The Impact of Supervisor Support on the Job Satisfaction of Immigrant and Minority Long-Term Care Workers

Previous research has demonstrated the effect of supervisor support on the job satisfaction of long-term care workers (LTCW); however, much less is known about how this effect differs among race/ethnicity or immigration groups. We examined how supervisor support mediates the associations between race/ethnicity, immigration status, and job satisfaction among nursing assistants (NAs). Data of 2749 NAs were extracted from the National Nursing Assistant Survey (2004). Findings indicated that NAs of non-Hispanic Black and other races and immigrant workers were more likely to be satisfied with their jobs compared to White and non-immigrant workers, and the associations were mediated by NAs’ perceived supervisor support. Minority or immigrant LTCW may be more sensitive to supervisory support and more grateful if they received support from supervisors. Managers should be aware of these racial differences and by being supportive they may improve NAs job satisfaction and reduce turnover rates.


Introduction
The need for long-term care workers (LTCW) will grow significantly as the American population ages (Center for Excellence in Assisted Living, 2015). According to national population projections, the population aged 65 and over will reach 83.7 million by 2050, almost double the population for 2012 (Health Resources and Services Administration [HRSA], 2013) and it is estimated that approximately 70% of people will need Long-Term Support Services (LTSS) at some point after the age of 65 (Health Resources and Services Administration, 2013). Therefore, the recruitment and retention of long-term care workers is a critical issue. Foreign-born individuals make up a substantial percentage of the LTC workforce, and one out of every five long-term care (LTC) employees was born abroad (PHI, 2018); in addition, approximately 56% of LTCW are minorities (PHI, 2018). This study aimed to examine the associations between race/ ethnicity, immigration status, and job satisfaction, and whether supervisory support mediates these relationships.

Job Satisfaction of Long-Term Care Workers
Job satisfaction is conceptualized as a worker's evaluation of his/her job or job experiences and employees either express satisfaction in their job or dissatisfaction (Parvin & Kabir, 2011;Weiss, 2002). Poor working conditions have resulted in job dissatisfaction among long-term care workers (Castle et al., 2006(Castle et al., , 2007Gil, 2021;Jang et al., 2017). Job satisfaction is seen as an important issue because it has been linked to high turnover levels among LTCW, as well as other negative outcomes, such as "absenteeism, low performance, lower morale, low contribution to the team, less coordination, and less orientation towards an organizational objective" (Varma, 2017). Research supports that nursing home aides with low levels of job satisfaction are more likely to turn over within 1 year (Castle et al., 2007;Donoghue, 2010;Jang et al., 2017). The job satisfaction experienced by LTCW is also important for the well-being of not only the workers but also of the residents and families they serve (Hawes, 2020). Research has found that work environments with higher employee satisfaction provide better service to patients (Plaku-Alakbarova et al., 2018). Given the projected demand for long-term care workers, job satisfaction is a timely and important issue.

Supervisor Support and Job Satisfaction of Long-Term Care Workers
The social environment in the workplace may impact the job satisfaction of long-term care workers (Woodhead et al., 2016). Within the theory of organizational social capital, positive social relationships are a means to produce individual and community goals (Bourdieu, 1983a;Putnam, 1993Putnam, , 2000. For example, social relationships can help individuals to access the resources (i.e., information) that are needed to accomplish one's job (Adler & Kwon, 2002;Baker, 2007;Gittell, 2008). Workplaces with strong, supportive social networks are more likely to have employees with higher levels of job satisfaction (Hombrados-Mendieta & Cosano-Rivas, 2013). An important part of the social working environment in the workplace is the relationships employees have with their supervisors. Employees with strong supervisory support tend to feel more confidence in taking appropriate risks, exploring new ways of doing things and having individual power to explore goals (Paterson et al., 2013). Research has established that supervisory support can help achieve feelings of community (Fukui et al., 2019), belonging (Okolie et al., 2021), recognition (Baqir et al., 2020), and intrinsic rewards (Stumpf et al., 2013). This is consistent with the theory of organizational social capital in that individuals within the community are empowered and supported by their supervisor to accomplish tasks they would otherwise have less confidence to do.
However, not all individuals or social groups benefit equally from organizational social capital. Some groups may have less social capital than others, due to the social position of individuals within the group; in addition, individuals may be excluded from the social networks through which organizational social capital is accessedresulting in unequal access to social resources (i.e., social capital is not built up to the same extent by all employees) (Bourdieu, 1983b). Thus, the social networks that people participate in will determine the extent to which they can draw on social capital. This study assumes that minority and immigrants' job satisfaction can differ from White and USborn workers' job satisfaction because of the different social experiences immigrant and minority workers may encounter at work.

Race and Ethnicity Status, Supervisor Support, and Job Satisfaction
There are mixed findings on the relationship between race and ethnicity and supervisor support. Researchers have observed that in collectivist cultures employees tend to have a strong sense of loyalty to their supervisors (Moyes et al., 2007). Some research indicates that having close relationships with supervisors is valued particularly high among the Latino population (Chen, 2001). Yet, other studies have found that Black and White workers and Hispanic and non-Hispanic workers did not differ in their job satisfaction or their opinions about supervision (Wright & Saylor, 1992). No research has examined these relationships within the longterm care workforce.

Immigration Status, Supervisor Support, and Job Satisfaction
Studies have examined the association between social support drawn from supervisors and job satisfaction among immigrant workers employed in numerous workforces including traffic enforcement (Baruch-Feldman et al., 2002), hospital personnel (Harris et al., 2007, and white-collar employees (Tierney et al., 2002), but no known studies have explored the relationship between supervisor support and the impact on immigrant LTCW. Furthermore, research among other workforce groups, related to social support and the job satisfaction of immigrant workers, has yielded mixed results. Some research suggests that social support in the workplace is highly valued among immigrants in seeking a sense of belonging and feeling respected (Sloane et al., 2010).
Other research suggests that there may be negative social interactions between immigrant employees and supervisors including more physical and psychological distress, and getting more negative performance evaluations.
Although some research has addressed race/ethnicity, immigrant status, and job satisfaction among certified nursing aides (CNAs), no models have been formed to depict how supervisor support may impact the relationships between race/ethnicity and immigration status and job satisfaction of long-term care workers. Based on the above discussion, our research questions and hypotheses are listed as follows: Research Questions: 1) What are the direct relationships between race, ethnicity, and immigration status on job satisfaction? 2) Does higher supervisory support explain the relationships between certain racial/immigrant groups and job satisfaction?
Hypothesis 2a: Organizational social capital via supervisory support mediates the relationship between immigration status (i.e., being an immigrant) and job satisfaction.
Hypothesis 2b: Organizational social capital via supervisory support mediates the relationship between race/ethnicity (i.e., racial-ethnic minority status) and job satisfaction.

Data Source and Sample
For this research, data is extracted from the National Nursing Assistant Survey related to CNAs' demographics, race/ ethnicity, immigration status, job satisfaction, and supervisor support. The NNAS (2004) is the first and latest national survey (n = 3017) of certified CNAs employed in nursing facilities. The NNAS data is used by researchers studying CNAs and their work environments (Adams, 2010;Bishop et al., 2008;Sloane et al., 2010;Temple et al., 2009). The survey evolved out of growing concern about the high turnover of CNAs . Survey questions were developed to provide data regarding demographics, family life, education/training, as well as work life, including job satisfaction, for researchers interested in studying this workforce. Of the 3017 completed interviews, those who had left their jobs at the time of the interview were not included in the analyses (n = 120), reducing the sample to 2897. Due to missing values, the final sample size for this study was 2749.

Measures
Job Satisfaction. Job satisfaction was measured using CNAs' responses to the question: "overall, how satisfied are you in your job, are you…?", which had four response categories: extremely satisfied, somewhat satisfied, somewhat dissatisfied, and extremely dissatisfied. We created a dummy variable, with 1 = extremely or somewhat satisfied, and 0 = extremely or somewhat dissatisfied. Cases with "Don't know" or "Refused" were coded as missing.
Immigration Status. Immigration status was categorized into two groups: US-born and immigrants. Participants answered the following questions: "Are you a citizen of the United States?" and "Were you born a citizen of the United States, or did you become a citizen of the US through naturalization?" Participants who were not a citizen of the US or became a citizen of the US through naturalization were categorized as immigrants.
Race/Ethnicity. Race is a single-item variable categorized by responses to the following question: "I'm going to read you a list of…race categories. Please choose one or more races that you consider yourself to be." In addition, the NNAS also asks interviewees to distinguish if they are Hispanic or Latino/Latina (yes or no). Responses to these two questions were coded as 1 = Non-Hispanic white, 2 = Non-Hispanic African American, 3 = Non-Hispanic Other (the "other" category included Asians, American Indians, Alaska natives, Native Hawaiians, and Pacific Islanders), and 4 = Hispanic.
Supervisory Support. The NNAS has a series of 10 questions assessing the quality of the supervisor (whether the supervisor provides clear instructions, treats all NAs equally, deals with NA complaints and concerns, is open to new ideas, is supportive of progress in the NA's career, helps NAs with tasks when needed, listens when the NA is worried about resident's care, supports NAs working in teams, disciplines NAs not performing well, and tells the NA when doing a good job); responses to these 10 questions were coded as 1 = strongly disagree, 2 = somewhat disagree, 3 = somewhat agree, and 4 = strongly agree. Mean scores of the 10 items were calculated. Final scores were treated as missing values if more than five items were missing. Score ranged from 1 to 4, with higher scores indicating higher supervisory support.
Control Variables. Control variables included in this research include gender, primary language, level of education, household income, marital status, and age. Gender was dichotomized into responses of "male" or "female." For primary language, respondents answered, "what do you consider to be your primary language?" Responses to this question were coded as 0 = Spanish/other and 1 = English. Education and household income served as proxy measures of socioeconomic status. Education was included in the model as actual years of education completed and was split into two categories, a high-school diploma or less, or more than a highschool education. For household income, participants were asked to indicate the range that best described their total income in the past year. We combined those who had income greater than $40,000 into one category as the distribution was positively skewed; thus, this variable included five categories: less than $10,000, $10,001-$20,000, $20,001-$30,000, $30,001-$40,000, and more than $40,000. Marital status was dichotomized (0 = not married, 1= married/living with partner) based on answers to the following question: "Are you currently married, living with a partner in a marriage-like relationship, separated, divorced, widowed, or have you never been married?" Experience of discrimination was asked by "On your current job, have you ever been discriminated against because of your race or ethnic origin?" Answers were coded as 1 = Yes, 0 = No.

Analytic Strategy
SPSS 27 was used to analyze the dataset. Univariate analyses (frequencies and percentages) were conducted for all variables to provide a detailed description of the sample. Bivariate analyses, including analysis of variance (ANOVA), independent sample t-test, and chi-square test, were conducted to investigate the individual relationships amongst variables of interest. Further, a series of binary logistic regression models were conducted to examine the associations between race/ethnicity/citizenship and job satisfaction. First, we examined the extent to which the independent variables (race/ethnicity or immigration status) predicted job satisfaction, controlling for covariates. The second set of models was adjusted to add the mediator (i.e., supervisory support). The PROCESS macro v.3.5.3 for mediation analysis (Hayes, 2022) was used to test the mediation effect of supervisory support in the associations between race/ethnicity and job satisfaction, and between immigration status and job satisfaction among the NAs. Direct and indirect effects and 95% confidence intervals (CIs) were estimated using 5000 bootstrapped samples. Finally, interactions between race/ethnicity and immigration status were conducted, we generated eight categories (4 races/ethnicities × 2 immigration statuses), and examined their associations with job satisfaction.

Sample Characteristics
Univariate analyses (frequencies and percentages) were conducted for dependent, independent, and control variables to provide a detailed description of the sample. Table 1 shows the respective means, standard deviation, and percentages for the variables of interest. 80% of the sample reported extremely or somewhat satisfied with current job. Supervisor support responses averaged to 3.18 (out of 4). The sample was predominately composed of those who were US-born citizens (86.9%), with 13.1% of respondents identifying as immigrants. Examining the racial/ethnic breakdown of the sample, 54.7% of respondents were White, 29.5% were Black, 5.8% were Non-Hispanic Other, and 10% were Hispanic. Examining sociodemographic variables, the highest percentage of workers was between 26 and 40 (36.8%), followed by 30.3% of workers between the ages of 41 and 55, 23.6% of the sample between the ages of 16 and 25, and 9.3% of workers surveyed were aged 56-65. 92.3% of the sample was female, and 51.3% were married. The majority (75.5%) had a High-School education or less. Examining income, 42% of the sample had a household income below $20,000, whereas only 16.4% of respondents had a household income above $40,000. Most respondents (91.9%) spoke English as their primary language. Seven percent of the respondents reported being discriminated against due to their race/ethnic origin on their current job. Table 2 describes bivariate analyses of main study variables. Supervisor support and job satisfaction are examined by race/ethnicity groups and immigration status. Results from ANOVA suggested significant between-group differences for race/ethnicity on supervisor support (F = 4.79, p < 0.01). Tukey's HSD post-hoc test showed that NAs from non-Hispanic other races reported significantly higher supervisory support than non-Hispanic White NAs. Examining group differences for supervisor support using independent sample t-test, immigrant workers reported higher ratings than US-born workers (t = 4.97, p < 0.001). Chi-square tests were employed to examine the bivariate associations between job   satisfaction and race/ethnicity/immigration status. There was no between-group difference for race/ethnicity or immigration status on job satisfaction. Table 3 reports binary logistic regressions for associations of results of race, ethnicity, and job satisfaction mediated by supervisory support. In model 1, we examined the association between race/ethnicity and job satisfaction, individual characteristics of the employees (age, gender, education, income, marital status, primary language, and felt discrimination) were kept under control. Results showed significance for non-Hispanic Other employees (OR = 2.03, p = 0.005). Specifically, findings indicate that non-Hispanic Other workers were 103% more likely to be satisfied at work compared to White workers. Results were marginally significant for non-Hispanic Black workers (OR = 1.22, p = 0.094), and indicated that these workers were 22% more likely to be satisfied at work compared to White workers. In addition, Model 1 indicates that workers in the age groups 16-25 were more likely to report job satisfaction compared to those in the 26-40 age range (OR = 1.42, p = 0.009). Examining income, workers making $20,001-$30,000 were less likely to report job satisfaction compared to the reference income group of less than $10,000 (OR = 0.69, p = 0.039). Meanwhile, workers who reported being discriminated were less likely to be satisfied with their current job (OR = 0.36, p < 0.001).

Multivariate Analysis
In Model 2 of Table 3, the supervisor support variable was added. The associations between non-Hispanic other race and job satisfaction was no longer significant. The mediation analysis using the PROCESS macro showed that the direct effect of non-Hispanic other races on job satisfaction was 0.5 (p = 0.065; 95% CI: À0.03-1.03), and the indirect effect of non-Hispanic other race on job satisfaction through supervisor support was 0.27 (95% CI: 0.11-0.44), 33% of the total effect. Thus, there was evidence that supervisor support mediated the relationship between Non-Hispanic other workers and job satisfaction. Table 4 reports the binary logistic regression results for the association between immigration status and job satisfaction mediated by supervisory support. Again, in the first step, job satisfaction was regressed on immigration status, controlling for individual characteristics (age, gender, education, income, marital status, primary language, and discrimination). Immigrants were 75% more likely than US-born workers to be satisfied with their current job (OR = 1.75, p = 0.009). But when adding supervisor support to the model (model 2, Table  4), this variable no longer influenced job satisfaction (OR = 1.24, p = 0.356). The mediation analysis showed that the direct effect of immigrant status on job satisfaction was 0.22 (p = 0.365, 95% CI: À0.24-0.67) and the indirect effect through supervisory support was 0.34 (95% CI: 0.21-0.48), 61% of the total effect. This indicates that supervisor support mediated the relationship between immigration status and job satisfaction.
In Table 5, we combined race/ethnicity groups and immigration status, and generated eight categories (non-Hispanic White US-born citizens, non-Hispanic White immigrants, non-Hispanic Black US-born immigrants, non-Hispanic Black residents, non-Hispanic Other US-born citizens, non-Hispanic Other immigrants, Hispanic citizens, and Hispanic immigrants). We examined the associations between these categories and job satisfaction. Controlling for covariates, non-Hispanic Black immigrants (OR = 1.89, p = 0.016) and non-Hispanic other race immigrants (OR = 3.66, p = 0.003) were more likely to be satisfied with their current job comparing with non-Hispanic White US-born citizens.
In sum, the results indicate that NAs of other races or were immigrants were more likely to be satisfied with their current job, especially, Black immigrants and immigrants of other races were most likely to be satisfied with their jobs. Also, supervisory support mediated the associations between employees who are other races and job satisfaction and immigrant status and job satisfaction.

Discussion
Given the rapidly growing population of older adults and the increasing likelihood that Americans will need long-term services and supports, understanding the factors that influence job satisfaction among the long-term care workers who provide these services has important implications for policy and practice. The current study contributes to the existing knowledge on the factors that impact job satisfaction of longterm care workers in several ways. Job satisfaction in the workplace is influenced by many factors, as documented in previous studies (Bishop et al., 2009;Decker et al., 2009;Wiener et al., 2009). In particular, the social environment in the workplace is likely to affect the job satisfaction of longterm care workers (Woodhead et al., 2016). In line with previous research that highlights the importance of positive social relationships at work (Adler et al., 2008), our findings support research that specifically emphasizes the importance of supervisor support in creating a positive work environment (Bishop et al., 2009).
Considering the current and projected shortage of longterm care workers, one solution is to recruit and rely upon foreign-born workers as well as racial and ethnic minorities to fill this gap. Therefore, this study captured the unique characteristics and experiences of immigrant and minority workers by exploring potential racial/ethnic and immigrant differences and examining how supervisor support may predict job satisfaction of these workers. Our hypotheses are based on previous findings that race/ethnicity may influence job satisfaction among long-term care workers via social relationships at work (Chung-Bridges et al., 2015;Zakoscielna, 2013).

Race/Ethnicity
Regression analysis results from this study showed NAs of other races (i.e., Asians, American Indians, Alaska natives, Native Hawaiians, and Pacific Islanders) reported higher levels of supervisory support and were more likely to be satisfied with their jobs comparing to non-Hispanic Whites. These findings provide support for research that shows that collectivist culture employees tend to have a strong sense of loyalty to their supervisors. Collectivist cultures emphasize the needs and goals of the group, so it is logical that interconnectedness at work would be particularly important (Triandis, 2001). Most Asian, American Indian, Alaska Native, Native Hawaiian, and Pacific Island cultures are predominantly collectivist in nature. Furthermore, this research supports findings from research by Sloan and colleagues that showed that Black workers report significantly more positive emotions in the workplace than White workers (Sloan et al., 2013). Sloan and her co-authors also found that giving social support to coworkers was associated with increased levels of positive emotion among Black workers.

Immigration Status
Directly supporting the proposed hypotheses, we found that supervisory support explained the relationships between immigration status and job satisfaction. These findings establish that social support in the workplace is particularly necessary for immigrants and social support in the workplace is highly valued among immigrants in seeking a sense of belonging and feeling respected (Sloane et al., 2010). Perhaps this is due to their expectations to the working environment being lower, so they are more satisfied. They may also have limited employment opportunities and increased vulnerabilities such as language, and cultural barriers might add to the emphasized importance of supervisory support which in turn, would lead to higher job satisfaction. Interestingly, our findings also indicated that non-Hispanic Black immigrants and non-Hispanic Other immigrants were more likely to be satisfied with their current job compared to non-Hispanic White US-born citizens. This further confirmed our conclusion in the above paragraph, that minority immigrants may have the most limited choices in job hunting, and therefore they are more inclined to appreciate their job opportunities. This finding suggests that immigration and minority status is substantial in shaping job satisfaction as predicted, and specifically that supervisor support within the workplace does work differently based on race/ethnicity and immigration status.
Supervisory relationships that are supportive play a role in the job satisfaction of long-term care workers, in particular some minority (i.e., Black and non-Hispanic other) and immigrant long-term care workers. These findings therefore imply that the role that managers and supervisors play in making sure an organization fosters shared values and provides a foundation for communication within the organization is important to LTCW job satisfaction. To address this important concern, respect and acknowledgment of employees should be nurtured, and supervisors should display an inclusive attitude toward all workers, despite nationality.

Policy and Practice Implications
As illustrated by this manuscript and supported by a number of other studies, LTCW are a group that warrants particular attention, especially as a growing number of paid workers will be needed to provide LTSS in the upcoming decades. A wide variety of possible factors are associated with job dissatisfaction within this workforce.
Regardless of race/ethnicity or immigration status, CNAs have low-status employment, coupled with poor work conditions, challenging job tasks and low wages (Zakoscielna, 2013). Tailoring existing workplace environments to promote job resources among LTCW is a practical way to reduce the negative and costly consequences of stress within the workplace. The key policy and practice implications for these findings are most appropriately discussed in relation to recruiting and maintaining satisfied LTCW, especially minority and immigrant workers. As the field of LTSS is faced with declines in the number of native-born workers willing to work, several countries, including Canada, the U.K., and the Netherlands, are adopting workforce development strategies that feature the recruitment of foreign-born individuals (Stone & Bryant, 2019). This study suggests that managerial practices specifically aimed at fostering social capital among immigrant workers in addition to non-Hispanic other (i.e., Asian, American Indian, Alaska native, Native Hawaiian, and Pacific Islander) and non-Hispanic Black workers would be beneficial to their job satisfaction. When leadership invests in immigrant CNAs, job satisfaction of these workers will likely improve.

Study Limitations
This study has several limitations. First, this is a secondary data analysis. Although there is an advantage to having access to such a large national sample, this study was limited to only those variables available in the national survey. For example, job satisfaction was measured using only one applicable question from the survey. Ideally, this construct would be composed of more indicators and survey questions to develop a more detailed scale of job satisfaction. Furthermore, the NNAS was administered in 2004, and the age of this dataset is also a limitation. There have been no updates of this national study, so it was not possible to analyze more recent national data on NAs. Current anti-immigrant rhetoric and policy changes since the time of the surveys may make current NAs reluctant to share immigration status, therefore reducing concerns about the time lag between the data collection and analysis. Finally, the dataset has individual-level data only, with no way to link these individual data with facility-level information that might further contextualize the work environment, such as information about a facility's overall racial/ ethnic makeup. Therefore, analysis regarding the breakdown of race/ethnicity or immigration status within each organization, which would have been useful in assessing the level of social support a minority worker might have within their organization, cannot be determined.

Future Research
As previously described, the amount of research being conducted on the effects of race/ethnicity and immigration status on supervisor support and the impact on job satisfaction is limited, making the field for future research wide open. Future research that examines race/ethnicity and immigration status and organizational social capital that would merge with facility-level data to incorporate the specific characteristics of the organization would be beneficial. This would allow further examination assessing the level of social support a minority and/or immigrant worker might have within their organization. In addition, a further examination of how LTSS administrators could ensure that all racial/ethnic and immigration groups perceive strong levels of supervisor support is recommended. Qualitative research is recommended to further examine how racial/ ethnic and immigrant groups might process supervisor support differently.

Conclusion
In sum, this study supports prior research that supervisor support is important to improving job satisfaction. Management should pay attention to minority and immigrant workers employee's satisfaction. By understanding relationships with supervisors and using them to guide nursing home management practices, we can improve CNA job satisfaction, and decrease staff turnover rates. This would not only be beneficial to long-term care workers but also to nursing home residents and to long-term care in its entirety.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.