Neighbors in the Pews: Social Status Diversity in Religious Congregations

For the most part, Americans interact with other people like themselves—those with similar social and economic backgrounds. This homogeneity of social networks contributes in turn to social stratification and to the unequal distribution of social capital and civic integration. Religious congregations offer a rare opportunity for Americans to interact across social status lines. I use data from the 2001 U.S. Congregational Life Survey, which includes survey responses from relatively large samples of attendees nested within a large random sample of congregations, to examine the prevalence of income and education diversity in religious congregations. In contrast to racial diversity, which is minimal, there are high levels of social status diversity in most congregations. Status diversity in congregations also varies with congregational characteristics, such as religious tradition, age of the congregation, and racial makeup of the congregation; neighborhood characteristics, such as urbanity and proportion racial minority; and region of the country. I conclude by discussing the implications of the opportunities for cross-status interactions in religious congregations. Schwadel: Neighbors in the Pews 3 There is little diversity of social status in most American voluntary organizations (McPherson, Smith-Lovin, and Cook 2001). For the most part, people interact with others like themselves, that is, with people who have similar social and economic backgrounds (Byrne 1971; Feld 1982; Knoke 1986; McPherson and Smith-Lovin 1987). The lack of interaction between Americans of different social status contributes to social stratification and to the unequal distribution of social capital and civic integration (Popielarz 1999). Although most voluntary organizations are homogeneous in terms of social status, religious congregations could offer the opportunity for Americans to interact across status lines. Just as interracial congregations promote more amicable relations between AfricanAmericans and whites (Yancey 1999), social status diversity in congregations could contribute to better relations among people of different social strata. While low-status Americans are unlikely to participate in most voluntary organizations or to be politically active (Verba, Schlozman, and Brady 1995), there appears to be little social status difference in religious service attendance (Alston and McIntosh 1979; Hoge and Carroll 1978; Mueller and Johnson 1975). Of course, the fact that low-status Americans attend religious services does not mean that low-, middle-, and high-status Americans attend the same congregations. In this article, I ask whether religious congregations are, like most voluntary organizations, predominantly homogeneous in terms of social status. I empirically explore the prevalence of income and education diversity in U.S. religious congregations and factors that are associated with more or less diversity. I estimate measures of diversity in congregations using survey responses from relatively large samples of attendees nested within a large random sample of congregations. This analysis supplies the first generalizable estimates of social status diversity in U.S. congregations. In contrast to racial diversity, the results demonstrate that most religious congregations are highly status diverse, providing opportunities for Americans to interact with people from different social strata. WHY CONGREGATIONS SHOULD BE STATUS DIVERSE The traditional view is that “eleven o’clock on Sunday morning is the most segregated hour in America” (Wagner 1979: 9; see King 1958 for the original reference). Sociologists have long viewed religious congregations as homogenous social contexts (e.g., Lenski 1953, 1963; Niebuhr 1929; Pope 1942; Bryan Wilson 1969). According to McGavran (1980: 223), people “like to become Christians without crossing racial, linguistic, or class barriers.” More recently, Verba, Schlozman, and Brady (1995: 333) concluded that “religious congregations tend to be relatively socially homogeneous. Those who worship together are likely to share not only their faith but also their race or ethnicity and social class.” Although the view that religious congregations exhibit high levels of status 4 Interdisciplinary Journal of Research on Religion Vol. 5 (2009), Article 2 homogeneity is widespread, this proposition remains largely untested (for exceptions, see Dougherty 2003; Reimer 2007). Trends in denominational affiliation and neighborhood segregation suggest that congregations may be more status diverse than is generally assumed. Religious denominations in the United States have become increasingly status diverse over the last few decades (Roof and McKinney 1987; Wuthnow 1988, 1996). This does not mean that congregations within these denominations are necessarily status diverse; to assume so would be a form of the ecological fallacy (Robinson 1950), which assigns to individuals in a population the average characteristics of that population. Increasing denominational diversity, however, provides greater opportunities for status diversity in congregations. Evidence from residential segregation research also suggests that congregations are likely to be status diverse. Neighborhoods tend to be highly segregated by race, which inhibits racial diversity in religious congregations (e.g., Emerson and Kim 2003; Hadaway, Hackett, and Miller 1984). Conversely, neighborhoods are only moderately segregated by social status (Cook, Shagle, and Degirmencioglu 1997; Farley 1977; Jargowsky 1996). Since most Americans live reasonably close to the religious institutions they attend (Chaves 2004), relatively low levels of residential status segregation should translate into relatively high levels of status diversity in religious congregations. Other sociologists have proposed that religious congregations could be status diverse (e.g., Demerath 1965; Stark and Finke 2000). These authors note that case studies of specific congregations provide evidence of status diversity (e.g., Bultena 1949; Charles Lee Wilson 1945). In his recent comparison of congregational and denominational social class, Reimer (2007: 590) concludes, “Class matters at the congregational (and denominational) level, but class boundaries are porous and nonexclusive.” More directly relevant to the current research, Dougherty (2003) uses key informant data to show that congregations are more status diverse than race diverse. Although this research is suggestive of social status diversity in religious congregations, it is based on key informant data in which a congregational leader estimates characteristics of congregants. As recent research demonstrates, samples of congregants provide more reliable portraits of congregations than do key informant estimates (Schwadel and Dougherty 2010). The current research improves on previous research by analyzing diversity in congregations using data from samples of congregants and by examining factors that are associated with social status diversity in congregations. Schwadel: Neighbors in the Pews 5 FACTORS RELEVANT TO CONGREGATIONAL STATUS DIVERSITY In addition to establishing the mean level of social status diversity in congregations, I examine neighborhood, regional, and congregational characteristics that might influence congregational diversity. Previous research, using key informants’ estimates of the racial distributions in their congregations, suggests that congregations affiliated with certain religious traditions are more racially diverse than other congregations are (e.g., Dougherty 2003; Dougherty and Huyser 2008; Emerson and Woo 2006; Hadaway, Hackett, and Miller 1984). This might also be true of social status diversity. Catholic parishes tend to be large, and their placement is regionally circumscribed, meaning that Catholics who want to attend a Catholic church have fewer options for where to attend than do most Protestants. The Catholic Church’s distinctive mode of church planting should translate into relatively high levels of status diversity in Catholic parishes. In contrast to Catholic parishes, attributes of the Protestant community suggest that Protestant congregations are not very diverse (Emerson and Smith 2000). The lack of new members in most mainline Protestant denominations is producing an increasingly older, white mainline population (Roof and McKinney 1987), which could lead to low levels of status diversity in mainline congregations. Furthermore, the traditional association between lower social status and evangelical or conservative Protestantism (e.g., Demerath 1965; Niebuhr 1929) implies that there is relatively little status diversity in most evangelical congregations. Other congregational factors that are relevant to diversity in congregations include size, age, and congregational growth. Research on racial integration in congregations, for example, suggests that size is an important indicator (e.g., Northwood 1958). More attendees could create greater possibilities of diversity because of the change in scale. The age of the congregation might also influence diversity. Newer congregations, forming in a cultural context that emphasizes diversity, might seek to appeal to a diverse group of attendees, while the hereditary nature of affiliation with older congregations might lead to a more homogenous group of attendees. Conversely, church-sect theories suggest that congregations often form in response to the needs of a specific social class (Niebuhr 1929; Stark and Finke 2000), which would make newer congregations less status diverse than older congregations. Diversity might also be negatively related to congregational growth. For instance, Wagner (1979) suggests that diversity is antithetical to congregational growth because people prefer to attend religious services with other people like themselves. The remaining congregational factors that I explore measure the social status and racial composition of the congregations. Social status diversity in a congregation is clearly related to the social status makeup of the congregation. How social status diversity is related to the status distribution of the congregation, however, 6 Interdisciplinary Journal of Research on Religion Vol. 5 (2009), Article 2 remains a question. A larger than average proportion of attendees from the highest and lowest income and education categories could lead to greater status diversity, owing to the relatively small number of highand low-status people. On the other hand, if low-status and high-status Americans are largely segregated into their own congregations, then greater proportions of low-status and high-status attendees will lead to less status diversity. Social status diversity might also be positively related to racial diversity (Dougherty 2003). In addition to congregational factors, I explore three characteristics of the neighborhoods where the congregations are located that are likely to affect congregational diversity. First, social status in the United States varies considerably along the urban-rural continuum (Rodgers and Weiher 1988), and previous research suggests that urbanity positively affects racial diversity in congregations (e.g., Dougherty 2003; Dougherty and Huyser 2008; Emerson and Woo 2006), which may also be true for status diversity. The prevalence of minorities in the area is the second neighborhood factor. A high proportion of minorities in a neighborhood can lead to a racially homogenous congregation (Emerson and Woo 2006) and might also influence the social status distribution in the congregation. The third neighborhood characteristic is the geographic mobility of the residents. Geographic mobility varies by social status (Chesney, Wood, and Gombeski 1980). Thus greater geographic mobility in a neighborhood can influence status diversity in congregations. The final potential correlate of diversity is region of the country. The western portions of the United States are generally the least racially segregated (Farley and Frey 1994), and white Americans living in the West are more likely to report attending church with African-Americans than are white Americans living in the South, East, or Midwest (Hadaway, Hackett, and Miller 1984). These regional differences in racial diversity might influence regional differences in social status diversity in congregations.