Religious Diversity and Community Volunteerism Among Asian Americans

Transcription

1 Religious Diversity and Community Volunteerism Among Asian Americans ELAINE HOWARD ECKLUND JERRY Z. PARK We examine whether religious membership and participation foster community volunteerism among a religiously diverse group of Asian Americans. We use data from the Social Capital Community Benchmark Survey (SCCBS), the only data set that contains both a large, national sample of Asian Americans and detailed questions on religious and civic participation. Asian-American Protestants, Catholics, and adherents of non-christian religions are involved in community volunteerism to varying degrees. Surprisingly, however, fewer Hindus and Buddhists volunteer when compared to the nonaffiliated. We use these results to propose theoretical concepts that take into account the impact of a religion s structure as well as the double-minority status faced by nonwhite and non-christian Asian Americans on the likelihood of volunteering. Our findings indicate that accepted predictors of community volunteerism may operate differently among new nonwhite immigrants and their children than in the general U.S. population; this provides building blocks for future research on religion and civic participation among nonwhite and non-christian populations. INTRODUCTION With the current religious and racial transformation of the United States, the study of American civic life warrants the inclusion of new populations. Much of the change in U.S. demographics is as a result of the new immigration (ca. mid-1960s to mid-1990s), which has increased the number and diversity of nonwhite racial minorities as well as the number of non-christian religions in the country. Where American religion was once almost exclusively Western Judeo Christianity, it now includes non-christian religions such as Buddhism, Hinduism, and Islam (Wuthnow and Hackett 2003). American Muslims nearly outnumber Jews, congregating in an estimated 3,000 mosques (Ostling 2001). And some researchers report the U.S. presence of at least 4 million American Buddhists (Numrich 2000). 1 Both religion and race are often closely intertwined with community volunteerism, the dominant form of U.S. civic involvement (Musick, Wilson, and Bynum 2000). Little research, however, specifically examines the various ways in which the increasing presence of non-christian religions and nonwhite racial groups will shape the future of American civic life. With data resulting from the Social Capital Community Benchmark Survey (SCCBS) we are able, for the first time, to explore the relationship between religious involvement and community volunteerism among Asian Americans, an important yet understudied population. Here we ask what factors influence Asian-American community involvement and focus specifically on religiosity, a dominant influence on civic participation in the general population (Putnam 2000). We find that both Catholic and Protestant Asian Americans volunteer more than the nonreligious, but surprisingly, Asian- American Buddhists and Hindus actually volunteer less than those with no religious affiliation. In addition, contrary to studies of the white American population, gender, education, and income Correspondence should be addressed to Elaine Howard Ecklund, Department of Sociology at the University at Buffalo, SUNY, 430 Park Hall, Buffalo, NY, Elaine Howard Ecklund is Assistant Professor of Sociology at the University at Buffalo, SUNY, 430 Park Hall, Buffalo, NY. ehe@buffalo.edu Jerry Z. Park is Assistant Professor of Sociology at Baylor University, One Bear Pl , Department of Sociology, Baylor University, Waco, TX jerry park@baylor.edu Journal for the Scientific Study of Religion (2007) 46(2):

2 234 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION are not significant influences on volunteering for this population. In the conclusion we propose a theoretical model that takes into account current research on religion and civic participation as well as that on human capital, arguing that some Asian Americans face a double-minority status as both non-protestants and nonwhites. This status may influence their ability to link with American volunteer organizations. RELIGION, RACE, AND CIVIC PARTICIPATION Religion and community volunteerism have a connected, although complex, relationship (Becker and Dhingra 2001). 2 Because of low sample sizes, studies of religion s influence on volunteerism for members of American non-christian religions, such as Buddhism, Hinduism, and Islam, as well as nonwhite Americans within Christianity are largely absent from most analyses of religion and civic participation. 3 Since Asian Americans are a fast growing Census and politically defined U.S. racial group, they are a centrally important population to study in order to gain a better understanding of American civic life (Barnes and Bennett 2002; Espiritu 1992; Lien, Conway, and Wong 2003). In addition, they form the majority of the non-christian world religions in America, and a significant number belong to the two dominant forms of American Christianity, Catholicism and Protestantism (Min and Kim 2002). Only a few ethnographic studies have investigated the relationship between religion and community volunteering for Asian Americans. Carolyn Chen s study of an immigrant Taiwanese Protestant church and an immigrant Taiwanese Buddhist temple (Chen 2002) and Elaine Howard Ecklund s study of second-generation Korean Americans in monoethnic and multiethnic churches (Ecklund 2005a, 2006) both find that the distinct religious discourses of these different types of religious organizations change the meaning and form of community volunteering. While there may be variation in the types and forms of voluntary activity, this initial evidence suggests a positive relationship between religion and volunteering for Asian Americans irrespective of religion. Research on volunteering reveals that individuals, particularly those who participate actively in religious communities, generally have relationships with others who are demographically similar (Wuthnow 2003; Wuthnow, Hackett, and Hsu 2004). Work on community participation has shown that ethnic minority religious groups may not be asked to volunteer as frequently as largely white groups (Musick, Wilson, and Bynum 2000; Wilson and Musick 1997). Consequently, Asian Americans who are members of minority religions may face a double-minority status as both religious and racial outsiders. On one hand, because other groups of Americans do not think of them as typical volunteers, Asian Americans may not be asked to volunteer as often as white Christians. On the other hand, this status could act to foster civic integration as non-christian Asian Americans seek to prove their patriotism by being especially involved in American civic life. THE ASIAN-AMERICAN CASE Asian Americans are an ideal case to study the impact of religious membership and participation on community volunteerism for members of diverse religious groups. They are a fast growing Census and politically defined U.S. racial group (Espiritu 1992; Okamoto 2003). Between 1990 and 2000, the Asian-American population experienced between 48 and 72 percent growth and reached a total estimated population of between 10.2 and 11.9 million, making Asian Americans the fastest growing group of new Americans (U.S. Census 2000). And Asian Americans form a diverse population, including both the native born and immigrants. For Asian Americans, loosely defined as a racial group, the double-minority status is especially significant. For example, the current interest in new religions in America has typically focused on three major non-christian world religions (Buddhism, Hinduism, and Islam), all of which are dominated by members of nations on the Asian continent.

3 RELIGIOUS DIVERSITY AND COMMUNITY VOLUNTEERISM 235 There have been few comparative studies, however, that examine the impact of different religious traditions on community volunteerism for this group (Chen 2002). What remains unanswered is whether religion merely provides resources internal to specific ethnic Asian-American (and mostly immigrant) communities or whether religious participation facilitates active engagement with other sectors of American society (Ecklund 2005b). Addressing this question may provide some insight into the extent and significance of religion s mediating influence on the broader civic integration of nonwhite communities. By studying Asian Americans, we focus on a nonwhite group whose influence on American civic life is rarely considered. Different from previous research, our concern here is with examining Asian Americans as a whole, with an emphasis on Asian Americans as a case of religious diversity. We examine Protestantism and Catholicism to study the influence of religion on civic participation among the two largest divisions within American Christianity. We consider also the experiences of two of the major minority U.S. religions, Buddhism and Hinduism. 4 By studying Asian-American members of non-christian religions, our analysis offers one of the first glimpses into the world of groups that are both racial and religious minorities. Previous research has shown that religious membership and activity for African Americans promotes civic participation (Musick, Wilson, and Bynum 2000) and we might expect similar results among Asian Americans. An important distinction, however, between Asian Americans as a whole and African Americans is the diversity of religious traditions within Asian America. While religion for most African Americans is typically Protestant, religion for Asian Americans encompasses a variety of traditions including Protestantism, Catholicism, Buddhism, Hinduism, and Islam. To what extent is the impact of religious activity on community volunteerism different for members of these religions when compared to Protestants and Catholics? For example, Carolyn Chen s comparative case study suggested that Buddhist and Protestant Taiwanese Americans differ in the rationale and type of civic participation, but that both groups value volunteering (Chen 2002). Hypotheses Based on ethnographic data as well as studies of religion s influence on volunteering among the general population (Wilson and Janoski 1995), we hypothesize, first, that religiously affiliated Asian Americans will have higher rates of volunteering than nonaffiliated Asian Americans. Second, because of the positive incorporation of many, although certainly not all, Asian Americans into the economic and educational institutions of American society (Kitano and Daniels 2000), we anticipate that other basic social characteristics such as gender, education, and income will have the same positive effect on community volunteerism as seen in studies using white American samples. Third, given the Christian context of religious life and voluntary experience in the United States (Wuthnow, Hackett, and Hsu 2004), however, we hypothesize that Protestant and Catholic Asian Americans overall will be more likely to volunteer than Asian-American members of non-christian religions. DATA,MEASURES, AND METHODS Analyses are based on the SCCBS, a publicly accessible data set consisting of a two-tier sample of adult Americans conducted late in the year Weighting to adjust for demographic disparities and responses to income and education level questions produced an overall sample of 711 Asian-American respondents for our analyses. 6 The survey was a random-digit-dialing phone survey and achieved a 29 percent response rate for both community samples and the national sample. 7 While this response rate seems low, it is nearly identical to the results of the 2004 survey experiment conducted by the Pew Research Center (see where they found very little difference in response to a range of variables between a survey that resulted in a 27 percent response rate and one that achieved a 51 percent response rate.

4 236 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION We chose to focus specifically on Asian Americans as a group rather than examining specific groups within Asian America for two central reasons. First, low sample sizes for several Asian ethnic groups (e.g., Indian, Korean, Vietnamese, N < 36 each) made statistical tests unresponsive to possible group differences. Second, we had little theoretical reason to expect ethnic group differences in volunteer rates among Asian Americans. 8 The most central weakness of the data set is the inability to compare immigrant and nativeborn Asian Americans because no question about immigrant status was asked in the SCCBS. Because over 70 percent of Asian Americans living in the United States are foreign born, 9 in the conclusion of the article we extrapolate on how the findings presented here may have implications for the influence of immigrants and their children on American religion and civic life. Despite these drawbacks the SCCBS is the only data set that allows systematic comparison between those with Christian and with non-christian religious affiliations and is also the only data set that allows the possibility of examining religion s influence on civic participation for a large, national sample of Asian Americans. These benefits far outweigh its limitations. Dependent Variable Community Volunteerism Community volunteerism is a central form of civic participation in American society and includes involvement in local communities and neighborhoods (Wilson and Janoski 1995). Our dependent measure of civic participation from the SCCBS was participation in formal volunteering organizations. Survey respondents answered a series of questions about participation in neighborhood projects, arts, health, youth, and humanities organizations. 10 Given the variety of ways in which volunteering was measured, none of which is more or less correlated with religious practice, we took the sum of these responses and created a dichotomized variable where 1 = volunteered for any of the following organizations and 0 = not volunteering for any of these. 11 Independent Variables Religious Participation and Affiliation Religious measures included six separate indicators of both affiliation and practice: (1) religious worship attendance; (2) religious participation aside from worship; (3) religious financial giving; (4) religious group membership such as a church, temple, or mosque; (5) religious group participation apart from a worshipping congregation; and (6) religious volunteering. We reduced religious attendance from a five-point continuum ranging from never to weekly or more to three dichotomous variables, where each measure captures a small range of attendance responses, but sustains substantive distinctions. These include: less often than a few times per year to a few times per year ; once or twice a month to almost every week ; and every week or more often. Religious participation aside from worship was a dichotomous indicator, where 1 = yes and 0 = no. Religious financial giving was represented by the reported amount the respondent contributed to his or her local church or other religious causes, ranging from none to more than $5,000. We collapsed these responses into a dichotomous measure reflecting no giving and giving. Religious membership, religious volunteerism, and religious group membership, aside from congregational attendance, were also coded 1 = yes and 0 = no. Because the sample size was not large enough to make comparisons between different Protestant traditions, we merged all Protestant Christian traditions together (evangelical, mainline, and African American), following a modified version of the scheme developed by Steensland et al. (2000). 12 We sustained the Catholic and nonaffiliated categories from the original schema, and

5 RELIGIOUS DIVERSITY AND COMMUNITY VOLUNTEERISM 237 divided the non-christian religious groups into two major world religions, Buddhism and Hinduism, with a residual category of all other religions. 13 Class and Gender We included class and gender as control variables within each analysis, two variables that are significant predictors of civic participation for other groups of Americans. For example, higher income and education, taken together as a measure of class, are both strongly correlated with increased civic participation (Musick, Wilson, and Bynum 2000; Putnam 2000; Smith 1994). Further, women are consistently more likely than men to volunteer (Wilson and Musick 1997). In this study, gender was defined dichotomously (1 = female ). We determined class according to the respondent s highest level of attained education and present family household income, measuring education using four dichotomous variables each representing the highest reported level of education by the respondent: high school or less ; some college ; college university ; and graduate/professional degree. Income was measured using six self-reported measures of income to indicate household income: less than $20K ; 20 30K ; 30 50K ; 50 75K ; K ; and 100K or more. To analyze the data, we first conducted a series of significance tests comparing the various levels of civic participation and distribution of religious resources for respondents of each religious tradition. Next we applied logistic regression techniques to determine the likelihood that each independent measure influenced civic participation in the overall sample. In the analyses we first present the independent effect of each measure, followed by one model showing only the effects of the control variables. Each of the remaining six models includes these control measures while adding in each measure of religiosity. FINDINGS Religious Affiliations Table 1 presents the descriptive statistics for overall participation and religious practice according to religious affiliation. Catholic Asian Americans were the largest group of religious affiliates, comprising about 24 percent of the sample, followed by the other religious groups in descending order: Protestants (20 percent); Buddhists (11 percent); and Hindus (about 9 percent). The other religious Asian Americans together comprise another 7 percent, and the nonaffiliated were the largest share of the sample at nearly 29 percent. These figures are in relative agreement with Jasso et al. s (2003) pilot study of recent immigrant Americans, which included a subset of Asian Americans (figures available upon request) (Jasso et al. 2003; Cadge and Ecklund 2006). When analyzing predictors of the dependent variable, community volunteerism, Asian Americans of various religious affiliations report different rates of participation. While 51 percent of the sample reported having participated in at least one nonreligious volunteer organization, Catholic, Protestant, and other religion Asian Americans reported higher participation levels than the average (54, 69, and 65 percent, respectively) while Hindu, Buddhist, and nonreligiously affiliated Asian Americans fell below the average (40, 26, and 45 percent, respectively). Overall, our analyses reveal that the two Christian religions have greater civic participation levels compared to the non-christian religious traditions. Most notably, the members of the two non-christian religions, Hinduism and Buddhism, volunteered less than the nonaffiliated. 14 On most measures of religious practice, Protestant Asian Americans reported the highest levels among all religious groups in this sample. This may reflect a particular bias in the survey toward asking questions about the types of practices, such as congregational attendance, that are more prevalent among Protestants. This issue aside, it bears noting that the majority of Asian- American Protestants and Catholics reported religious service attendance at least once per month

6 238 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION TABLE 1 ASIAN-AMERICAN RELIGIOUS TRADITIONS, COMMUNITY VOLUNTEERISM, AND RELIGIOUS RESOURCES (PERCENT YES ), SOCIAL CAPITAL BENCHMARK SURVEY 2000 Other Non- Chi-Sq Protestant Catholic Hindu Buddhist Religion affiliated Total Percent of sample Volunteered for at least 1 org Age (Mean) Female Education High school or less Some college/associates BA Grad/prof. degree Income Less than 20K K K K K K Religious attendance A few times a year or less Once per month to almost weekly Weekly or more often Religious participation (% yes) Religious giving (%yes) Religious organization membership (% yes) Mem. relig-affil. org. (% yes) Religious volunteering (% yes) p < 0.05, p < 0.01, p < or more (about 78 and 71 percent, respectively). In contrast, most Asian-American Hindus and Buddhists reported religious attendance of a few times a year or less. These differences continue in religious participation outside of worship services: about 61 percent of the Protestants reported participating, whereas all other groups reported no more than 35 percent participating. About 82 percent of Catholics reported some amount of financial giving to religious causes or organizations, the highest of all groups. Hindu respondents had the lowest proportion reporting any giving at 57 percent. About 71 percent of Protestants and Buddhists reported religiously-based financial giving. Membership in a religious organization was the highest among Protestant Asian Americans (almost 70 percent), followed by Catholics (57 percent), Hindus (34 percent), and Buddhists (19 percent). Few respondents of any religious group with the exception of Protestants (33 percent) reported membership in a religious organization outside of their congregation. About 17 percent of Catholics and less than 2 percent of Buddhists and Hindus reported the same. Volunteering

7 RELIGIOUS DIVERSITY AND COMMUNITY VOLUNTEERISM 239 for a religious organization was also fairly low. Almost a third of Protestant Asian Americans indicated religiously-based volunteering, whereas less than 24 percent of any other group reported such volunteering. Human Capital Factors A few differences among religious traditions appear when analyzing the independent variables of age, gender, education, and income, variables that are shown by other research to have the greatest influence on volunteering. Buddhist, Catholic, and Protestant Asian Americans had a higher percentage of women (60, 57, and 56 percent, respectively) compared to the percent of women in the overall sample (about 50 percent). Members of the other religions group and the nonaffiliated had a much lower percentage of women (42 and 38 percent, respectively). In terms of education, the distribution across levels of degree attainment was roughly similar. With the exception of Hindu respondents, most groups had the largest proportion report some college education. This was also true for the nonaffiliated. Nearly one-half of the Hindu respondents reported attaining a graduate or professional degree, however. Income distribution presented a similar picture with respect to the religious groups. The majority of most religious subgroups reported an income in the middle range (from $30K to $75K). But while Protestant and Hindu Asian Americans reported a household income in the middle range, between 22 and 27 percent report an income of 100K or more. In other words a sizeable portion of Asian-American Protestants and Hindus have the kinds of human capital resources that foster civic participation in the general population. Hindu Asian Americans notably have more human capital with their combined higher education and income levels when compared to the general population. In Table 2 we show the independent effects of a variety of measures that often influence community volunteerism. Surprisingly, age, gender, and education, all factors that influence the likelihood of volunteering for the general population, generally have no significant relationship to community volunteerism among Asian Americans in this sample. These independent effects are fairly stable across subsequent models when we include a variety of religious involvement measures. These findings are surprising given that comparative studies analyzing these measures among white Americans generally show a significant and positive relationship to civic participation. Our second hypothesis is therefore refuted, since age, gender, and education do not have the same relationship to community volunteerism for this population as they do for the general population. The only exception to this pattern was income and here we found only marginal support as the significance of increases in income changed when various religious controls were added. As expected, in bivariate modeling higher income levels were usually associated with a higher likelihood of volunteering (except for those in the middle category earning between $30 and $50K). While this effect was stable after introducing other controls and religious tradition, income was no longer significant when accounting for religious participation, religious giving, and membership. In short, the most consistent group along income levels was that reporting earnings in the highest bracket ($100K or more); among these Asian Americans only at very high levels of income attainment is income associated with an increased likelihood of volunteering. Religious Affiliation When examining religious affiliation, we found that Protestant Asian Americans were most likely to have participated in a voluntary organization relative to the nonaffiliated. Buddhist Asian Americans, however, were significantly less likely to volunteer relative to the nonaffiliated. In Table 2, Model 1, when we control for demographic and socioeconomic characteristics such as age, gender, education, and class, we find that Protestant and other religious affiliation increase the likelihood of nonreligious voluntary group participation among Asian Americans.

8 240 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION TABLE 2 ODDS RATIOS OF ASIAN-AMERICAN COMMUNITY VOLUNTEERISM: SOCIAL CAPITAL BENCHMARK SURVEY 2000 (FULL SAMPLE) Independent Model Model Model Model Model Model Model Effects Age Age a Gender: Female Class: Education b 2 years of college years of college Post-BA Income a 20 30K K K K K Religious tradition c Protestant Catholic Hindu Buddhist Other religion Religious resources Religious attendance d Once/mo. to almost weekly Once a week or more Religious organization participation Religious giving Religious organization membership Religious association Religious voluntarism Constant Nagelkerke R N p < 0.05, p < 0.01, p < a Less than 20K excluded. b High school or less excluded. c Other religion refers to other non-christian religions not already specified. Comparison group is the nonaffiliated in Model 1 only. Subsequent models use Other religion as the comparison group. d Attend a few times a year or less excluded.

9 RELIGIOUS DIVERSITY AND COMMUNITY VOLUNTEERISM 241 In Models 2 through 7, the comparison group is the residual category of Asian Americans who subscribe to other religions. In these models only Buddhist and Hindu respondents were significantly different from Asian Americans in the other religion category. With the exception of controlling for religious voluntarism, both of these groups were significantly less likely to volunteer. Surprisingly, in Model 2, religious attendance had no significant effect in changing the likelihood of civic voluntarism relative to nominal attendance. In Model 3, among religiously affiliated Asian Americans, religious participation apart from worship services increased the likelihood of volunteering. In Model 4, religious giving also significantly increased the likelihood of nonreligious volunteering for religiously affiliated Asian Americans. In Model 5 in Table 2, religious membership had no significant effect in predicting the likelihood of nonreligious volunteering. Similarly, participation in a religious group apart from a house of worship (Model 6) had no effect on nonreligious volunteering. In Model 7, religious volunteering greatly increased the likelihood of nonreligious volunteering, and only Buddhist affiliation in this model significantly decreased the likelihood of volunteering. DISCUSSION AND CONCLUSIONS It is clear that religious and racial diversity in American society is increasing. Yet, we know little about whether greater diversity encourages or discourages social integration, members of different groups working together to build the broader civic institutions of American society. These findings move toward answering this larger question by exploring the relationship between religion and community volunteering for Asian Americans, a religiously diverse, nonwhite group that is typically ignored in studies of community participation. Surprisingly, a higher percentage of nonaffiliated Asian Americans volunteer when compared with Hindu and Buddhist Asian Americans, thus refuting our first hypothesis that religious Asian Americans would overall be more likely to volunteer. In addition, Asian Americans in the other religions category volunteer more than Hindus, Buddhists, and Catholics. Therefore, when we use Asian-American religious others as the comparison group (e.g., Models 2 7) we should expect that having a Hindu, Buddhist, or Catholic affiliation would decrease the likelihood of volunteering. Only Buddhist and Hindu Asian Americans, however, appeared significantly less likely to volunteer when compared with members of the other religions. One possible explanation is the organizational nature of Hinduism and Buddhism. Members of these two religions are not as likely to encounter one another in their houses of worship given the more individual nature of their practices (Kurien 2001), meaning that group-based pressures to volunteer may not occur as often for members of these two religions. Prema Kurien and other scholars who examine Hindu community life, in particular, suggest that such forms of religion are particularly vibrant among Asian immigrants in the United States (Kurien 1999; Kurien 2001). Our analyses reveal low levels of membership, participation, and attendance among Asian-American Hindus, however. Future research might consider such findings as important evidence that measures of religious participation used in current studies of religiosity are not accurately capturing the diversity of religious forms among Hindus and other members of non- Christian religions. If some non-christian religions are found to be less organizationally-based than are Christian religions in the United States this would perhaps discourage links with formal volunteer organizations; volunteer organizations may find it difficult to link with the semistructured religious environments of some non-christian religions. It is also possible that Asian-American Hindus and Buddhists may actually face a doubleminority status, as members of a minority racial group and a minority religion. For members of these traditions, in addition to having lower levels of religious involvement according to the indicators of religiosity used in the SCCBS, their practices may not translate very easily into voluntary contributions to the larger community. They may face even more difficulty than

10 242 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION African-American Christians who have a single-minority status in being asked to volunteer (Musick et al. 2000). We should point out here that, in part based on increasing immigration, Protestantism is no longer a numerical majority, a demographic reality that may change the cultural reality of what it means to be a religious person in the United States (Smith and Kim 2005). These kinds of nuances cannot be uncovered with the existing data but certainly point the way for future research. Neither of the previous explanations account for the lack of participation among Catholic Asian Americans, however. Perhaps Catholic Asian-American communities presently face internal concerns that detract from additional service outside the boundaries of the parish. Smith and Denton (2005) have argued that religious socialization and education is lacking in Catholic parishes. If this finding is applicable to nonwhite Catholic populations, it may also suggest that Catholic teachings of service to the wider community are not well conveyed to Asian-American Catholics. With minor exceptions, these results do not support our second hypothesis that human capital factors such as income and education would have the same influence on volunteering among Asian Americans as among the general population (Wilson and Musick 1997). In particular, Hindu Asian Americans, who in this sample had a greater pool of human capital as a group when compared with other religiously affiliated Asian Americans, were actually less likely to volunteer. Although these data do not allow us to compare immigrant and nonimmigrant Asian Americans, it is probable, given the high percentage of immigrants in Asian America (especially the high numbers of immigrants among Asian-American Hindus and Buddhists; see Jasso et al. 2003), that Asian Americans in the SCCBS were largely immigrant. It may be that immigrant factors, such as an inability to speak English fluently, mitigate human capital factors like high income, leading to Asian-American immigrants being less likely to volunteer than those in the general population. 15 Further research should more systematically test this hypothesis. These results point the way for future studies to account for greater religious diversity with more representative samples of small but growing minority religious groups. Including important demographic information such as immigrant generation and ethnic enclave proximity will benefit scholars and community leaders in understanding the contexts that religious and nonreligious Asian Americans inhabit and how these contexts enable or discourage civic participation. Such projects should also examine what kinds of identities and forms of participation lead to civic integration for Asian Americans, particularly for the growing numbers who are members of non- Christian religions. Scholars must think not only about the benefits these new religions provide their members, but also about the willingness of other organizations within civil society to form bridges with members of new religions. As we have demonstrated here, researchers should not assume emerging diverse communities will simply fit the models of civic participation found among other groups. Rather, the diversity in forms of participation among non-christian religions might generate different kinds of volunteer habits when compared with Christian religions. As the American population continues to become more diverse, it is all the more pressing to understand the complex influence of religion, race, and ethnicity on civic participation and the consequent development of the American community. ACKNOWLEDGMENTS Early versions of this article were presented at the 2003 Annual Meetings of the Society for the Scientific Study of Religion and the 2004 American Sociological Association Annual Meetings. We express our thanks to Monique R. Payne, Eugene Walls, and Robert Wuthnow, who read and commented on earlier drafts. NOTES 1. Care should be exerted when estimating the number of individuals in minority religious groups. There are very few surveys that specifically examine the percentage of individuals in the general population that adhere to a minority

11 RELIGIOUS DIVERSITY AND COMMUNITY VOLUNTEERISM 243 religion. Researchers who make such extrapolations often do so on the basis of very small numbers of survey respondents. See, for example, Jasso et al. (2003). 2. In general, participating in a religious community increases the likelihood of volunteering in one s local community. Yet, with respect to religious identification, research suggests that Catholic identification discourages community volunteerism relative to other Christian traditions (Lam 2002), while conservative Protestant identity encourages giving to the poor relative to self-identified Protestants or Catholics (Regnerus, Smith, and Sikkink 1998). When affiliation interacts with religious behavior such as church attendance, studies suggest that church attendance decreases conservative Protestants likelihood of volunteering (Wilson and Janoski 1995; Wuthnow 1999), while it increases the likelihood of mainline and Catholic volunteering (Wuthnow 1999). Other work finds that religious affiliation has no effect on the extent to which churchgoing Protestants volunteer (Park and Smith 2000). 3. Work by Chen (2002), Ecklund (2005a, 2005b), and Wuthnow and Hackett (2003) is notable exceptions to this statement. 4. While Judaism is also a significant minority religion in the United States, the SCCBS sampled only one Asian- American Jewish respondent, making it impossible to include Judaism as a distinct religious group in our analyses. Further, although Islam is a growing religion in the United States, there were not enough Muslims (N = 19) in the SCCBS to make individual analyses of this population possible. 5. The first tier included a random, representative sample of adult Americans and the second tier included a larger set of respondents from 30 communities in the United States, yielding a complete total of about 29,000 respondents. 6. An additional strength of the data set included interviews with some Chinese respondents in Cantonese. 7. See Social Capital Community Benchmark Survey Executive Summary, available at communitysurvey/docs/exec summ.pdf, for a further discussion of survey method. 8. In earlier analyses, we conducted bivariate logistic regression using ethnic categories alone and in combination with religious affiliation and found that Vietnamese and Cambodian respondents were somewhat less likely to volunteer, but when considered together with religious affiliation, had no effect on the dependent variable (figures available upon request). 9. According to our analyses of data from the 2000 Census, there are 31 million foreign born in the United States, of whom 26.4 percent are from Asia (8.2 million). The total Asian population in the United States is 11.9 million. Therefore, about 70 percent of the Asian population in the United States in 2000 was foreign born. 10. The specific question in the SCCBS read: I m going to list some of the types of organizations where people do volunteer work. Just tell me whether you have done any volunteer work for each in the past twelve months: (1) for health care or fighting particular diseases, (2) for school or youth programs, (3) for any organizations to help the poor or elderly, (4) for any arts or cultural organizations, (5) for any neighborhood or civic group. We also included responses to the following question: Which of the following things have you done in the past twelve months: worked on a community project? All response categories consisted of either yes or no with additional codes for those who volunteered don t know or refused to answer. 11. In an earlier analysis of the sum total of these measures we noted that responses were skewed to the left such that a large spike occurred for volunteering for one organization and dropped precipitously afterward. This created a nonnormal curve, which violated OLS assumptions that the dependent variable be normally distributed. To address this we reduced these responses to a simple dichotomous measure. 12. This scheme collapses the numerous denominational responses into larger traditions for ease of comparability while maintaining substantive sociological and theological differences among faith traditions (Steensland et al. 2000). This blunt approach at merging Protestant Asian Americans together may suppress any potential differences between volunteering for mainline Protestant and evangelical Protestant Asian Americans. To date, no survey has been able to document the actual distribution of Asian Americans by Protestant tradition, but ethnographic studies suggest that most Protestant Asian Americans are evangelical (Min and Kim 2002). 13. We recognize the concern of some that ethnic differences are coupled with religious affiliations based on a confluence of historical, cultural, and political factors. In analyses not presented here we explored the degree of association between religion and ethnicity and found only a few instances of significant overlap. For example, Muslim respondents (N = 19) identify as either Asian Indian (21 percent) or some other Asian nationality apart from the major ones available in the SCCBS (79 percent). Hindu respondents (N = 65) surprisingly were not largely Asian Indian (34 percent) but rather of some other Asian ethnicity (64 percent). These analyses suggest that there is only a minor association between religion and ethnicity and that the overlap is not an overwhelming one-to-one correlation between these religions and Asian-Indian ethnicity. The low count of Muslim respondents did not allow us to meaningfully explore ethnic differences between Muslims and Hindus. 14. Because of small sample sizes, these findings should be viewed with caution. 15. This insight was suggested, in part, by an anonymous reviewer. REFERENCES Barnes, J. S. and C. E. Bennett The Asian population: In Census 2000 Brief. Washington, DC: U.S. Census Bureau.

12 244 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION Becker, P. E. and P. H. Dhingra Religious involvement and volunteering: Implications for civil society. Sociology of Religion 62: Cadge, W. and E. H. Ecklund Religious service attendance among immigrants: Evidence from the new immigrant survey-pilot. American Behavioral Scientist 49:1 22. Chen, C The religious varieties of ethnic presence. Sociology of Religion 63: Ecklund, E. H. 2005a. Models of civic responsibility: Korean Americans in congregations with different ethnic compositions. Journal for the Scientific Study of Religion 44: b. Us and them: The role of religion in mediating and challenging the model minority and other civic boundaries. Ethnic and Racial Studies 28: Korean American evangelicals: New models for civic life. New York: Oxford University Press. Espiritu, Y. L Asian American panethnicity: Bridging institutions and identities. Philadelphia: Temple University Press. Jasso, G., D. S. Massey, M. R. Rosenzweig, and J. P. Smith Exploring the religious preferences of recent immigrants to the United States: Evidence from the new immigrant survey pilot. In Religion and immigration: Christian, Jewish, and Muslim experiences in the United States, edited by Y. Y. Haddad, J. I. Smith, and J. L. Esposito, pp Walnut Creek, CA: Alta Mira Press. Kitano, H. H. L. and R. Daniels Asian Americans: Emerging minorities. Englewood Cliffs, NJ: Prentice Hall. Kurien, P Gendered ethnicity: Creating a Hindu Indian identity in the United States. American Behavioral Scientist 42: We are better Hindus here : Religion and ethnicity among Indian Americans. In Religions in Asian America: Building faith communities, edited by P. G. Min and J. H. Kim. Walnut Creek: Alta Mira Press. Lam, P.-Y As the flocks gather: How religion affects voluntary association participation. Journal for the Scientific Study of Religion 41: Lien, P.-T., M. M. Conway, and J. Wong The contours and sources of ethnic identity choices among Asian Americans. Social Science Quarterly 84: Min, P. G. and J. H. Kim, eds Religions in Asian America: Building faith communities. New York: Alta Mira Press. Musick, M., J. Wilson, and W. B. Bynum Jr Race and formal volunteering: The differential effects of class and religion. Social Forces 78: Numrich, P. D How the swans came to Lake Michigan: The social organization of Buddhist Chicago. Journal for the Scientific Study of Religion 39: Okamoto, D. G Towards a theory of panethnicity: Explaining Asian American collective action. American Sociological Review 68: Ostling, R. N Survey of U.S. Muslims finds a growing faith. Intelligencer Record B13. Park, J. and C. Smith To whom much has been given... religious capital and community voluntarism among churchgoing Protestants. Journal for the Scientific Study of Religion 39: Putnam, R Bowling alone: The collapse and revival of American community. New York: Simon and Schuster. Regnerus, M. D., C. Smith, and D. Sikkink Who gives to the poor? The influence of religious tradition and political location on the personal generosity of Americans towards the poor. Journal for the Scientific Study of Religion 37: Smith, C., with M. L. Denton Soul searching: The religious and spiritual lives of American teenagers. New York: Oxford University Press. Smith, D. H Determinants of voluntary association participation and volunteering: A literature review. Nonprofit and Voluntary Sector Quarterly 23: Smith, T. W. and S. Kim The vanishing Protestant majority. Journal for the Scientific Study of Religion 44: Steensland, B., J. Z. Park, M. D. Regnerus, L. D. Robinson, W. B. Wilcox, and R. D. Woodberry The measure of American religion: Toward improving the state of the art. Social Forces 79: Wilson, J. and T. Janoski The contribution of religion to volunteer work. Sociology of Religion 56: Wilson, J. and M. Musick Who cares? Toward an integrated theory of volunteer work. American Sociological Review 62: Wuthnow, R Mobilizing civic engagement: The changing impact of religious involvement. In Civic engagement in American democracy, edited by T. Skocpol and M. P. Fiorina, pp Washington, DC: Brookings Institute Press Overcoming status distinctions? Religious involvement, social class, race, and ethnicity in friendship patterns. Sociology of Religion 64: Wuthnow, R. and C. Hackett The social integration of practitioners of non-western religions in the United States. Journal for the Scientific Study of Religion 42: Wuthnow, R., C. Hackett, and B. Y. Hsu The effectiveness and trustworthiness of faith-based and other service organizations: A study of recipients perceptions. Journal for the Scientific Study of Religion 43:1 17.