Introduction

The diverse experiences of older people are increasingly drawing the attention of gerontologists interested in achieving a more informed portrait of elders in the United States. It is now acknowledged that there exist “different worlds of aging” and that heterogeneity within social categories (i.e., health status among older blacks by social class) may be more insightful than comparisons made between categories (health status of older blacks compared to older whites) (Dilworth-Anderson & Burton, 1999; Henderson, 1994; Morgan & Kunkel, 2001). This study addresses the diversity of aging experiences by examining the effects of immigrant status, religious affiliation, and available resources on the well-being of Arab-American elders.

Arab Americans, Aging, and Well-being

Arab American refers to those individuals whose ancestors arrived from an Arab speaking country which include 22 nations in North Africa and west Asia. As a result, Arab Americans are quite diverse in terms of religion and national origin. Faiths include both Christian and Muslim. Lebanon is the number one country of origin for Arab immigrants to the United States, followed by Syria and Egypt (Brittingham & de la Cruz, 2005).

Being an elder and caring for elders is relatively new for Arab Americans who have for the most part been a relatively young cohort of immigrants (Zogby, 1990). Increases in the elder population and accompanying health related issues pose new areas of concern. A key reason to examine the associations between resources, religion, immigrant status, and well-being relates to the situation where Arab Americans now, more than ever, face critical judgments and stigmatized identities due to recent world events. Arab Americans are officially classified together with non-Hispanic whites. Yet, increasingly attention is directed toward them due to the events and aftermath of September 11, 2001. Stereotyped perceptions of Arab Americans become the primary means by which most of the United States understands Americans of Arab origin. Some writers refer to Arab Americans as an invisible minority in the United States (Salari, 2002; Samhan, 1999). The need for research that empirically documents lives of individuals with ancestry from an Arab-speaking country will provide important insights into the pragmatics of their attitudes and experiences. Their status as an immigrant/ethnic group shapes opportunity structures and access to resources. Two of the most critical areas to consider with regard to resources include human and social capital. The capital available to Arab-American elders via these two sources and whether they operate differently depending on immigrant status and religious affiliation constitute areas in need of examination.

Theoretical Framework

The rich theoretical and empirical tradition of well-being research has been extended in recent years to consider the special case of immigrants. Research on immigrant well-being addresses the experiences of those from Asian, Pacific Islander, Caribbean and Hispanic origins, demonstrating that immigrants appear to have better physical health than their U.S. born counterparts (Cho, Frisbie, Hummer, & Rogers, 2004; Frisbie, Cho, & Hummer, 2001; Rogers, Hummer, & Nam, 2000). Immigrant elders, on the other hand, are at risk of depressed mood or affect due to the stresses of immigration, acculturation, and high rates of family disruption (Aranda & Miranda, 1997; Mui, 1996; Plawecki, 2000; Wilmoth & Chen, 2003). Older immigrants from Arab-speaking countries, nevertheless, are not visible in the literature (see Salari, 2002).

Moreover, theoretical developments in well-being research increasingly look to the processes through which identified predictors influence such outcomes (Bisconti & Bergeman, 1999; Fernandez-Ballesteros, Zamarron, & Ruiz, 2001). The role of human and social capital as resources that may be differentially influenced by immigrant status and also distinctively linked to well-being has mainly received attention among younger immigrants (e.g. Rumbaut, 2005; Zhou & Bankston, 1998). Conceptualized as resources, human and social capital may constitute key explanatory mechanisms for understanding well-being outcomes (Coleman, 1990; Portes, 1998), however, little research specifically examines the link that both human and social capital may have to the association between immigrant status and psychological well-being for example, and rarely have researchers examined these resources as potential explanatory mechanisms among an older immigrant group. The objective of this study is to examine whether human and social capital constitute a pathway by which to understand well-being differences that may exist according to immigrant status and religious affiliation among older Arab Americans.

Immigrant Status

Adaptation among immigrants to a new cultural context often produces stress that negatively affects psychological well-being (Berry, Kim, Minde, & Mok, 1987). Yet some studies on immigration and psychological well-being among older age groups are equivocal in their findings. One set of findings suggest that elder Mexican American immigrants report more depression than U.S. born elders of Mexican descent (Black, Markides, & Miller, 1998; Gonzales, Haan, & Hinton, 2001). On the other hand, Cuellar, Bastida, and Braccio (2004) report no significant difference between immigrants and U.S. born Mexican Americans, or length of residence on level of depression. One consistent finding, however, is that immigrants report more stress compared to their U.S. born counterparts (Cuellar et al., 2004; Gonzales et al., 2001). It may be that older Arab-American immigrants experience higher incidents of depressed affect and lower life satisfaction compared to U.S. born Arab Americans due to the challenges associated with living and adapting to a new cultural context.

Religious Affiliation

The diversity of religious affiliation among Arab Americans is often obscured by the belief that all immigrants from Arab speaking countries are Muslim. While the most recent immigrants from Arab speaking countries tend to be Muslim (Abu-Laban, 1991; Orfalea, 1988), the majority of Arab Americans are Christian (Samhan, 2001). A major empirical question involves whether Muslim and Christian Arab Americans differ with regard to well-being. The small number of studies that compare Muslims and Christians focus on acculturation, and find little difference in those experiences (Faragallah, Schumm, & Webb, 1997; Read, 2003). Muslims may differ with regard to well-being, however, not only because their religion differs from that of the dominant culture, but given the anti-Muslim sentiment that exists in the United States post 9/11, it may be that Muslims report lower well-being than their Christian counterparts (Arab American Institute Foundation, 2002; Salari, 2002).

Resources

Human and social capital may account for links between immigrant status, religious affiliation and well-being. Human capital refers to characteristics that an individual holds (e.g., education) and social capital refers to the relationships one forges with others. Each represents qualitatively distinct resources. Their associations with well-being highlight the potential pathways through which immigrant status and religious affiliation influence well-being among Arab-American elders. Presented first is a discussion of human capital, followed by attention to the role of social capital.

According to Coleman (1990:304) “...human capital is created by changing persons so as to give them skills and capabilities that make them able to act in new ways.” Education level, income, and language constitute critical aspects of human capital for immigrant/ethnic groups in the United States. Such resources may be acquired over the life course due to various opportunities and experiences. High rankings on socioeconomic indicators such as education level and income are often associated with better well-being, particularly among older populations (Miech & Shanahan, 2000). Language fluency is salient for immigrant/ethnic groups, providing integration opportunities and promoting cultural understanding (Barresi, 1987; Lubben & Bacerra, 1987). Recently, however, research on immigrant well-being suggests that bilingual abilities provide a more productive pathway for well-being, allowing individuals to accommodate demands of both host and home culture expectations. Not having to forfeit their native cultural ways in order to acquire the skills of their new society helps members of immigrant/ethnic groups to achieve bi-cultural adaptation, which correlates with positive well-being (Barry, 2001; Cabassa, 2003; Cuellar et al., 2004; Portes & Hao, 2002). This may be particularly salient for Arab Americans who are facing increasingly hostility from the dominant culture due to recent socio-political events.

Human capital among Arab Americans may differ most by immigrant status. While the U.S. Census Bureau documents a highly educated profile for many Arab Americans with greater than average incomes, there also exist pockets of low educated, low income families, usually those who are recent arrivals to the United States (Brittingham & de la Cruz, 2005; Read, 2004; Samhan, 2001). Other research on Arab Americans suggests that human capital characteristics such as education and income do not differ a great deal by religious affiliation (Read, 2004).

Whereas human capital refers to characteristics acquired through the development of skills and abilities that become part of the individual, social capital, “is created when the relations among persons change in ways that facilitate action” (Coleman, 1990:304). In other words, the resource emanates from the relationship between individuals. The nature of family relationships is of heightened interest when examining immigrant/ethnic groups, particularly because leaving the home country frequently produces transformations in family relations. Spousal and parent–child interactions frequently change upon immigration, often becoming more negative (Yee, 1994). Additionally, among elders negative social interactions may have a stronger effect on well-being than do positive relations (Rook, 1984). Thus, although social support and social integration are associated with better health in a general sense, the association is likely to be multifaceted and to vary with specific construct indicators. In the current study, by focusing on two important sources of support within the family, spouse and adult child, the goal is to avoid generalizations that aggregate across important differences in type and source of support from family members.

There have been two studies conducted on Arab-Americans elders that touch on social relations, yet each focuses on one religious group. A qualitative study describing how socio-cultural traditions affect aging and retirement among middle-class Christian Arab Americans living in the northeastern part of the United States found that elders expected that children will care for them as they age yet simultaneously did not want to become a burden (Shenk, 1991). Moreover, Shenk illustrated within group differences demonstrating that a proportion of elders’ perceived needs were indeed met by family while another segment indicated that family did not meet their expectations. In a needs assessment of elderly Muslim Arab Americans residing in Dearborn, Michigan, Sengstock (1996) reports that while elders indicate high levels of satisfaction with the relationship they have with their children, satisfaction with their spousal relationship was less prevalent. Comparing Christian and Muslim elders with regard to social resources and well-being will provide a more nuanced understanding of Arab Americans.

Potential Mechanisms and Study Goals

Identifying potential pathways of improving health and well-being among Arab Americans will contribute to theoretical understandings of immigrant aging generally as well as to the practical issues of developing service delivery and care options. The current study focuses on a mediation hypothesis, asserting that the link between immigrant status and religious affiliation with well-being is explained, in part, by both human and social capital. According to the mediating hypothesis to be tested, possessing less human and social capital represents one of the causal pathways between immigrant status and lower levels of well-being. This hypothesis is suggested by prior theoretical and empirical work that links the availability of capital (both human and social) to immigrant status and well-being among younger age groups (Rumbaut, 2005; Zhou & Bankston, 1998).

These resources are examined separately to ascertain the unique contributions each form of capital exerts on well-being. An examination of the significance of each to the well-being of Arab-American elders will contribute to a more thorough understanding of how different worlds of aging, in the context of immigrant status and religious affiliation, exist among an understudied but increasingly targeted immigrant/ethnic group in the United States. Research suggests that gender and age may exert effects on well-being, although some have demonstrated these effects to be minimal in old age (Akiyama & Antonucci, 2002; Fernandez-Ballesteros et al., 2001; Mirowsky & Ross, 1992). Consequently, the present analyses will control for the potential effects of gender and age on well-being.

Specifically, the analysis explores

  1. 1.

    Whether immigrant status and religious affiliation are significantly associated with psychological well-being among Arab-American elders.

  2. 2.

    Whether human and social capital mediate the documented associations.

Design and Methods

Study design

Participants completed face-to-face survey interviews. Based on language preference, two-thirds of the interviews were conducted in English (by a native English-speaking Arab American), and one-third in Arabic by a native Arabic-speaking bi-lingual interviewer. Interviews occurred at the participants’ homes and lasted for approximately an hour. The data were collected in 2001–2002, with the first interview conducted after the events of 9/11, beginning in November 2001.

Sample

Participants included 101 Arab-American elders aged 56 and above from the metropolitan Detroit area. Michigan is home to the largest and most visible Arab-American community of any state; in fact, the Arab-American population grew in Michigan by 65% from 1990 to 2000 (http://www.aaiusa.org/AAIF/Census/demographics/MIdemographics.pdf, retreived May 26, 2006). Census data suggest that there are 151,493 Arab Americans in Michigan, however, community-based estimates place the number closer to 490,000 (Zogby Study, 1997). Based on place of birth of the participant’s father, the largest national origin group in this study was Lebanese (57%), followed by Syrian (19%), and the third largest came from Iraq (10%). The national origins of the sample mirror U.S. Census data that report the largest proportion of Arab Americans are Lebanese and the second largest Syrian. Analysis of the ancestry question on the long form of the U.S. Census form also reveals that the largest numbers of immigrants to Michigan in recent years have come from Iraq (http://www.aaiusa.org/AAIF/Census/demographics/MIdemographics.pdf, retreived May 26, 2006). Approximately 45% of the sample was men, and about two thirds were immigrants. Reflecting the educational distribution of Arab Americans in the U.S. Census (Zogby, 1990), the sample is about evenly divided between those with more than a high school education and those who report less than a high school education.

Study participants were accessed primarily though religious institutions. One mosque and two churches served as initial recruitment sites. The religious leader of each establishment was informed about the study and then referred the principle investigator to key community leaders who aided in identifying potential participants. Snowball sampling followed. Recruitment from two area social service agencies occurred later in the study to ensure representation of those immigrants who spoke Arabic better than English.

Measures

Dependent measure

Well-being is considered a multi-dimensional construct, and assessed in two ways meant to identify the affective and cognitive dimensions of well-being. The affective dimension was measured using an item from the WHOQOL-BREF multidimensional measurement instrument for quality of life. This question was included in the survey both because it is meant to be used in cross-cultural research (WHOQOL Group, 1998), and to allow for potential comparative analysis with data carried out in Arab speaking societies. Respondents were asked, “In the last two weeks, how often have you had negative feelings such as feeling down, despair, anxiety, depression?” The response categories were on a five-point scale ranging from never, sometimes, often, quite often, and always.

In order to measure the cognitive aspect of well-being, an indicator routinely used in numerous social research studies to measure life satisfaction was included (George, 2006). Respondents were asked ‘How satisfied are you with your life as a whole these days?’ on a seven point scale. The variable was coded so that 7 indicated the highest life satisfaction score and 1 the lowest.

Immigrant status

Immigrant status was measured by two assessments: place of birth and for those not born in the United States length of residence, following the example of Finch, Hummer, Reindl, and Vega (2002). Elders first responded to the question: “Where were you born (city and state/country)?” Immigrant status was then categorized into three strata according to the mean number of years immigrants lived in the United States (x = 30.12 sd = 16.83; only 7 out of the 62 immigrants in this study reported living in the United States less than 10 years). This assessment allowed for a consideration of whether duration in the United States is significant for immigrant well-being. Indicators included those who immigrated during middle age or later, less than 30 years ago (n = 28); those who immigrated before middle adulthood, 30 years ago or more (n = 34); and those born in the United States (n = 38). The variable was coded such that 1 represented those born in the United States, and 0 represented those born outside the country. Similarly, those who reported living in the United States for 30 years or longer were coded as 1, and all others 0.

Religion

The participant’s religion was measured by asking the elder to self report in response to the question, “What is your religion, if any?” Responses were dummy-coded for purposes of this analysis where 0 represents Christian and 1 represents Muslim.

Human capital

Human capital was indicated by three separate measures. Education, a continuous variable representing the number of years of schooling completed, ranged from 0 up to 17+. Income was measured by asking the respondent to estimate total pre-tax household income from all sources (jobs, interest, rents, and so forth) for him/her and all family members with whom s/he lived for the previous year. An income range was presented, divided into four categories, and thus becomes an interval level variable for this analysis, where 1 = <$20,000 a year, 2 = $20–$39,999 a year, 3 = $40–$59,999 a year and 4 = $60,000 or above a year. To assess language elders were asked: “What language(s) do you speak in your home?” Response categories included: Arabic, English, both Arabic and English, and Other. The language variable was coded ‘bilingual’ where 1 represents those who reported speaking both Arabic and English in their home, and 0 represents those who reported speaking either English only or Arabic only. Similarly, ‘English’ was coded as 1 representing those who reported speaking English only in their home and 0 representing those who reported speaking Arabic only or both English and Arabic. No respondent indicated speaking an ‘other’ language in their home.

Social capital

Social capital referred to type and quality of support in family relations, specifically with spouse and child. Participants were asked first about their relationship with the child upon whom they rely most, and second about their spouse. Elders responded on a scale that ranged from 0 (not at all) to 10 (absolutely) first if they perceived emotional support available: “I can share my very private feelings and concerns with (spouse/child).” Next, instrumental support was measured by having elders respond to the statements: “I feel that (spouse/child) would help me out financially if I needed it” and “I feel that (spouse/child) would take care of me when I am sick.” Because descriptive results showed high positive skewing, these variables were recoded so that 1 indicates absolutely, and 0 indicates the rest, that is those who designated 9 and below.

Quality of support was assessed by asking two questions: “(Spouse/Child) gets on my nerves” and “(Spouse/Child) makes too many demands on me.” These variables were recoded so that 0 indicates not at all, and 1 indicates the rest, that is those who designated 1 and above. Table I provides a frequency distribution of responses by relationship.

Table I Frequency Distribution of Social Capital
Full size table

Control variables

Gender was a dummy variable with 0 representing male and 1 representing female. Age was measured continuously (in years) based on a request to indicate date of birth.

Methods of analysis

A bi-variate correlation of study variables with the dependent variable was conducted first, followed by a regression model constructed for each dependent variable of interest. The first set of regression analyses involved testing for the effects of immigrant status on each of the two dimensions of well-being (perceived negative feelings and life satisfaction) controlling for gender and age. The next set of analyses included human capital, divided into two overarching areas: socioeconomic and language, to test whether they mediate the association between the independent variable (immigrant status) and well-being (negative feelings and life satisfaction). Another set of regressions were conducted separately regarding social capital available from child and spouse to examine whether type of support and then quality of support mediate the association between the independent and dependent variable(s). Results of the regression analysis are presented only when models are significant (p < .05). Finally, a formal assessment of the indirect effects of human and social capital when they were significant in the regression models was conducted by testing whether the independent variable (i.e., immigrant status) was significantly associated with the proposed mediator (i.e., education) and then applying a nonparametric bootstrapping procedure to test effect-size estimation of the mediating variable (Preacher & Hayes, 2004). This approach is especially well-suited for small samples.

Results

Correlation results (see Table II) demonstrated that religious affiliation, whether one is Muslim or Christian, was not significantly associated with the dependent variables. Religion was therefore not included in the remaining analyses. Being U.S. born, however, was associated with both lower frequency of negative feelings and higher life satisfaction. Length of time in the United States among immigrants had no significant associations with the dependent variables. Following from these results, immigrants, regardless of duration in the United States, were combined for all further analysis (0 = U.S. born; 1 = Immigrant).

Table II Nonparametric Correlation of Dependent Variable with Study Variables N = 101
Full size table

More frequently perceived negative feelings correlated with low life satisfaction, as well as with one of the control variables: younger age. Higher levels of life satisfaction correlated with another of the control variables: gender (being male).

All human capital indicators correlated with perceived negative feelings with the exception of bilingualism. Speaking English only in the home was associated with less frequent negative feelings, as were high education and high annual family income. Life satisfaction, on the other hand, correlated only with high education and high income.

With regard to relations with key family members, perceived negative feelings was associated with the respondent perceiving that s/he may confide in both their adult child and spouse, as well as with the perception that their spouse makes no demands on the respondent. Higher life satisfaction correlated with the perception that their child does not at all get on her/his nerves and with the perception that sick care is available from a spouse.

Tables III and IV present the two models predicting the effects of immigrant status, human capital, and social capital on each measure of well-being.

Table III Regression Analysis Summary—Negative Feelings
Full size table
Table IV Regression Analysis Summary—Life Satisfaction
Full size table

Predicting negative feelings

Controlling for gender and age, being an immigrant is associated with more frequent reports of feeling down (p < .01).

Models 2 and 3 present results from the mediating analysis, suggesting that of the human capital variables tested, income and English spoken in the home were significantly associated with perceived negative feelings at the trend level (p < .10). Those who reported lower income levels and do not speak only English at home were more likely to report feeling down. Note also that the association between immigrant status and perceived negative feelings was no longer statistically significant, suggesting that income and language are important resources and may account for associations between immigrant status and elder well-being. The boot-strapping procedure to test effect-size estimation of the mediating variable indicated that while lower income was significantly associated with immigrant status (p < .001), the indirect effect of income level was not significantly different from 0. The true indirect effect was estimated to lie between −0.1121 and 0.5433 with 95% confidence, illustrating no mediation effect. Speaking only English at home was also significantly associated with being U.S. born and did reveal a true indirect effect significantly different from 0. The indirect effect was estimated to lie between 0.0301 and 0.7719 with 95% confidence, demonstrating that language, specifically speaking English in the home, mediated the association between immigrant status and negative feelings.

The only effect of social capital significantly associated with negative feelings was the support perceived to be available from the child upon whom the elder relies most. A perception that the elder may confide in his/her child was positively associated with perceived negative feelings. The association between immigrant status and negative feelings became a trend. Tests of mediation indicated that being an immigrant was also associated with the perception that the elder may confide in their child at a trend level (p < .10), yet the indirect effect of perceived emotional support from child was not significantly different from 0. The true indirect effect was estimated to lie between −0.0074 and 0.3508 with 95% confidence, illustrating no mediation effect. No significant associations were present when social capital from spouse or negative aspect of relations with child and spouse were included in the models.

Predicting life satisfaction

Paralleling results with perceived negative feelings, immigrant status was significantly related to life satisfaction so that being U.S. born was associated with higher ratings of life satisfaction (p < .05), controlling for gender and age.

Higher education was slightly (p < .10) associated with more life satisfaction, and immigrant status was no longer statistically significant when human capital indicators were included in the model. The macro results indicate that immigrant status was associated with lower education levels (p < .001), and additional analyses revealed that the indirect effect of education level was significantly different from 0. The true indirect effect was estimated to lie between −0.7168 and −0.0128 with 95% confidence, revealing that education level mediated the association between immigrant status and life satisfaction.

Only negative aspects of relations with child and spouse were significant (or demonstrated a trend toward significance) in the regression models. Specifically, perceiving that either a child (p < .05) or spouse (p < .10) gets on the elder’s nerves was associated with lower life satisfaction, while perceiving that a spouse makes too many demands (p < .10) was associated with higher life satisfaction. Analyses revealed that being U.S. born was not associated with whether or not a child is perceived to get on nerves, and accordingly additional analyses revealed that the indirect effect of each measure was not significantly different from 0. The true indirect effect for child gets on nerves was estimated to lie between −0.2347 and 0.2438 with 95% confidence. Being U.S. born was associated with spouse getting on the elder’s nerves and spouse making too many demands, but the true indirect effect for spouse gets on nerves lies between −0.0987 and 0.5668; and for spouse makes too many demands between −0.4698 and 0.1743 with 95% confidence, revealing no true mediation effects.

Discussion

An examination of well-being among Arab-American elders is particularly timely due to the potential that events following 9/11 are correlated with Arab Americans reporting higher incidents of psychological distress (Howell & Shryock, 2003) and increased mental health care utilization (Amer, 2005). Affective and cognitive indicators of well-being were examined in an attempt to provide preliminary evidence about how immigrant status, religious affiliation, and available resources may play a role in the lives of Arab-American elders. Results provide vital information about the diversity of Arab Americans by highlighting how such characteristics vary within this group, but also hint at whether they operate in similar ways as reported in the extant literature. In the following paragraphs the implications of these results are considered.

The finding that religious affiliation does not correlate with well-being is not especially surprising, given recent findings suggesting no significant effects of religious affiliation among Arab Americans on an array of outcomes ranging from adaptation to gender attitudes (Faragallah et al., 1997; Read, 2003). While other studies focus on younger Arab Americans, this analysis corroborates the lack of significant effects of religious affiliation on an older sample. Islam is often believed to promote traditionalism more so than other Abrahamic faiths, yet this analysis suggests that the immigrant experience supersedes or is more prominent than religious affiliation. What might be more influential is the degree of religiosity (very religious or not) and/or religious behaviors (church/mosque attendance or daily prayer) as they relate to well-being as opposed to simply whether one is Christian or Muslim.

Immigrant duration in the United States has little effect on well-being, but whether or not one is born in the United States does seem to matter. Elders who immigrated may be vulnerable in more ways than their U.S. born counterparts in that they are more likely to report negative affect and report lower levels of life satisfaction. These results may represent the challenges immigrants face when they leave their homeland, either voluntarily or under duress, and enter a new culture. For instance, one may be pushed out of their country by persecution, political instability, danger, fear for their children’s future, or some other hardship. Those immigrants may actually be suffering from post traumatic stress disorder, depending on the circumstances. Conversely, one may be pulled to another land in order to experience more opportunities, or join children and other family members. In either case, immigrants may suffer mental distress or even endure disappointment originating from the forces that initiated the migration. Furthermore, the significant association between immigrant status and well-being among older Arab Americans mirrors findings from other studies on immigrant elders. Research on Hispanic/Latino and Asian elders demonstrate that immigrants are indeed at heightened risk of depressed mood or affect (Aranda & Miranda, 1997; Mui, 1996; Plawecki, 2000; Wilmoth & Chen, 2003).

Human capital does offer one potential pathway by which Arab-American immigrants may experience both affective (perceived negative feelings) and cognitive (life satisfaction) dimensions of well-being. The findings demonstrate that while human capital is significant for both dimensions, each are sensitive to different factors, further elucidating the importance of examining multiple indicators of human capital as well as both affective and cognitive aspects of well-being. For instance, findings illustrate that the link between perceived negative feelings and being an immigrant disappears when English spoken in the home is considered. This finding illustrates that language acquisition is critical for Arab-American immigrant elder well-being at the affective level. The significance of language may lie in its potential to impede meaningful interaction not only with larger society, but also with family members. Language discrepancies may alter the quality of family relations for elders, as younger generations acquire new ways of thinking through English language fluency (Yee, 1994). Education provides a significant resource for immigrants with regard to life satisfaction, even when including income in the analysis. This finding suggests that socio-economic indicators of human capital comprise separate and independent resources, providing a more detailed understanding of the nature and extent of those resources available and their influence on well-being. Indeed, education level may be a critical indicator of access to resources among older cohorts (Broese van Groenou & van Tilburg, 2003).

Findings suggest that the indicators of social capital utilized in this study, perceived social relations with adult child and spouse, do not mediate the relationship between immigrant status and well-being for Arab-American elders. Although the models predicting well-being are significant when some aspects of social capital are included, the estimation of indirect effects reveals no statistically significant link. These findings suggest that the immigrant-well-being association in this sample does not hinge on the type or quality of social relations with spouse or child, and instead may be mediated by any or all of a host of other factors, from greater exposure to stressors to a lack of a sense of control (Jackson, Williams, & Torres, 2003; Lefley, 1999) to human capital factors as suggested above.

Social capital does not mediate the association between immigrant status and well-being, yet results demonstrate significant associations between social capital and well-being. Perceiving the ability to confide in a child is associated with more frequent reports of depressed affect. While this finding is counterintuitive, it may signify a finding in the literature that talking more about a problem increases depression levels (Nolen-Hoeksema, Morrow, & Frederickson, 1993). Or conversely, it may signify that children are a significant source of comfort for Arab-American elders, as opposed to age peers or spouse, when elders are feeling down. In other words, they feel free to discuss their negative feelings with children should the need arise. Results also demonstrate that perceiving a child gets on nerves is associated with lower levels of life satisfaction among Arab-American elders. This finding suggests that while relying on a child does occur, it may not be optimal for them to be the only source of support in times of need. It is not only the availability of support that matters, but the nature of the relationship, which is key to well-being (Antonucci, Akiyama, & Lansford, 1998; Rook, 1984). A child undoubtedly represents an important support resource, however, when tensions are high, additional sources of support may prove vital to elder life satisfaction. This finding suggests an instance where external sources of aid to elders and their families may be most beneficial, particularly interventions that augment those circumstances when family acts as a sole resource in times of need. While children certainly provide critical sources of support for elders, their contributions may be enhanced by supplements from other means.

Finally, limitations and suggested avenues for future research should be addressed. Because the analysis is conducted on a convenience sample, it may not be representative of Arab Americans, prohibiting any generalizations. Additionally, the small sample size prevents the use of more sophisticated analysis. Finally, all data are self-reported; no objective measures of well-being or social capital are assessed. Despite such limitations, these findings do provide critical preliminary evidence about immigrant status, religious affiliation, well-being, and available resources among an understudied ethnic group, thereby guiding the direction of future research on the experiences of older Arab Americans.

This study provides empirical evidence about the pragmatic situations facing Arab-American elders. Arab Americans are a growing population in the United States and portrayed as one of the most successful ethnic groups, yet pockets of lower-class, less educated individuals also exist (Read, 2004; Zogby, 1990). The influence of human capital on well-being constitutes an essential area of inquiry. Also, as is often the case among stigmatized groups, reliance on ‘your own kind’ supersedes any attempt to seek external or formal sources of aid in times of need (Lareau, 2002). Many published works on Arab-American families highlight traditional ideals concerning social capital, lacking systematic documentation of the nature and quality of social relations, an important resource for health and well-being. Empirical research that examines human and social capital as fundamental pathways through which identified vulnerabilities affect well-being among older Arab Americans is warranted. The findings from this study indicate that resources and their distribution are associated with noteworthy variations in well-being, highlighting one aspect of diversity among Arab-American elders.