Abstract
Despite the influx of immigrants in the American workplace, little is known about their well-being. The authors built on literature pertaining to gender-specific stressors and organizational support theory to examine a model of stress for immigrants. Analysis of a national, archival data set (N = 150) demonstrated that, consistent with research involving U.S. citizens, immigrant women reported higher stress than immigrant men. The authors extended the literature, however, by showing that the sex–stress relationship is not straightforward for this population. Specifically, supervisor support moderated the effect of sex such that male–female stress differences were twice as large when supervisors were less supportive.
Despite the influx of immigrants in the American workplace, little is known about their well-being. The authors built on literature pertaining to gender-specific stressors and organizational support theory to examine a model of stress for immigrants. Analysis of a national, archival data set (N = 150) demonstrated that, consistent with research involving U.S. citizens, immigrant women reported higher stress than immigrant men. The authors extended the literature, however, by showing that the sex–stress relationship is not straightforward for this population. Specifically, supervisor support moderated the effect of sex such that male–female stress differences were twice as large when supervisors were less supportive.
Stress, the experience of mental or physical tension, is a universal construct in that it affects everyone. Although it results from a variety of sources, many of which are not work related, employee stress levels are of prime concern to organizations because they influence important organizational outcomes like employee well-being, burnout, counterproductive work behavior, intentions to quit, employee health, and health care costs (Chen & Spector, 1992 ; Colligan & Higgins, 2005). Stress also has clear career management implications for individuals because of its deleterious effects on vocational decisions such as the desire to change professions, career choice satisfaction, perceived career outlook, and even career advancement (Heponiemi et al., 2009 ; Hutri & Lindeman, 2002 ; Liang & Hsieh, 2005 ; Yang, Che, & Spector, 2008). Furthermore, the relationships between stress and career-related outcomes are robust, as they have been examined using vastly different samples (e.g., students and retirees; Reynolds & Constantine, 2007 ; Robson & Hansson, 2007) across various countries (Heponiemi et al., 2009 ; Yang et al., 2008) with similar work-related outcomes.
Although research on stress in the organizational sciences is relatively commonplace, significant gaps remain in the literature. For example, little is known about demographic differences in employees’ stress levels (Liu, Spector, & Shi, 2008). Furthermore, there is limited research on the experience of stress by immigrant workers in the United States, which is important because the influx of immigrant workers in the past decade has resulted in immigrant workers comprising roughly 15% of the U.S. workforce (Camarota, 2005). Although immigrant workers are a part of the collective U.S. workforce, their stress levels might be influenced by factors purely related to their immigrant status, thereby making their experience of stress different from that of nonimmigrants. For instance, visa issues, emotional and social adjustment, and integration of immigrants into American society are issues that are unique to immigrants. Our current knowledge of stress in the American workforce does not account for the abovementioned sources of stress and, hence, cannot be generalized to the immigrant population. Therefore, it is important to consider these unique sources of stress within the immigrant population in addition to other sources of stress that are characteristics of the job itself.
Of the limited research in this area, one study suggests that immigrants experience elevated levels of stress (relative to citizens) as a result of a combination of factors such as adjustment to the host country, social support, and job-related factors (Kennedy & McDonald, 2006). Although it is true that immigrants from different cultures and countries may not face the same set of challenges while living and working in the United States, there are other challenges that appear universal across immigrants, regardless of culture. For example, all immigrants at a given point in time deal with issues related to their residential status in the United States (e.g., visa, green card, etc.). Often, these issues are tied to their sponsoring employers, meaning that the residential status of immigrants in the United States is dependent on the willingness of the employer to retain the immigrant employee by supporting his or her immigration application. As a result, the pressure on the immigrant employee to perform well and retain employment is higher than on U.S. natives (de Castro, Gee, & Takeuchi, 2008). Other challenges that are likely to be universal for immigrants are adjustment (social and psychological), assimilation into different systems such as financial, medical, and educational systems, and acceptance by host nationals into mainstream society and work-related groups. Despite the prevalence of these immigrant-specific stressors, little is known about the specific sources of stress, the boundary conditions affecting the relationships between organizational stressors and the experience of stress, or sex differences in the experience of stress among immigrant workers, and how they influence overall well-being (de Castro et al., 2008).
If we consider the idea that most immigrants choose to migrate to another country for better opportunities (e.g., employment, education, and lifestyle), it is reasonable to assume a good deal of spillover between life and work stress within the immigrant population. This, in conjunction with the fact that not all immigrant-specific stressors are work related, leads us to examine general stress as opposed to limiting our focus to that attributable to workplace stressors alone. We address the gaps in the literature on immigrant employees by exploring whether (a) commonly observed sex differences in U.S. citizen samples generalize to this population and (b) supervisor support serves as a moderator of the prospective sex–stress relationship among immigrants. In doing so, we build on literature summarizing gender-specific sources of stress to explain differential effects of sex in the proposed model and apply organizational support theory (Eisenberger, Huntington, Hutchison, & Sowa, 1986) to explicate the moderating role of supervisor support in the sex–stress linkage. We begin by discussing sex differences in the experience of stress. This is followed by a description of the proposed boundary condition representing the heart of the study’s theoretical contribution to the literature.
Sex Differences in the Experience of Stress
There are many reasons for sex differences in employee stress levels. For instance, Jick and Mitz (1985) described three sources of differential responses to stressful situations between men and women: (a) genetic (i.e., inherent tendencies to behave and perceive work- and life-related events in certain way), (b) structural (i.e., “glass ceiling effect,” therein invisible factors such as gender-based stereotypes prevent women from career advancement), and (c) social (i.e., cognitive appraisals of stressful environments are shaped by gender-specific roles, expectations, and norms). As genetic sources are difficult, if not impossible, to change, we focus on the latter two sources to explain why women likely report higher levels of stress than their male counterparts.
Research on sex differences has shown that (a) women report greater stress than men (Davis, Matthews, & Twamley, 1999) and (b) the sources of this stress (i.e., stressors) differ for the two sexes (e.g., McDonough & Walters, 2001). Liu, Spector, and Shi (2008) suggested that female workers tend to experience more strain and depression as a result of greater sensitivity to job stressors than their male employees. Similarly, Jick and Mitz (1985) posited that women experience lower levels of emotional and mental well-being as compared to men and are more restricted in their activities due to minor health ailments such as headaches. Conversely, male workers tend to engage in more drinking behavior and are more prone to coronary heart disease than their female counterparts. Furthermore, female workers in general tend to enter the workforce with higher levels of stress than male workers because of factors such as lower salary ranges, higher expectations in a “male-dominated” workforce, and elevated standards set by themselves and others for their performance. Research also supported the greater prevalence of burnout and general stress among women than men, largely as a result of interpersonal issues, work–family conflict, social life, role conflict, and role ambiguity (Liu et al., 2008 ; Rupert & Morgan, 2005). Liu et al. further explain that the social construction of gender in society places expectations on women that focus on interpersonal relationships and rendering of emotional support to friends, family, and colleagues. This heightens women’s role responsibility in and outside the workplace, thereby elevating their stress levels. In sum, research on sex differences in stress supports the idea that women experience higher levels of stress than men because of the value they place on social and interpersonal issues, work–family conflict, and role conflict and ambiguity. To date, however, the stress literature has not investigated this phenomenon in the immigrant population. Consequently, we hypothesize the following:
Hypothesis 1: In general, female immigrant employees experience higher levels of stress than male immigrant employees.
The Role of Supervisor Support
Although there is some evidence for the differential experiences of stress between men and women, researchers have not provided much insight concerning what employers can do to minimize these differences. Although organizations may not be able to manage and control all of the different sources of stressors that their employees experience, they may endeavor to minimize the effects of these stressors. Accordingly, we examine supervisor characteristics (i.e., support) to determine their impact on stress differentials between men and women. We explore supervisor support as a boundary condition in the sex–stress relationship to emphasize its importance as an organizational factor that may be managed to help attenuate differential experiences of stress among male and female immigrant workers.
Organizational support theory may prove useful in explaining the moderating role of supervisor support in the proposed model (Eisenberger et al., 1986 ; Rhoades & Eisenberger, 2002). According to this theory, employees tend to attribute humanlike characteristics to the organization; thereby forming perceptions of support from the organization toward themselves (Eisenberger et al., 1986). Furthermore, employees tend to perceive support or the lack thereof, as a function of how they are treated by individuals representing the organization (e.g., supervisors; Shanock & Eisenberger, 2006). This theory describes the employee–employer relationship as a social exchange wherein both parties have certain expectations about each other’s obligations and commitments (Eisenberger et al., 1986 ; Robinson, 1996). Notably, these exchanges are built on implicit and subjectively defined obligations. Incumbents do their employers a favor (e.g., engaging in organizationally valued behaviors such as helping others), hoping that their favors will be repaid by the organization (Blau, 1964 ; Cropanzano & Mitchell, 2005). Social exchange theory emphasizes the importance of norms of reciprocity, with employees and the employer being aware of their own and the other party’s commitments and obligations (Cropanzano & Mitchell, 2005 ; Gouldner, 1960). Social support is thought to fulfill employee socioemotional needs such as the needs for self-esteem, approval, and affiliation (Armeli, Eisenberger, Fasolo, & Lynch, 1998). Because of the implicit and subjective nature of the social exchange, violation of these expectations has negative consequences for employee well-being and other organizationally relevant outcomes (Zhao, Wayne, Glibkowski, & Bravo, 2007). More specifically, lack of support has been linked with more negative outcomes for both male and female employees (Vermeulen & Mustard, 2000).
Research is consistent with these theoretical assertions. In particular, employee perceptions of support are linked to a variety of individually and organizationally important outcomes. For example, perceived organizational support (POS) and perceived supervisor support (PSS) are associated with improved task and contextual performance (Shanock & Eisenberger, 2006), lowered turnover (Eisenberger, Stinglhamber, Vandenberghe, Sucharski, & Rhoades, 2002), weakened depressive symptoms (Dormann & Zapf, 1999), increased job, family, and life satisfaction, and decreased stress, intentions to quit, and work-to-family conflict (Thompson & Prottas, 2006).
Although research supports the positive effects of POS and PSS on various work-related outcomes, it remains unclear what effects these variables have on sex differences in stress. Research on relative deprivation (Crosby, 1984) indicates that the effects of a resource tend to be more pronounced among those who have been denied access to (i.e., deprived of) that resource. Within the current context, a host of literature indicates that female employees perceive less support and more discrimination than their male counterparts (e.g., Avery, McKay, & Wilson, 2008 ; Ohlott, Ruderman, & McCauley, 1994). Due to this pattern, and women’s knowledge of it, they are prone to be especially sensitive to support from their organization or supervisor. Additionally, women appear to place higher value on social support and relationships as compared to their male counterparts, thus, making them more responsive to social support at work (Schwarzer & Gutiérrez-Doña, 2005). Accordingly, though higher levels of support generally may correspond in lower levels of stress, this linkage is probably especially pronounced among women. The result would be considerably smaller sex differences in stress when supervisors are supportive, which leads us to propose the following two hypotheses:
Hypothesis 2: Supervisor support is negatively related to stress.
Hypothesis 3: Supervisor support moderates the relationship between sex and stress levels such that sex differences are smallest for those with supportive supervisors.
Method
Subjects and Procedure
We used a part of the archival data from National Survey of the Changing Workforce (NSCW) conducted in 1997 by Louis Harris and Associates using the questionnaire developed by the Families and Work Institute (Bond, Galinsky, & Swanberg, 1998). Although this data set is 13 years old, we argue that it is relevant to study the current research questions due to 2 reasons. First, although the cultural and sociopolitical context for immigration has changed since 1997 (e.g., 9/11, passage of the Patriot Act, stricter immigration rules, etc.), the conceptual meaning of our research variables (e.g., supervisor support and psychological distress) has remained the same. Second, the NSCW (1997) data set has been used extensively in recent publications to examine a variety of workplace phenomena (e.g., dual careers, work–family interface, and person–job fit; Behson, 2005 ; Hall & MacDermid, 2009 ; Kalleberg, 2008 ; Winslow, 2005).
The NSCW researchers conducted telephone interviews throughout the United States resulting in 3,552 completed surveys, using stratified and unclustered random probability sample generated by random-digit-dial method. Participants were offered $20 for their participation. The survey primarily assessed work–family balance, job stress, organizational, and supervisor support. Men represented 50% of the sample. The mean age was 41 years (SD = 12.23). The overwhelming majority of respondents were American citizens with only 225 people (6%) reporting they were not U.S. citizens. People varied in the level of their education with 35% having high school diploma, 33% finishing some postsecondary degree, and 32% having 4-year college degree or higher. Most were legally married (56%), followed by single (38%) and those living with a partner (6%). Respondents varied in their occupations: 25% employed in production/operation/repair, 18% professionals, 16% managers, 14% administrative support, 12% sales, 11% service, and 4% technical.
For the current study, we used the sample of 150 participants. Specifically, the respondents were asked whether they were born in the United States or their parents were U.S. citizens. Two hundred twenty five people responded negatively to this question, and, thus, were categorized as immigrants. 1 We, then, eliminated those respondents that had missing data on the variables of interest. This strategy resulted in the usable data of 150 people. The subsample consisted of 51% of men with the mean age of 39 years (SD = 11.56). Among these foreign-born workers, 57% were naturalized citizens, 58% indicated that English was their main language, and 75% reported they were fluent in English. The respondents varied in their racioethnicity: 35% White, 28% Other, including mixed, 24% Asian, 12% Black, and 1% Native North American. In addition, 33% self-identified as Hispanic/Latino. Their length of residence in the United States was approximately 20 years (SD = 11.83). The respondents were reasonably well educated with 43% having 4-year college degree or higher, followed by those having high school diploma (29), and 28% of them finished some postsecondary degree. They held a variety of jobs: 22% employed as professionals, 21% administrative support, 18% employed in production/operation/repair, 12% managers, 11% service, 10% sales, and 6% technical.
Measures
The Families and Work Institute developed the measures for the NSCW (1997) data set (Bond et al., 1998). These measures are very similar to the existing and validated in the literature measures.
Supervisor support
We measured supervisor support with 9 items, each assessing the degree to which a supervisor values his or her subordinates’ contributions and cares about their subordinates’ well-being. The items are as follows: “My supervisor keeps me informed of things I need to do job well,” “My supervisor has realistic expectations of my job performance,” “My supervisor recognizes when I do a good job,” “My supervisor is supportive when I have a work problem,” “My supervisor is fair when responding to employee personal/family needs,” “My supervisor accommodates me when I have family/personal business,” “My supervisor is understanding when I talk about personal/family issues,” “I feel comfortable bringing up personal/family issues with my supervisor,” and “My supervisor cares about effects of work on personal/family life.” The respondents indicated their opinion using a 4-point scale ranging from 1 (strongly disagree) to 4 (strongly agree). The higher values indicated higher supervisor support. The internal consistency coefficient based on the full sample, including nonimmigrants (N = 3552), was comparable (α = .89) to the one reported based on the subsample of immigrants (α = .88).
Furthermore, Behson (2005) used the same NSCW (1997) data set, where he examined supervisor support (α = .89) and stress. Notably, he obtained a similar correlation between these two variables (r = −.23, p < .05) as we did (r = −.27, p < .001). In addition, the items comprising this scale are similar to those of a widely used scale of perceived supervisor support (sample item: “Help is available from my supervisor when I have a problem;” Eisenberger, Cummings, Armeli, & Lynch, 1997). Prior research provides validity evidence for the PSS. For example, Ng and Sorensen (2008) reported a correlation of .52 between PSS and job satisfaction. In our sample, we obtained a similar correlation between supervisor support and job satisfaction (r = .54, p < .001). In addition, supervisor support was not significantly related to tenure or gender (Ng & Sorensen, 2008). Similarly, in our research, there were nonsignificant relationships between supervisor support and tenure (r = −.01, p = .922) or supervisor support and gender (r = −.10, p = .242) in immigrant-based subsample. Likewise, in full sample, supervisor support was not significantly related to either tenure (r = −.02, p = .340) or gender (r = .02, p = .267). Together, this evidence provides further support for the convergent and discriminant validity of our measure of supervisor support.
Overall stress
We measured overall stress with 2 items capturing how often people have experienced psychological distress. The items include “Frequency of being bothered by minor health problems” and “Frequency of feeling nervous/stressed during the past three months.” The higher values indicated higher level of experienced stress. The respondents indicated their opinion using a 5-point scale ranging from 1 (never) to 5 (very often). The internal consistency coefficient based on the full sample, including nonimmigrants (N = 3552), was relatively comparable (α = .67) to the one reported based on the subsample of immigrants (α = .76).
In addition, this scale is similar to the existing scale of overall psychological stress (e.g., people indicate whether they feel “nerve-wrecking;” Stanton et al., 2001). In support of the convergent validity of our measure, overall stress was negatively related to job satisfaction and positively to lateness in the full data set (r = −.25, p < .001, r = .09, p < .001, respectively) and immigrant-based subsample (r = −.32, p < .001, r = .12, p = .158, respectively). These correlations are comparable to those obtained by Stanton and colleagues in a sample of aerospace employees (N = 4,322; 73% men, ranging in age from 25 to 29). The correlations between stress and job satisfaction ranged from −.10 to −.47 (for different version of the measure), and for stress and intent to quit (which is a withdrawal behavior as lateness) they ranged from .12 to .36. Furthermore, recent meta-analysis suggested that education and marital status were poor predictors of overall stress (Dowden & Tellier, 2004). Based on this meta-analytical evidence and supporting discriminant validity of our measure of overall stress, this construct was not significantly related to either marital status (r = −.01, p =.957) or education (r = .15, p =.077) in immigrant-based subsample. Similar correlations were obtained in full sample (r = .01, p =.604; r = .01, p =.737, respectively).
Additionally, to provide further support for the discriminant validity of the two constructs (i.e., supervisor support and overall stress), we tested two measurement models. Before testing these two models, we closely examined our items and identified two supervisor support items (i.e., “My supervisor is understanding when I talk about personal/family issues” and “I feel comfortable bringing up personal/family issues with my supervisor”) whose content overlaps substantially. In other words, these 2 items likely share variance due to their conceptual similarity (e.g., discussing personal issues with one’s supervisor). Accordingly, following Thompson’s (2004) recommendations, we correlated the residual variances of these 2 items. The first model combined the two scales to form one factor and the results indicated that the model fit the data poorly, χ2(43, N = 2652) = 1474.222, p <. 001, root mean square error of approximation (RMSEA) = .11, Confirmatory Fit Index (CFI) = .87, Tucker-Lewis Index (TLI) = .83, standardized root mean square residual (SRMR) = .06. We, then, tested the model with two factors (i.e., supervisor support and overall stress), which exhibited a significantly better fit, χ2(42, N = 2652) = 882.602, p <. 001, RMSEA = .09, CFI = .92, TLI = .90, SRMR = .04. Thus, the results of the confirmatory factor analysis provided further support for the discriminant validity of our research measures.
Controls
We used subordinates’ age and tenure as control variables because of their potential to account for variance in overall stress (Cheng & Chan, 2008 ; Pinquart & Sörensen, 2003). In the modern market place, job insecurity may substantially decrease people’s well-being (Strazdins, D’Souza, Lim, Broom, & Rodgers, 2004 ; Tennant, 2001). It is a unique stressor for immigrants’ health (Maiden, 2001). We included job security to demonstrate that the effects of sex-explained variance in the overall stress above and beyond job security. We measured job security with 1 item, assessing the “Likelihood of losing job and looking for new job in next two years.”
Results
Table 1 presents descriptive statistics and the correlation matrix. We conducted hierarchical, moderated multiple regression (MMR) analyses to test the hypothesis. We first centered the predictor variables (Aiken & West, 1991). At the first step, we entered the control variables—subordinates’ age, tenure, and job security. Of the control variables, only job security was significantly related to stress (B = −.22, p = .02; see Table 2 ). At the second step, we entered the main effects of sex and supervisor support.
Means, Standard Deviations, and Correlations Among All Variables
Note. N =150. *p < .05.
**p < .01.
Hierarchical Regression Analysis for Overall Stress
Note. N = 150. *p < .05.
**p < .01.
Consistent with Hypotheses 1 and 2, suggesting women and those perceiving less support would display greater stress, the effects of sex (B = .80, p <.001) and support (B = −.30, p = .03) were significant. At the third step, we entered the two-way cross-product term—Sex × Supervisor Support, which contributed unique incremental variance (ΔR 2= .02). In support of Hypothesis 3, supervisor support moderated the effect of sex on stress (B = −.51, p = .04). Specifically, sex differences were twice as large when support was low (B = 1.15, p < .001) compared to when it was high (B = .49, p = .03). These results, which are illustrated in Figure 1 , supported Hypothesis 3. Thus, all three of the study hypotheses received support.
Interaction between sex and supervisor support in predicting overall stress.
Supplemental Analyses
As immigrants of various ethnic backgrounds may have different workplace experiences (e.g., lowered supervisor support), we tested our model across different ethnic groups to detect any potential differences. Specifically, we reran the hierarchical, MMR analyses based on five different subsamples (i.e., White, Black, Asian, Other/mixed, and Hispanic). Given the very small sample sizes, we report only the unstandardized coefficients without associated pvalues and urge readers to interpret these results with caution. In general, the regression coefficients for both main effects and the Sex × Supervisor Support interaction were comparable across these subsamples. There were, however, notable exceptions. Specifically, the sex coefficients were similar for the White and Black groups (B = .36 and B = .37, respectively), whereas sex differences were more pronounced among Asian, Other/mixed, and Hispanic groups (B = .98, B = 1.46, and B = .85, respectively). Furthermore, the regression coefficients for supervisor support were virtually equivalent across White, Black, Hispanic, Asian, and Other/mixed groups (B = −.44, B = −.40, B = −.56, B = −.31, B = −.39, respectively). Finally, the interaction term’s regression coefficients varied considerably but were all in the same direction across the White, Black, Asian, Other/mixed, and Hispanic subgroups (B = −.29, B = −1.50, B = −.14, B = −.51, B = −.56, respectively).
Discussion
The purpose of this study was to examine the relationship between sex and stress in the context of immigrants in the workforce. In addition to the main effects, we examined the moderating role of supervisor support on this relationship. We applied the organizational support theory (Eisenberger et al., 1986 ; Shanock & Eisenberger, 2006) and research on gender-specific sources of stress (Jick & Mitz, 1985) to explain the proposed model. In support of our hypotheses, we found that, among immigrants, women experience greater stress than men. In addition, employees perceiving higher levels of support reported less stress. As hypothesized, however, these effects were conditional: Supervisor support moderated the effects of sex on stress. Stress differences were smaller when supervisors were seen as more supportive because the effects of support were more pronounced among women than men. Below we discuss implications of our findings.
Implications
From the theoretical perspective, our findings are consistent with prior research, indicating that women respond differently to stressors than men, resulting in heightened stress (Davis et al., 1999 ; Liu et al., 2008 ; Rupert & Morgan, 2005). We, however, went beyond this literature by showing that the same pattern of results holds for immigrants. Research on coping strategies suggests that men handle stressors by confronting a problem; whereas, women respond to stressors emotionally and by venting their problems with friends and family (Tamres, Janicki, & Helgeson, 2002). Such differences in coping strategies result in increased stress among women relative to men (Tamres et al., 2002). Although we did not measure coping strategies directly, our results are congruent with this research in that immigrant women reported higher levels of stress than immigrant men.
Furthermore, the results of our supplemental analyses indicated that women of Asian, Hispanic, and mixed racial origin are particularly susceptible to experience psychological strain. Furthermore, Hispanic women reported lower levels of supervisor support. These findings are consistent with the literature on cultural values and its impact on psychological values of Hispanic immigrants. Specifically, emotionally positive, supportive close familial relationships play an integral role in the life of Hispanic people, especially women, a phenomenon called familialism (Campos et al., 2008 ; Zea, Quezada, & Belgrave, 1994). For Hispanic immigrant women, the familial ties are likely to be weakened, resulting in increased psychological distress (Grzywacz et al., 2007) and, potentially, perceived lack of supervisor support. Similarly, Asian cultural values focus on building harmonious social bonds with their families, coworkers, and other groups (Lun & Bond, 2006). Not surprisingly then, having immigrated to the United States—one of the most individualistic cultures (Hofstede, 1996)—female Asian immigrants are prone to experience heightened levels of psychological distress because of a lack of group (e.g., family and community) support and cohesion (Lueck & Wilson, 2010). For immigrant women of mixed racial origin, they may experience elevated levels of psychological strain because of the social identity dilemma (e.g., pressure to self-identify with preexisting racial or social group; Miville, Constantine, Baysden, & So-Lloyd, 2005 ; Taylor, 2010). Future research should test these theoretical propositions explicitly.
In addition to the main effect, we examined a boundary condition of the linkage between sex and stress. Consistent with the organizational support (Eisenberger et al., 1986) and relative deprivation (Crosby, 1984) theories, our results indicated that supervisor support moderated the effect of sex on stress. Conjunctively, these two theoretical premises suggest (a) supervisor support is associated with positive individual and organizational outcomes because of the increased social exchange and (b) those traditionally deprived of it (e.g., women) exhibit stronger effects of support. In concert with these assertions, we found that supervisor support is particularly important for immigrant women (relative to men) in mitigating stress. Specifically, we showed that sex differences were half as large when working with supportive supervisors.
From the practical perspective, our results offer guidance for managers striving to improve the work experience of their subordinates. Specifically, these findings illustrate the importance of supervisor behaviors in helping immigrant employees deal with the stressors they face. Because immigrant women appear to be more sensitive to social support, it is likely that those who are highly stressed but receive support will show a significant decrease in overall stress. Although such support could come from organizations providing stress counseling services for their personnel, our findings suggest organizations pay more attention to the supervisory styles of managers to help mitigate immigrant employees’ stress in general and sex differences in stress in particular. For example, supervisors can be trained to provide emotional as well as career-related support to their subordinates regarding their professional issues at work. They also can offer emotional and instrumental (e.g., flextime) support when their subordinates experience family-related problems. Moreover, with regard to stress induced by immigrant status at work, organizations are likely to benefit from training supervisors to be aware of the unique issues experienced by immigrants. Often, nonimmigrant supervisors are not entirely informed about the stress and adjustment issues experienced by their immigrant employees, leaving them unprepared to provide the necessary support to help offset these stressors.
Especially in the current sociopolitical context of globalization, frequent changing of immigration policies, loss of employment due to the current economic recession, and hiring freezes for job seekers with visas, it is imperative that organizations provide assistance to help their immigrant employees manage stress related to these issues. Furthermore, it is likely that immigrant employees will experience discrimination or hostility from their nonimmigrant counterparts because of current job market conditions that (a) are a source of stress to all job seekers in general and (b) often increase intergroup (i.e., natives vs. immigrants) competition for what are perceived to be scarce resources (e.g., jobs, promotions, etc.). In such situations, supervisors may act to reduce the occurrence of such incidents by ensuring that immigrants do not become targets and all employees feel adequately supported.
For career counselors, our findings indicate the importance of having a thorough understanding of the stressors unique to female and male immigrant employees. By understanding these issues, counselors may help immigrant employees navigate the treacherous terrain they often face in their efforts to establish careers in the United States. For instance, counselors can (a) assess immigrant employees’ coping skills and help them to develop any shortcomings they identify in this regard, (b) aid in identifying supportive employers, and (c) provide assistance with handling stress-inducing visa and green card issues. Furthermore, helping them to locate or develop supportive environments and mentoring groups also may prove beneficial. New immigrant workers likely would benefit from being mentored by immigrant employees already established in the United States by gaining comfort and confidence, making their adjustment into the workplace easier. Finally, counselors within organizations could help establish diversity programs that allow nonimmigrant employees to interact with and understand the issues faced by their immigrant counterparts to help create a more harmonious work environment, thereby reducing the likelihood of acts of discrimination or hostility toward immigrant workers.
Limitations, Future Research, and Conclusion
Despite the strengths of our study (i.e., unique sample of immigrants), we should note several limitations. First, the cross-sectional design does not allow for causal conclusions to be drawn. Second, all data are self-reported, which makes it possible that common method variance affected our findings. It is important, however, to note that common method variance cannot account for interactions (Evans, 1985), which was the primary contribution of our research. Third, there was potential for self-selection bias, making it possible that the people who chose to participate significantly differed from those who were not interested in participating. An additional limitation of the study is that the sample included immigrants that were highly educated. Finally, the use of NSCW (1997) data set may have affected applicability of our results to the current workplace phenomena. Although the conceptual similarity of the variables and extensive use of this data set in recent publications (e.g., Hall & MacDermid, 2009) suggests some degree of applicability to the current workplace, we encourage future researchers to test our model in the modern marketplace.
The results of our study offer important avenues for future investigation. Organizations would benefit greatly from future research delineating the different sources of stress experienced by immigrant employees in and outside the workplace. It is possible that the context of the stressor experienced could influence its impact on immigrants’ stress levels. Other inquiry could include examining the effects of different supervisory styles, career counseling approaches, or workplace interventions on immigrant employee stress levels. For instance, participative leadership approaches could lead immigrants to feel more included, thereby helping to offset prospective feelings of isolation and exclusion. Alternatively, counseling strategies that prepare immigrant employees for likely cultural differences in supervisory approaches at work could help these individuals to maximize the impact of support provided by their managers. Finally, macro-level research in this area could consider the impact of organizational diversity training focused on enhancing awareness concerning immigrant issues, as it is possible that such interventions could moderate the effect of international diversity on organizational effectiveness (especially in companies containing a significant immigrant presence). Research on immigrant populations is highly valuable due to the growing use of immigrants in organizations here in the United States.
Immigrant populations make up a significant proportion of the U.S. workforce and more attention must be paid to their well-being at work. Workforce diversity is becoming a well-researched topic and, in the process, this ever-growing literature is proving its potential value to organizations (see Herring, 2009 for an excellent recent example). The findings presented here are of particular value to organizations because of the insight they provide into the experiences of immigrant workers who are a valuable yet understudied resource in today’s workforce. By identifying means of relieving stress, organizations can minimize undesirable outcomes (e.g., counterproductive work behaviors, burnout, and high turnover) among their immigrant employees, thereby creating a positive and productive work environment for all employees. In sum, it appears as though the answer to the question of whether your supervisor stresses you out among immigrants depends on the employee’s sex and the supervisor’s supportiveness.
Footnotes
The author(s) declared no conflicts of interest with respect to the authorship and/or publication of this article.
The author(s) received no financial support for the research and/or authorship of this article.