U.S. companies operating in China, the U.S.
managers may have to modify their style in order to be effective in
dealing with employees in China. For example, they may have to more
explicitly communicate the importance of employee behavior (as compared
to extraneous factors beyond control) for work performance. We also
found that internal work locus of control was positively related to
subjective well-being. It implies that there may be some benefits of
including work locus of control as one of the personality traits for
employee selection. Similarly, based on the role of family-work conflict
in mediating the effect of nationality on SWB, organizations that
provide the work-family balance to minimize the family-work conflict are
more likely to have happier employees.
Our findings indicate that though there may be differences between
the U.S. and China on WLOC and family-work conflict that explain part of
the differences in SWB, there are similarities in the mechanisms through
which WLOC and family-work conflict affect SWB. That is, whether in the
U.S. or in China, social support and active coping serve a similar role
in enhancing SWB. Therefore, organizations that foster a culture where
the shared norm is to support one's co-workers are likely to find
positive effect on employee well-being. Similarly, employee relationship
management and employee assistance prog