The Immigrant Effect: Short-Term Mission Travel as Transnational Civic Remittance

The Immigrant Effect: Short-Term Mission Travel as Transnational Civic Remittance Abstract Short-term mission (STM) travel is a popular religious and civic practice done by religious congregations, but the local conditions that facilitate its production are poorly understood. We analyze organizational factors behind STM travel, with special focus on the role of recent immigrants within congregations. We use data from the third wave of the National Congregations Study. Our results show large differences by religious tradition, as well as the influence of foreign clergy, youth ministry, college-educated members, recent immigrants, and immigrant service orientation. We identify an immigrant effect, theorizing how immigrant presence and identity influence U.S. congregations’ transnational engagement, especially within religious traditions with relatively low levels of recent immigrants. By connecting research on congregational civic engagement with that on transnational immigrant religion, we argue that about 30% of STM travel is a form of civic remittance in which recent immigrants and their U.S. congregations aid foreign communities. Short-term mission (STM) travel by U.S.-based congregations to international locations has received markedly increased scholarly attention in recent years (Adler and Offutt 2018; Bakker 2014; Beyerlein et al. 2011; Howell 2012; Priest et al. 2006; Wuthnow 2009). This large movement of people and resources facilitated by congregations has been presented as a flow of civic, religious transnationalism, primarily moving in the opposite direction of another well-known transnational flow, immigration (Adler and Offutt 2018; Offutt 2015; Offutt and Miller 2016; Priest et al. 2006; Wuthnow and Offutt 2008). Yet, no research has systematically examined the factors that differentiate why some U.S. congregations produce such distinctive transnational civic engagement and others do not. In this article, we argue that this general lacunae, in turn, has led scholars to miss the role that immigrants have in prompting transnational civic engagement by some U.S. congregations. Instead of two disconnected flows passing each other by, we argue that immigration and STM travel are two related aspects of religious transnationalism. In the past two decades, a body of transnational immigration research has shown the range of personal ties that immigrants retain to sending communities and has demonstrated the role of U.S. religious congregations in shaping immigrant assimilation (Cadge and Ecklund 2007; Ebaugh and Chafetz 2002; Hirschman 2004; Kurien 2013, 2014; Levitt 2001, 2007; Menjívar 2010; Warner 2007; Warner and Wittner 1998; Yang and Ebaugh 2001). Within this growing literature, some studies have pointed to the limited civic engagement propensities of immigrant congregations, noting that their patterns of informal, localized services tend to bond immigrants together (Cadge 2008; Ebaugh and Pipes 2001; Hirschman 2004; Levitt 2008; Ley 2008). At the same time, other studies have demonstrated the sizeable engagements that immigrants have with their sending communities, particularly through social and economic remittances (Kurien 2014; Levitt 1998; Menjívar 1999; Vertovec 2004). While these two sets of findings reveal much about immigrants and congregational activity, together they minimize the role of congregations in facilitating immigrants’ formalized transnational engagement, while also neglecting ways that immigrants may motivate transnational engagement among congregations more generally. We translate concepts from research on congregational civic engagement to analyze why congregations direct substantial organizational focus and resources to STM travel. Our results show significant, sizeable differences in STM trip production by a number of factors, including the presence of recent immigrants and organizational styles of serving immigrants. We theorize that these results point to an immigrant effect: the power of immigrants’ transnational connections, identities, and motivations to shape U.S. congregations’ transnational civic activity (Schnable 2015b). For nearly one-third (30%) of congregations, then, STM travel is an extension of a transnational process already set in motion by immigrants themselves. Given this, we suggest that STM travel and other transnational civic activities by U.S. congregations could be understood as a form of civic remittance when immigrants are involved. Similar to other remittances, civic remittances are a resource that immigrants can use to influence the material, cultural, and religious life of their home communities (Levitt 2001). This research contributes to an understanding of congregations as producers of civic action beyond their locales, adding to knowledge about the political, social, and cultural impact of religious organizations on public life. In particular, our research addresses a problem identified by Numrich and Kniss (2007:220), that “civic engagement patterns have garnered less attention from scholars of immigrant congregations than from scholars of nonimmigrant congregations.” Part of our contribution is to expand research attention about immigrants and congregational civic engagement, in both the “immigrant congregation” setting and beyond. We also expand knowledge about how religious transnationalism works by showing the connection between two transnational flows of people that have tended to be analyzed separately (though see Schnable 2015b). In a recent review of religious transnationalism, Offutt and Miller (2016:536) argued that transnational activities like STM travel are “a genuinely new wrinkle in how [U.S.] congregations operate and [they] should force sociologists to rethink what we know about the cultures and environments of local religious organizations.” Our research sheds light on this “new wrinkle,” showing how religious transnationalism might influence local congregations through immigrants, but also how local U.S. congregations in turn help construct religious transnationalism. BACKGROUND OF STM TRAVEL STM travel by U.S. residents through congregations was nearly nonexistent in 1950s (Howell 2012; Wuthnow 2009). The expansion of mass travel infrastructure, the emergence of diversified transnational religious organizations (Bakker 2014; McCleary and Barro 2008), and shifts in global religious demographics (Wuthnow 2009) increased the ease of STM travel, while changes in religious discourse about foreign others (Hefferan 2007; Howell 2012) and the increased valuation of identity-shaping religious practices among religious Americans (Bakker 2014; Trinitapoli and Vaisey 2009) led to a drastic increase in STM popularity. By the mid-2000s, STM travel engaged 1.6 million U.S. church-going adults each year (Wuthnow 2009). This is likely an undercount of the actual total of congregation-based traveling religious persons since it does not include adults that attend worship services less regularly or those under the age of 18 to whom trips are often targeted (Priest et al. 2006; Probasco 2013). Indeed, using retrospective data, Wuthnow (2009) showed that STM travel by teenagers has increased five-fold since 1980s. As further evidence of STM’s popularity and familiarity, the National Study of Youth and Religion (NSYR) indicates that in 2002, 29% of U.S. teenagers between the ages of 13 and 17 reported having gone on a religious mission trip of some type (Trinitapoli and Vaisey 2009). STM travel is differentiated from other modes of religious travel by a reliance on local U.S. religious organizations to mobilize traveling groups (Adler and Offutt 2018; Priest et al. 2006); by interactions with foreign communities through local religious organizations native to foreign communities (Bakker 2014; Wuthnow 2009); and by the usage of partnership language to articulate a form of moral relationship (Adler and Offutt 2018; Bakker 2014; Wuthnow 2009). Ammerman (2005) characterizes this direct, local-to-local pattern as “pervasive” among American congregations, with denominational, parachurch, and international aid organizations helping to broker connections between local organizations and foreign communities (Kinney 2015; Wuthnow 2009). One common feature of STM travel is the delivery of aid, whereby “the majority of trips involve some kind of service project in an impoverished community” (Probasco 2013:204), as well as finanical capital, educational materials, and/or medical supplies to facilitate these projects (Adler and Offutt 2018). Alongside the delivery of aid, STM trips may also involve “learning about cultural and social justice issues, and[/or] engaging in evangelization efforts among local populations” (Beyerlein et al. 2011:781). What this STM activity accomplishes for foreign organizations and communities—beyond charitable contributions and the ideal of religious partnership—is a topic of continued debate and depends on differences within the broad category (Adler and Offutt 2018; Hefferan 2007; Offutt 2011; Priest et al 2006; Ver Beek 2006; Wuthnow 2009). Our purpose is to focus on the organizational characteristics that may influence congregational STM travel to begin with. In the following section, we introduce four such factors. FACTORS INFLUENCING CONGREGATION-BASED STM TRAVEL Religious Tradition Religious tradition categorizes congregations according to shared historical formation and social ideas (Steensland et al. 2000; Woodberry et al. 2012). Religious tradition has been shown to influence a range of congregation-based social service activity and political engagement (Barnes 2011; Beyerlein and Chaves 2003). Religious tradition matters for STM production because traditions combine distinct orientations to “the world,” preferred modes of action for transforming “the world,” and different organizational structures that facilitate international engagement. Basic correlations provide evidence that religious tradition influences STM trip production. The NSYR suggests variation in traveler participation by religious tradition, with Latter-Day Saint, Mainline Protestant, and Conservative Protestant youth more likely to travel (Trinitapoli and Vaisey 2009). Beyerlein et al. (2011) report in a footnote using 2006 National Congregations Study (NCS) data that, without controls, Evangelical Protestant congregations reported the highest annual rate of STM travel. Similarly, using representative data from congregations in Arizona, Adler and Offutt (2018) report that Evangelical Protestant congregations were the most likely to travel within the previous year as well as over a 5-year period. Here, we briefly theorize how religious tradition may structure STM travel variation. After the mid-20th century, “many Mainline Protestant denominations were out of the overseas missions business entirely…” (Ammerman 2005:293), which meant less opportunity for foreign connections. The Mainline Protestant style of foreign engagement shifted away from emphasizing the export of religious belief to the provision of foreign relief. This relative disengagement of personal encounter by Mainline Protestants suggests that congregations in the Mainline Protestant tradition may not produce STM travel with the highest frequency (Bakker 2014). However, because Mainline Protestant denominations and organizations exist throughout the developing world (Lynn 2016; McCleary and Barro 2008), and because congregant interest in international affairs is still high (Ammerman 2005; Schnable 2015b), we expect that Mainline Protestant congregations will produce short-term travel at a moderate level compared to other traditions. Also in the mid-20th century, the growing Evangelical Protestant tradition massively invested in foreign missionary fields, part of a renewed proselytization strategy (Ammerman 2005; Wuthnow 2009). Numerous well-funded, U.S.-based organizations developed to facilitate foreign engagement by Evangelical Protestant congregations (Thaut 2009). Today, Evangelical Protestant congregations support international work at relatively high rates through both denominational and parachurch structures (Ammerman 2005), even though individuals are significantly less likely to donate to international aid (Schnable 2015b). As Howell (2010) recounts in a partial history, the STM idea was embraced by Evangelical Protestant religious groups as a method of recruiting future missionaries (though that specific purpose generally failed), reflecting an individualist aid logic (Schnable 2015b). With the demographic center of Christianity shifting to the Global South, U.S. Evangelical Protestant religious groups increasingly have religious international peers (co-religionists) to connect with (Bakker 2014). McAlister (2008) characterizes this contemporary global focus among U.S. evangelicals as an “enchanted internationalism” in which foreign evangelicals are seen as partners in mission activity, but also spiritual authorities given their social sufferings. We expect Evangelical Protestant congregations to organize STM travel at the highest rates compared to other traditions. Catholic congregations in the mid-20th century United States shifted attention to overseas relief, particularly as dioceses and religious orders mobilized a missionary movement in Central and South America (Smith 1996). This movement was supported by prioritization of international development in Roman Catholic social encyclicals and the increased centrality of U.S.-based Catholic organizations and their monies in international development (Calderisi 2013). Catholic international organizations (e.g., Catholic Relief Services) and religious orders facilitated massive flows of financial and organizational support to the developing world, especially helping to build the capacities of foreign native Catholic organizations (Calderisi 2013). Local Catholic parishes have been an important financial source in this movement. As a result of this structure and legacy, Wuthnow (2009) reports that Catholic congregations are leaders in the support of international hunger relief and refugee work, while Schnable (2015b) reports that individual Catholics are relatively more likely to donate to international aid and support Catholic international efforts. However, parishes appear to transmit financial support for foreign activity through professionalized organizations (Ammerman 2005) and Catholic congregations are characterized by deflated rates of civic involvement by members (Bane 2005). We expect Catholic congregations to have a comparatively low rate of STM travel. Beginning in the mid-19th century, Black Protestant denominations and missionary societies actively sponsored foreign missionaries, primarily to Africa and the Caribbean (Cornelius 2002; Killingray 2003; Lincoln and Mamiya 1990). By World War I, however, Black Protestant denominations, which had been instruments of foreign-focused involvement for congregations, scaled down foreign-oriented missionary work (Ammerman 2005). This withdrawal was partially in response to the domestic pressures of African American organizational life under Jim Crow and partially due to colonial authorities in Africa seeking to prevent ideals of racial equality or pan-African solidarity (Jacobs 1980; Killingray 2003). Despite declines in missionary activity, “most Afro-Americans learned of Africa through their churches or missionaries who had been stationed in Africa” (Jacobs 1980:167). And, as a result, tens of thousands of congregations affiliated with U.S.-based Black Protestant denominations were formed in other countries, especially on the continent of Africa (Lincoln and Mamiya 1990). One exception to the pattern of post-World War I decline among Black Protestant groups was the growth of global missions activity by Pentecostal and Holiness groups beginning around 1900 (Case 2006). During the interwar years, the Holiness focus on redeeming Africa even laid the groundwork for a major domestic political movement like Garveyism (Roll 2010). There is some evidence that the Holiness tradition within Black Protestant Christianity even increased foreign mission activity after World War II (Cornelius 2002). Still, up until the present day, the percent of African American missionaries among the total Christian missionary population never came close to matching the percentage of African Americans in the U.S. adult population (Killingray 2003). Recent research on Black Protestant congregations suggests that, despite a strong public commitment to evangelism, their external activities still tilt relatively toward involvement in domestic activities that are directly related to overcoming the political and economic inequality of Black communities (Ammerman 2005; Barnes 2005). While there has been renewed emphasis on cooperative “foreign missions” activity between Black Protestant denominations, as well as some evidence of embracing STM (Sutherland 2003), we expect Black Protestant congregations to have a comparatively low rate of STM travel. Among non-Christian congregations, the activities and dynamics associated with “STM travel” likely occur in different styles. For minority status religious groups in the United States, such travel may incorporate visits to religious homelands alongside aid provision. For example, travel is widely used for Jewish identity construction among U.S.-based Jewish groups like Birthright Israel (Kelner 2010). Because of a less extensive institutional structure, we expect relatively lower rates of travel among non-Christian congregations in general. One specific aspect of any religious tradition that may foster transnational interaction is the congregational hosting of foreign clergypersons. These clergypersons may be missionaries, aid workers, or heads of foreign congregations (Wuthnow 2009), present in the U.S. for education, awareness raising, or fundraising (Offutt 2015). Similar to missionary speakers, they can cultivate knowledge and motivation for international aid (Schnable 2015b). Their presence varies by religious tradition and may help account for differences in prompting transnational engagement. To summarize, we expect that rates of STM travel will vary by religious tradition with the greatest likelihood of travel among Evangelical Protestant congregations, followed by Mainline Protestant, Catholic, Black Protestant, and Non-Christian congregations (in that order). We expect that congregations that host foreign clergypersons will be more likely to produce STM travel. Resources STM travel requires an extensive set of resources to be mobilized. For example, Wuthnow (2009:180) reports that the average short-term trip costs “at least $1,000 per person….” Three types of resources are likely to influence STM travel by congregations. Congregational financial funds are a main resource for STM travel, particularly because of their size and fungible character. Ammerman (2005) reports that congregations with small budgets are disproportionately uninvolved with foreign engagement activity. We expect that congregations with greater organizational income will be more likely to produce STM travel. While congregations themselves may provide funds for STM travel, case studies of individual STM trips show that the costs of travel are sometimes treated as external to organizational budgets, with would-be travelers required to fundraise from family, friends, and other congregants (Howell 2012; Wuthnow 2009). This would make STM travel especially available to congregations with wealthier members. We expect that congregations with greater levels of high-income persons will be more likely to produce STM travel because of greater extra-congregational financial resources. Increased congregational staff could also facilitate trip production since international travel requires extensive planning (Wuthnow 2009). Since STM travel is especially popular among youth (Trinitapoli and Vaisey 2009), a youth minister or official youth leader would likely facilitate trip production (Wuthnow 2009). We expect that congregations with more staff members as well as those with role specialization in a youth minister will be more likely to produce STM travel. Congregants’ Age and Education According to descriptions of STM trips, congregants’ age and educational profiles may influence congregational trip production (Ammerman 2005; Howell 2012; Probasco 2013; Wuthnow 2009). Teenagers have biographical availability at numerous times during the year and are considered to be at a prime developmental age for such a religious experience (Smith 2005). Young adults are also relatively biographically available and increasingly have familiarity with STM travel from their teenage years (Howell 2012; Ver Beek 2006; Wuthnow 2009). Wuthnow (2009) reports that young adults are twice as likely as middle-aged adults to go on such a trip. We expect that congregations with higher levels of teenagers as well as higher proportions of young adults will be more likely to produce STM travel. A second demographic characteristic, educational level, is correlated with trip participation, such that college education doubles the likelihood of an individual going on a STM trip (Wuthnow 2009). Educational level may indicate an orientation toward global cosmopolitanism or a desire to learn about global others (Snee 2013). We expect that congregations with higher-educated congregants will be more likely to produce STM travel. Immigrants in Congregations and Elements of Immigrant Congregations While previous research has suggested the connection of the preceding factors to STM travel, we know of no research that fully theorizes the connection of immigration and STM travel. We provide that here. Immigrant groups arriving in the United States since 1965 have been more religiously and ethnically diverse (Kivisto 2016; Warner and Wittner 1998; Yang and Ebaugh 2001). As many scholars have noted (Ebaugh 2004; Warner and Wittner 1998), the frequency, density, and diversity of transnational interactions of post-1965 immigrants to the United States are distinct. Writing in the late 1990s, Menjívar (1999:592) noted that “many of today’s immigrants can easily remain active in their homeland communities through travel and/or other means, such as remittances, telephone and video conferencing, and continued streams of immigrants….” Levitt’s (1998, 2001) landmark study in the Boston area helped to define the transnational, religious dimension of recent immigration by conceptualizing what she called social remittances: cultural elements of life, such as norms and practices, that could be carried by individuals, especially for intentional projects of creating transnational community. Extending this dynamic, Kurien (2014) has shown the sizeable “direct effect” immigrants can have on their sending communities, particularly in how they transform organizations in their homelands. Surprisingly, the small amount of research on transnational civic engagement by U.S. congregations that includes information about immigrants is not focused on ideal-typical “immigrant congregations.” For example, Alison Schnable (2015b) showed the self-reported giving of international aid by regular churchgoers is likely influenced by the presence of immigrants in a congregation. She suggested that “[i]mmigrant members of congregations can be sources of information about global problems and can open broader channels of collaboration…” (Schnable 2015b: 91). Bakker’s (2014) study of “twinning” relationships between United States and foreign congregations shows both of these possibilities. Though Bakker was not directly interested in this pattern, four of the twelve congregational cases in her research, across three different religious traditions, showed evidence of immigrant members playing a key role in establishing a “twinning” relationship. For example, a suburban Catholic Church comprised of upper-middle-class families twinned with a parish in Haiti due to an immigrant with continued Haitian relationships. A suburban, middle-class Mainline Protestant church began its twinning relationship through the efforts of a Liberian refugee family. Summarizing his cases of transnational engagement, Wuthnow (2009:144) noted that some “churches initiate direct programs in particular locations because an immigrant from that country becomes a member of their congregation.” From this evidence, we expect congregations with any amount of immigrants to be more likely than congregations with no immigrants to produce STM travel. Further, we expect that a specific type of immigrant—foreign-born clergypersons—will make congregations more likely to produce STM travel (Bakker 2014; Offutt 2015; Schnable 2015b). These hypotheses are about immigrants within congregations in general, but a close read of research literature about “immigrant congregations” reveals that many of those facilitate transnational civic activity by their members as well. For example, Levitt (1998:78) noted the presence of transnational “adopt-a-parish-type relationships” that offered “financial and material support” abroad. Menjívar (1999:604) described a congregation that “coordinated a project in El Salvador to help orphan children” as well as sponsored “missionary groups” that visited El Salvador. Ebaugh (2004) reported that immigrant congregations often supported their sending communities. Vietnamese Buddhist and Catholic congregations sent aid to “churches and temples in Vietnam” (Ebaugh 2004:223). The types of aid that these congregations sent abroad were distinct: targeted toward specific location and organizations, but not “religiously relevant” (Ebaugh 2004:175). Aid included items like financial capital, technology, technological expertise, and books. In each of these examples, the reported transnational activity is strikingly similar to STM travel: the movement of persons and resources from U.S.-based congregations to foreign religious organizations for localized aid beyond strictly “religious” activity. One conclusion from this evidence would be that “immigrant congregations” are more likely to produce transnational STM travel. However, there is debate about what characterizes an “immigrant congregation.” For example, Cadge (2008) has critiqued the category of immigrant congregation as collapsing heterogeneity. Chafetz and Ebaugh (2002) have advocated moving away from an ideal-typical approach to immigrant-focused congregations. Even moreso, Cadge and Ecklund (2007) suggest including various organizational characteristics as independent variables to adequately address the diversity of processes within the category of “immigrant congregation.” To this end, we focus on two dimensions of immigrants and congregations: immigrant generation and congregational style of serving immigrants. Research on immigrants, congregations, and civic engagement has often tied these two dimensions together. For example, when first-generation immigrants dominate a congregation, research suggests formal organized civic engagement may be weak. Ley (2008) characterized first-generation immigrant congregations as “hubs” that were not civically engaged, instead providing informal services like clergy counseling or job referrals to immigrant congregants. Ebaugh and Pipe’s (2001) analysis of first-generation immigrant religious groups in the Houston area concluded that few had organized social services, instead providing support through informal networks. In this line of interpretation, first-generation immigrants produce an internally focused, resource-poor congregation that avoids external engagement. However, we expect immigrant generation to influence the activity of STM travel in a more complex way. As hypothesized above, their presence in a congregation increases the possibility of congregation-based STM travel. There may be a tipping point, though, at which a high proportion of first-generation immigrants in a congregation might hinder transnational engagement and thus depress STM travel. For example, Ebaugh et al. (Chafetz and Ebaugh 2002; Ebaugh 2004) report that congregations defined by the heavy presence of first-generation immigrants often receive foreign support in early years, instead of sending it.1 These are the immigrant hub congregations that may be solely focused on domestic settlement and assimilation work. The case research suggests that the presence of the second generation—the children of original immigrants—or other nonimmigrant groups within a congregation, influences congregations to reach beyond the needs of the congregational community (Ebaugh 2004; Ebaugh and Pipes 2001; Ley 2008). In other words, a congregation with recent immigrants, but that is not dominated by such immigrants, should be more likely to produce STM travel. In the one study to examine immigrant generation and congregational STM travel activity, albeit without multivariate controls, Wuthnow (2009) found that high levels of first-generation immigrants were unrelated to STM travel, but that low levels of first-generation immigrants were significantly, positively associated with STM travel. Given this, we expect that congregations with any recent immigrants, when compared to congregations with no such immigrants, will be more likely to produce STM travel. We also expect that congregations with a higher proportion of recent immigrants will not be more likely to do so. The second dimension of congregations related to immigrants that we expect to influence STM production is the style of congregational service to immigrants, independent of the length of residency of immigrants or the presence of multiple generations of immigrants. Case studies suggest that congregations with an internal, informal type of focus may limit external activities, even transnational activities (Ebaugh and Pipes 2001; Menjívar 1999). We expect that congregations that serve immigrants solely through informal, internally focused services will be less likely to produce transnational STM travel because of an organizational identity focused on the immediate, localized needs of immigrants. By contrast, we expect congregations with formal, externally focused services to immigrants to be more likely to produce transnational STM, an example of a bridging identity attentive to social needs and community beyond the congregation (Kurien 2013). We expect congregations with no services to immigrants to be the least likely to produce STM travel, as they have no prioritized focus on the transnational identities and relationships of immigrants. DATA, MEASURES, AND METHOD We use data from the 2012 NCS (Chaves and Anderson 2014).2 The 2012 NCS provides an extensive set of variables relevant to STM travel and immigrants. The NCS is a representative sample of congregations constructed through respondents to the 2012 General Social Survey. Congregations were identified through a hypernetwork sample in which GSS respondents that reported attending religious services at least once a year were then asked to report the name and location of their religious congregation. The resulting set of congregations was contacted to participate in a key informant survey. Key informant surveys, including the NCS, have been validated as an accurate way to collect congregational information, particularly information related to directly observable characteristics instead of attitudes (Frenk et al. 2011). The response rate for the 2012 NCS is 73%.3 Dependent Variable We use a dichotomous variable that indicates whether in the previous 12 months a congregation had a group “to travel to another country to provide assistance to people in need” (0 = No, 1 = Yes). A particular strength of this question is its concentration on aid-focused travel—a common element of STM travel—without Christian-centric “mission” terminology. Independent Variables Religious tradition hypotheses are based off of Steensland et al. (2000, Woodberry et al. 2012). Religious tradition categories include White Evangelical Protestant, Catholic, White Mainline Protestant, Black Protestant, and non-Christian. The final category is heterogeneous, including Jewish, Muslim, and other non-Christian groups. The default NCS scheme categorizes a congregation as Black Protestant if it is affiliated with a traditionally Black denomination/movement or if it is affiliated with a White Evangelical or White Mainline denomination and has a membership that is 80% or more African American/Black. Out of concern that default coding merges congregations from the original Black denominations that may have distinct foreign activity patterns with Black majority congregations in other denominations, we recoded the latter congregations into their appropriate religious tradition categories.4 We measure foreign clergy speakers by whether a congregation indicated it “hosted a clergyperson or preacher who lives in another country” in the previous 12 months (0 = No, 1 = Yes). Our resource hypotheses use three variables. The first is financial resources, from the question, “What is the total amount of money your congregation received in income from all sources during your most recent fiscal year?” Answers in whole dollars were divided by $10,000. The second is the presence of high-income individuals. We use the percentage of regularly participating adults that live in households with incomes higher than $140,000 a year. The third is a dichotomous variable (0 = No, 1 = Yes) from the question, “Do you have a youth minister or other leader specially designated to coordinate activities for youth?” Hypotheses for membership demographics use three measures. For the measure of the presence of teenagers, we use the number of regularly participating teenagers “between 13 and 19 years old” in the congregation. For the measure of young adults, we use “What percent of the regular adult participants would you say are under 35 years old?” For a measure of college-educated adults, we use “About what percentage [of regular adult participants] would you say have four-year degrees or more?” Our hypotheses related to immigrant characteristics use three measures. To indicate whether a clergy member is foreign born, we reverse-coded the question, “Was the [head clergyperson] born in the United States?” to indicate foreign birth (0 = No, 1 = Yes). For a measure of immigrant generation, we draw from the question, “Of regular adult participants in your congregation, what percent would you have come to the United States within the past five years?” This measure indicates the proportion of the congregation comprised of first-generation individuals that recently immigrated. From this, we created a three-category measure. Our base category includes congregations with zero recent immigrants. Based on Wuthnow’s (2009) evidence about “low numbers” of immigrants, our next category indicates congregations with up to 5% recent immigrants. We created a final category that represents a congregation with a moderate to high portion of first-generation immigrants (5% ≤ 100%).5 Our final immigrant-related measure is built from three variables: whether in a congregation a “group met to offer services for immigrants, such as legal assistance, translation, or job assistance” (0 = No, 1 = Yes), whether the congregation had a group to teach English (0 = No, 1 = Yes), and whether the congregation answered an open-ended question about social service and human service programs by indicating a program for immigrants, migrants, or refugees (0 = No, 1 = Yes). The first question is an indicator of informal service provision oriented toward individual immigrant cases; the second is an indicator of internal service that aids assimilation; and the third indicates an organized, formal program that targets those outside the congregation. Congregations were coded as “internal” servers if they indicated “yes” to either or both of the first questions, but “no” to the third question; “external” servers if they indicated “yes” to the third question; and “nonservers” if they indicated “no” to all. These questions are not focused toward “recent” immigrants, so they may measure services provided to any immigrant groups, including first-generation immigrants with long residency as well as second-generation immigrants. Control Variables We control for congregational size using the number of regular adult attenders (divided by 100). Theological orientation is based on a set of dummy variables indicating the theological leanings of a congregation: “liberal,” “in the middle,” or “conservative,” which serves as the base category. To control for denominational influence, a dummy variable shows whether a congregation is affiliated with a denomination (0 = No, 1 = Yes). To control for general congregational external engagement, we indicate whether a congregation has any reported social service program (0 = No, 1 = Yes). We also control for percent of members that are over the age of 60, percent that are white, and percent that are female. We control for location in two ways. The first is by the inclusion of dummy variables gauging urban, suburban, or rural (reference category) location. The second is the inclusion of dummy variables indicating regional location, with Northeast/Atlantic as the reference category. Analytic Method The data are missing at random (MAR), with the income variable having the most missing values at 17% of cases. To retain as much case information as possible, missing values were imputed in Stata 14 using chained equations (White et al. 2011), which allow for models appropriate to diverse variable types (categorical, continuous, ordinal). The imputation model included all variables used in regression analyses in addition to variables associated with salient aspects of congregational difference (denominational affiliation, moral boundaries, context). Of 1,331 cases in the 2012 NCS, we were able to reliably impute missing data for all variables in 1,280 cases. All results are presented using 20 imputed data sets.6 We present univariate and bivariate statistics using a congregation-level weight that adjusts for sample selection probability due to size. Next, we use logistic regression to model the effect of variables on the likelihood of STM travel production. Our regression analysis is presented with no weighting, having passed the test from DuMouchel and Duncan (1983).7 RESULTS Table 1 presents all variables used in the analysis. In 2012, 27% of U.S. congregations reported STM travel.8 Table 1 Univariate Descriptive Statistics   Mean  SD  Foreign STM travel  .27  .45  Religious tradition   Roman Catholic  .06     White Evangelical Protestant  .48     Black Protestant  .18     White Mainline Protestant  .22     Non-Christian  .07     Hosted foreign clergy  .30  .46  Resources   % Adults in high-income households  6.93  12.64   Income (×$10000)  53.58  212.80   Youth minister/leader  .54  .50   No. of full-time staff  3.02  11.75  Demographics   % Adults with BA degree  32.18  27.13   % Young adults  24.79  18.24   No. of teenagers  22.31  71.56  Immigrant-related   % Recent immigrants: 0  .81     % Recent immigrants: 0 to 5  .14     % Recent immigrants: > 5  .05     No services for immigrants  .89     Internal service for immigrants  .01     External service for immigrants  .10     Foreign-born head clergy  .09  .29  Controls   No. of regular attenders (×100)  1.18  3.23   Has denomination  .76  .42   Has a social service program  .82  .38   Conservative theology  .63     Moderate theology  .25     Liberal theology  .12     % Senior adults  36.62  24.82   % White  62.46  42.47   % Female  61.98  12.85   Urban location  .50     Suburban location  .18     Rural location  .32     Mid-Atlantic/Northeast  .12     Midwest  .23     South  .51     Mountain/Pacific  .14      Mean  SD  Foreign STM travel  .27  .45  Religious tradition   Roman Catholic  .06     White Evangelical Protestant  .48     Black Protestant  .18     White Mainline Protestant  .22     Non-Christian  .07     Hosted foreign clergy  .30  .46  Resources   % Adults in high-income households  6.93  12.64   Income (×$10000)  53.58  212.80   Youth minister/leader  .54  .50   No. of full-time staff  3.02  11.75  Demographics   % Adults with BA degree  32.18  27.13   % Young adults  24.79  18.24   No. of teenagers  22.31  71.56  Immigrant-related   % Recent immigrants: 0  .81     % Recent immigrants: 0 to 5  .14     % Recent immigrants: > 5  .05     No services for immigrants  .89     Internal service for immigrants  .01     External service for immigrants  .10     Foreign-born head clergy  .09  .29  Controls   No. of regular attenders (×100)  1.18  3.23   Has denomination  .76  .42   Has a social service program  .82  .38   Conservative theology  .63     Moderate theology  .25     Liberal theology  .12     % Senior adults  36.62  24.82   % White  62.46  42.47   % Female  61.98  12.85   Urban location  .50     Suburban location  .18     Rural location  .32     Mid-Atlantic/Northeast  .12     Midwest  .23     South  .51     Mountain/Pacific  .14    Source: 2012 National Congregations Study. Imputed data (m = 20); weighted. View Large Table 1 Univariate Descriptive Statistics   Mean  SD  Foreign STM travel  .27  .45  Religious tradition   Roman Catholic  .06     White Evangelical Protestant  .48     Black Protestant  .18     White Mainline Protestant  .22     Non-Christian  .07     Hosted foreign clergy  .30  .46  Resources   % Adults in high-income households  6.93  12.64   Income (×$10000)  53.58  212.80   Youth minister/leader  .54  .50   No. of full-time staff  3.02  11.75  Demographics   % Adults with BA degree  32.18  27.13   % Young adults  24.79  18.24   No. of teenagers  22.31  71.56  Immigrant-related   % Recent immigrants: 0  .81     % Recent immigrants: 0 to 5  .14     % Recent immigrants: > 5  .05     No services for immigrants  .89     Internal service for immigrants  .01     External service for immigrants  .10     Foreign-born head clergy  .09  .29  Controls   No. of regular attenders (×100)  1.18  3.23   Has denomination  .76  .42   Has a social service program  .82  .38   Conservative theology  .63     Moderate theology  .25     Liberal theology  .12     % Senior adults  36.62  24.82   % White  62.46  42.47   % Female  61.98  12.85   Urban location  .50     Suburban location  .18     Rural location  .32     Mid-Atlantic/Northeast  .12     Midwest  .23     South  .51     Mountain/Pacific  .14      Mean  SD  Foreign STM travel  .27  .45  Religious tradition   Roman Catholic  .06     White Evangelical Protestant  .48     Black Protestant  .18     White Mainline Protestant  .22     Non-Christian  .07     Hosted foreign clergy  .30  .46  Resources   % Adults in high-income households  6.93  12.64   Income (×$10000)  53.58  212.80   Youth minister/leader  .54  .50   No. of full-time staff  3.02  11.75  Demographics   % Adults with BA degree  32.18  27.13   % Young adults  24.79  18.24   No. of teenagers  22.31  71.56  Immigrant-related   % Recent immigrants: 0  .81     % Recent immigrants: 0 to 5  .14     % Recent immigrants: > 5  .05     No services for immigrants  .89     Internal service for immigrants  .01     External service for immigrants  .10     Foreign-born head clergy  .09  .29  Controls   No. of regular attenders (×100)  1.18  3.23   Has denomination  .76  .42   Has a social service program  .82  .38   Conservative theology  .63     Moderate theology  .25     Liberal theology  .12     % Senior adults  36.62  24.82   % White  62.46  42.47   % Female  61.98  12.85   Urban location  .50     Suburban location  .18     Rural location  .32     Mid-Atlantic/Northeast  .12     Midwest  .23     South  .51     Mountain/Pacific  .14    Source: 2012 National Congregations Study. Imputed data (m = 20); weighted. View Large Table 2 presents the bivariate relationships of key independent variables with STM travel. Most of the factors we hypothesized are significantly related to STM travel at the bivariate level. Congregations with immigrants, especially relatively low levels of immigrants, are more likely to produce STM travel. In fact, these congregations represent 30% of congregations that produce STM travel, nearly twice the STM travel rate of congregations that do not have immigrants. Table 2. Religious Tradition, Resource, Demographic, and Immigrant-Related Variables by Foreign STM Travel in 2012   STM trip?  pa  No  Yes  Religious tradition   Roman Catholic  0.81  0.19  ***   White Evangelical Protestant  0.67  0.33   Black Protestant  0.79  0.21   White Mainline Protestant  0.79  0.21   Non-Christian  0.69  0.31   Hosted foreign clergy  0.22  0.5  ***  Resources   % Adults in high-income households  5.56  10.58  ***   Income (× $10,000)  32.49  109.99  ***   Youth minister/leader  0.49  0.66  ***   No. of full-time staff  1.59  6.82  ***  Demographics   % Adults with BA degree  30.47  36.82  ***   % Young adults  22.24  31.53  ***   No. of teenagers  18.94  31.08  ***  Immigrant-related   Recent immigrants    % Recent immigrants: 0  0.77  0.23  ***    % Recent immigrants: 0 to 5  0.58  0.42    % Recent immigrants: > 5  0.55  0.45   Services to immigrants    No services for immigrants  0.53  0.47  ***    Internal services for immigrants  0.76  0.24    External services for immigrants  0.75  0.25   Foreign-born head clergy  0.09  0.1      STM trip?  pa  No  Yes  Religious tradition   Roman Catholic  0.81  0.19  ***   White Evangelical Protestant  0.67  0.33   Black Protestant  0.79  0.21   White Mainline Protestant  0.79  0.21   Non-Christian  0.69  0.31   Hosted foreign clergy  0.22  0.5  ***  Resources   % Adults in high-income households  5.56  10.58  ***   Income (× $10,000)  32.49  109.99  ***   Youth minister/leader  0.49  0.66  ***   No. of full-time staff  1.59  6.82  ***  Demographics   % Adults with BA degree  30.47  36.82  ***   % Young adults  22.24  31.53  ***   No. of teenagers  18.94  31.08  ***  Immigrant-related   Recent immigrants    % Recent immigrants: 0  0.77  0.23  ***    % Recent immigrants: 0 to 5  0.58  0.42    % Recent immigrants: > 5  0.55  0.45   Services to immigrants    No services for immigrants  0.53  0.47  ***    Internal services for immigrants  0.76  0.24    External services for immigrants  0.75  0.25   Foreign-born head clergy  0.09  0.1    Source: 2012 National Congregations Study. Imputed data (m = 20); weighted. aWald test. †p < .10; *p < .05; **p < .01; ***p < .001, two-tailed tests. View Large Table 2. Religious Tradition, Resource, Demographic, and Immigrant-Related Variables by Foreign STM Travel in 2012   STM trip?  pa  No  Yes  Religious tradition   Roman Catholic  0.81  0.19  ***   White Evangelical Protestant  0.67  0.33   Black Protestant  0.79  0.21   White Mainline Protestant  0.79  0.21   Non-Christian  0.69  0.31   Hosted foreign clergy  0.22  0.5  ***  Resources   % Adults in high-income households  5.56  10.58  ***   Income (× $10,000)  32.49  109.99  ***   Youth minister/leader  0.49  0.66  ***   No. of full-time staff  1.59  6.82  ***  Demographics   % Adults with BA degree  30.47  36.82  ***   % Young adults  22.24  31.53  ***   No. of teenagers  18.94  31.08  ***  Immigrant-related   Recent immigrants    % Recent immigrants: 0  0.77  0.23  ***    % Recent immigrants: 0 to 5  0.58  0.42    % Recent immigrants: > 5  0.55  0.45   Services to immigrants    No services for immigrants  0.53  0.47  ***    Internal services for immigrants  0.76  0.24    External services for immigrants  0.75  0.25   Foreign-born head clergy  0.09  0.1      STM trip?  pa  No  Yes  Religious tradition   Roman Catholic  0.81  0.19  ***   White Evangelical Protestant  0.67  0.33   Black Protestant  0.79  0.21   White Mainline Protestant  0.79  0.21   Non-Christian  0.69  0.31   Hosted foreign clergy  0.22  0.5  ***  Resources   % Adults in high-income households  5.56  10.58  ***   Income (× $10,000)  32.49  109.99  ***   Youth minister/leader  0.49  0.66  ***   No. of full-time staff  1.59  6.82  ***  Demographics   % Adults with BA degree  30.47  36.82  ***   % Young adults  22.24  31.53  ***   No. of teenagers  18.94  31.08  ***  Immigrant-related   Recent immigrants    % Recent immigrants: 0  0.77  0.23  ***    % Recent immigrants: 0 to 5  0.58  0.42    % Recent immigrants: > 5  0.55  0.45   Services to immigrants    No services for immigrants  0.53  0.47  ***    Internal services for immigrants  0.76  0.24    External services for immigrants  0.75  0.25   Foreign-born head clergy  0.09  0.1    Source: 2012 National Congregations Study. Imputed data (m = 20); weighted. aWald test. †p < .10; *p < .05; **p < .01; ***p < .001, two-tailed tests. View Large In Table 3, we introduce sets of theoretical variables based on our hypotheses. Each set of theoretical variables in Models 1 through 5 passed an F-test that they were not jointly equal to zero. Since coefficients across logistic regression models with different specifications are not comparable, we focus our interpretation on Model 6, which includes all variables. Model 6 suggests some evidence for our hypotheses regarding religious tradition, hosting foreign clergy, youth ministers, education level of congregants, recent immigrants, and immigrant services. Table 3 Logistic Regression of Foreign STM Travel on Religious Tradition, Resources, Demographics, Immigrant-Related Variables, and Controls   Model 1  Model 2  Model 3  Model 4  Model 5  Model 6  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  Religious tradition   Roman Catholic (base = WEP)  −1.58  ***                  −1.92  ***    (0.16)                    (0.23)     Black Protestant (base = WEP)  −.92  ***                  −.40      (0.21)                    (0.32)     White liberal/moderate (base = WEP)  −.54  **                  −.37      (0.17)                    (0.22)     Non-Christian (base = WEP)  −1.36  ***                  −1.75  ***    (0.36)                    (0.45)     Hosted foreign clergy  1.21  ***                  .84  ***  (0.13)                    (0.15)    Resources   Income (×$10,000)      .00  *              .00          (0.00)                (0.00)     % Adults in high-income      .01  *              .01          (0.00)                (0.01)     No. of full-time staff      .02  **              .01          (0.01)                (0.01)     Youth minister/leader      .74  ***              .43  *        (0.15)                (0.17)    Demographics   % Adults with BA degree          .01  ***          .01  †          (0.00)            (0.00)     % Young adults          .02  ***          .00            (0.00)            (0.00)     No. of teenagers          .00  **          .00            (0.00)            (0.00)     % Recent immigrants: 0 to 5 (base = 0%)              .75  ***      .48  **              (0.13)        (0.18)     % Recent immigrants: >5 (base = 0%)              .51  **      .57  †              −.18        (0.28)     Foreign-born head clergy              −.27        .19                (0.19)        (0.23)     Internal service immigrants (base = none)                  .64  ***  .51  **                  (0.13)    (0.19)     External service immigrants (base = none)                  .42    −.17                    (0.30)    (0.38)    Controls   No. of adults                      .00                        (0.00)     Has denomination                      −.01                        (0.20)     Mod theology (base = conservative)                      −.21                        (0.17)     Lib theology (base = conservative)                      −.08                        (0.24)     Has a social service program                      .76  *                      (0.30)     % Age 60 and older                      −.01  **                      (0.00)     % White                      .00                        (0.00)     % Female                      −.01                        (0.01)     Urban location                      −.08                        (0.22)     Suburban location                      .21                        (0.25)     Midwest                      −.44  †                      (0.23)     South                      −.45  *                      (0.22)     Mountain/Pacific                      −.35                        (0.25)    Constant  −.39  ***  −1.35  ***  −1.51  ***  −.72  ***  −.60  ***  −1.12  †  (0.11)    (0.14)    (0.16)    (0.08)    (0.07)    (0.67)    N  1,322    1,284    1,306    1,314    1,323    1,280    Pseudo-R2  .1 0    .07    .04    .02    .01    .20      Model 1  Model 2  Model 3  Model 4  Model 5  Model 6  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  Religious tradition   Roman Catholic (base = WEP)  −1.58  ***                  −1.92  ***    (0.16)                    (0.23)     Black Protestant (base = WEP)  −.92  ***                  −.40      (0.21)                    (0.32)     White liberal/moderate (base = WEP)  −.54  **                  −.37      (0.17)                    (0.22)     Non-Christian (base = WEP)  −1.36  ***                  −1.75  ***    (0.36)                    (0.45)     Hosted foreign clergy  1.21  ***                  .84  ***  (0.13)                    (0.15)    Resources   Income (×$10,000)      .00  *              .00          (0.00)                (0.00)     % Adults in high-income      .01  *              .01          (0.00)                (0.01)     No. of full-time staff      .02  **              .01          (0.01)                (0.01)     Youth minister/leader      .74  ***              .43  *        (0.15)                (0.17)    Demographics   % Adults with BA degree          .01  ***          .01  †          (0.00)            (0.00)     % Young adults          .02  ***          .00            (0.00)            (0.00)     No. of teenagers          .00  **          .00            (0.00)            (0.00)     % Recent immigrants: 0 to 5 (base = 0%)              .75  ***      .48  **              (0.13)        (0.18)     % Recent immigrants: >5 (base = 0%)              .51  **      .57  †              −.18        (0.28)     Foreign-born head clergy              −.27        .19                (0.19)        (0.23)     Internal service immigrants (base = none)                  .64  ***  .51  **                  (0.13)    (0.19)     External service immigrants (base = none)                  .42    −.17                    (0.30)    (0.38)    Controls   No. of adults                      .00                        (0.00)     Has denomination                      −.01                        (0.20)     Mod theology (base = conservative)                      −.21                        (0.17)     Lib theology (base = conservative)                      −.08                        (0.24)     Has a social service program                      .76  *                      (0.30)     % Age 60 and older                      −.01  **                      (0.00)     % White                      .00                        (0.00)     % Female                      −.01                        (0.01)     Urban location                      −.08                        (0.22)     Suburban location                      .21                        (0.25)     Midwest                      −.44  †                      (0.23)     South                      −.45  *                      (0.22)     Mountain/Pacific                      −.35                        (0.25)    Constant  −.39  ***  −1.35  ***  −1.51  ***  −.72  ***  −.60  ***  −1.12  †  (0.11)    (0.14)    (0.16)    (0.08)    (0.07)    (0.67)    N  1,322    1,284    1,306    1,314    1,323    1,280    Pseudo-R2  .1 0    .07    .04    .02    .01    .20    Source: 2012 National Congregations Study. Imputed data (m = 20); unweighted. Coefficient estimates are unstandardized logits. †p < .10; *p < .05; ** p < .01; *** p < .001. View Large Table 3 Logistic Regression of Foreign STM Travel on Religious Tradition, Resources, Demographics, Immigrant-Related Variables, and Controls   Model 1  Model 2  Model 3  Model 4  Model 5  Model 6  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  Religious tradition   Roman Catholic (base = WEP)  −1.58  ***                  −1.92  ***    (0.16)                    (0.23)     Black Protestant (base = WEP)  −.92  ***                  −.40      (0.21)                    (0.32)     White liberal/moderate (base = WEP)  −.54  **                  −.37      (0.17)                    (0.22)     Non-Christian (base = WEP)  −1.36  ***                  −1.75  ***    (0.36)                    (0.45)     Hosted foreign clergy  1.21  ***                  .84  ***  (0.13)                    (0.15)    Resources   Income (×$10,000)      .00  *              .00          (0.00)                (0.00)     % Adults in high-income      .01  *              .01          (0.00)                (0.01)     No. of full-time staff      .02  **              .01          (0.01)                (0.01)     Youth minister/leader      .74  ***              .43  *        (0.15)                (0.17)    Demographics   % Adults with BA degree          .01  ***          .01  †          (0.00)            (0.00)     % Young adults          .02  ***          .00            (0.00)            (0.00)     No. of teenagers          .00  **          .00            (0.00)            (0.00)     % Recent immigrants: 0 to 5 (base = 0%)              .75  ***      .48  **              (0.13)        (0.18)     % Recent immigrants: >5 (base = 0%)              .51  **      .57  †              −.18        (0.28)     Foreign-born head clergy              −.27        .19                (0.19)        (0.23)     Internal service immigrants (base = none)                  .64  ***  .51  **                  (0.13)    (0.19)     External service immigrants (base = none)                  .42    −.17                    (0.30)    (0.38)    Controls   No. of adults                      .00                        (0.00)     Has denomination                      −.01                        (0.20)     Mod theology (base = conservative)                      −.21                        (0.17)     Lib theology (base = conservative)                      −.08                        (0.24)     Has a social service program                      .76  *                      (0.30)     % Age 60 and older                      −.01  **                      (0.00)     % White                      .00                        (0.00)     % Female                      −.01                        (0.01)     Urban location                      −.08                        (0.22)     Suburban location                      .21                        (0.25)     Midwest                      −.44  †                      (0.23)     South                      −.45  *                      (0.22)     Mountain/Pacific                      −.35                        (0.25)    Constant  −.39  ***  −1.35  ***  −1.51  ***  −.72  ***  −.60  ***  −1.12  †  (0.11)    (0.14)    (0.16)    (0.08)    (0.07)    (0.67)    N  1,322    1,284    1,306    1,314    1,323    1,280    Pseudo-R2  .1 0    .07    .04    .02    .01    .20      Model 1  Model 2  Model 3  Model 4  Model 5  Model 6  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  Religious tradition   Roman Catholic (base = WEP)  −1.58  ***                  −1.92  ***    (0.16)                    (0.23)     Black Protestant (base = WEP)  −.92  ***                  −.40      (0.21)                    (0.32)     White liberal/moderate (base = WEP)  −.54  **                  −.37      (0.17)                    (0.22)     Non-Christian (base = WEP)  −1.36  ***                  −1.75  ***    (0.36)                    (0.45)     Hosted foreign clergy  1.21  ***                  .84  ***  (0.13)                    (0.15)    Resources   Income (×$10,000)      .00  *              .00          (0.00)                (0.00)     % Adults in high-income      .01  *              .01          (0.00)                (0.01)     No. of full-time staff      .02  **              .01          (0.01)                (0.01)     Youth minister/leader      .74  ***              .43  *        (0.15)                (0.17)    Demographics   % Adults with BA degree          .01  ***          .01  †          (0.00)            (0.00)     % Young adults          .02  ***          .00            (0.00)            (0.00)     No. of teenagers          .00  **          .00            (0.00)            (0.00)     % Recent immigrants: 0 to 5 (base = 0%)              .75  ***      .48  **              (0.13)        (0.18)     % Recent immigrants: >5 (base = 0%)              .51  **      .57  †              −.18        (0.28)     Foreign-born head clergy              −.27        .19                (0.19)        (0.23)     Internal service immigrants (base = none)                  .64  ***  .51  **                  (0.13)    (0.19)     External service immigrants (base = none)                  .42    −.17                    (0.30)    (0.38)    Controls   No. of adults                      .00                        (0.00)     Has denomination                      −.01                        (0.20)     Mod theology (base = conservative)                      −.21                        (0.17)     Lib theology (base = conservative)                      −.08                        (0.24)     Has a social service program                      .76  *                      (0.30)     % Age 60 and older                      −.01  **                      (0.00)     % White                      .00                        (0.00)     % Female                      −.01                        (0.01)     Urban location                      −.08                        (0.22)     Suburban location                      .21                        (0.25)     Midwest                      −.44  †                      (0.23)     South                      −.45  *                      (0.22)     Mountain/Pacific                      −.35                        (0.25)    Constant  −.39  ***  −1.35  ***  −1.51  ***  −.72  ***  −.60  ***  −1.12  †  (0.11)    (0.14)    (0.16)    (0.08)    (0.07)    (0.67)    N  1,322    1,284    1,306    1,314    1,323    1,280    Pseudo-R2  .1 0    .07    .04    .02    .01    .20    Source: 2012 National Congregations Study. Imputed data (m = 20); unweighted. Coefficient estimates are unstandardized logits. †p < .10; *p < .05; ** p < .01; *** p < .001. View Large When compared to White Evangelical Protestant congregations, those in the Roman Catholic and non-Christian traditions are significantly less likely to produce STM travel. For example, the odds of a Roman Catholic congregation producing a STM trip were 85% lower (exp[−1.92] = .15) than those of a White Evangelical Protestant congregation, controlling for all other factors. The odds of a non-Christian congregation producing a STM trip were 83% lower (exp[−1.75] = .17) than those of a White Evangelical Protestant congregation. With relevant controls included, there is no significant difference in STM travel production among Protestant congregations, with Protestant congregations all more likely than non-Protestant congregations to produce STM travel.9 By examining pairwise contrasts, we confirmed this relative ordering, as well as no significant difference between Roman Catholic and non-Christian congregations. Overall, this ordering differs from our hypotheses by the absence of intra-Protestant variation and by the absence of difference between Roman Catholic and non-Christian congregations. Congregations that hosted a foreign clergyperson in the previous 12 months, no matter the religious tradition, were 2.3 times more likely to produce STM travel. Table 3 shows some limited evidence in support of our hypotheses regarding resource variables. The odds of a congregation producing a STM trip were 50% higher (exp[.43]) = 1.5) for congregations with a youth minister compared to those without (p < .05). We find some evidence for the effect of educational level (p < .10). Each one percent increase in adults with a college degree increased the odds of a congregational STM trip by about 0.7% (exp[.007] = .007). The resource variables of income, high-income households, and full-time staff fell just outside the cutoff of p <.10 for two-tailed tests, meaning they would be significant at p <.10 for one-tailed tests. Thus, the statistical evidence for them in this study is weak, but suggests their relevance. There is strong evidence for our hypotheses regarding the presence of immigrants. Congregations with a low level of recent, first-generation immigrants were more likely than congregations with no immigrants to produce an STM travel trip (p <.01).10 The odds of a congregation with a low level of recent immigrants producing STM travel were 60% higher (exp[.48] = 1.6) than those of a congregation with no recent immigrants. We note, too, that congregations with a higher level of recent immigrants (5%+) were also more likely than congregations with no recent immigrants to produce STM travel (p < .05). We had expected that congregations reporting external services for immigrants would be more likely to produce STM travel. However, the results show that they are not significantly different from congregations with no immigrant-serving programs at all. By contrast, congregations that reported only internal services for immigrants had 66% higher odds (exp[.512] = 1.67) than congregations with no immigrant-serving programs to produce STM travel (p < .01).11 Our results offer no support for age hypotheses or foreign-born clergy. Given the significance of both religious tradition and immigrant presence for trip production, we examined whether the effect of religious tradition on trip production was moderated by the presence of immigrants. Beyond the separate significance of these variables, we see three reasons why religious tradition might be moderated by immigrant presence.12 First, immigrant presence varies by religious tradition (Connor 2014). For example, about 28% of adult Catholics in the United States are immigrants, while immigrants make up less than 10% each of adult White Evangelical Protestants, White Mainline Protestants, and Black Protestants (Pew Forum 2014). Second, recent immigrants continue to influence all religious traditions (Connor 2012; Warner and Wittner 1998; Yang and Ebaugh 2001). Third, case studies suggest that the way that congregations connect immigrants to transnational communities, if at all, varies by religious tradition (Ebaugh and Chafetz 2002; Menjívar 2010; Mooney 2009). Table 4 examines this relationship, with the first column showing each religious tradition. The second column shows the distribution of congregations within a religious tradition across the three immigrant presence categories. Congregations in the Roman Catholic category and the non-Christian category demonstrate the highest presence of recent immigrants. The third column shows the rate of STM trip production within each immigrant category in a religious tradition. Table 4. Effect of the Interaction of Religious Tradition and Immigrant Presence on Foreign STM Travel Religious tradition  Congregations by category (%)a  Congregations within category with STM trip (%)b  Marginal effect of immigrants on STM trip (odds ratio)  p  Roman Catholic   Recent immigrants: 0  39  22  (base)     Recent immigrants: 0–5  39  16  1.30     Recent immigrants: >5  22  24  1.47    White Evangelical Protestant   Recent immigrants: 0  81  29  (base)     Recent immigrants: 0–5  13  50  3.79  ***   Recent immigrants: >5  5  57  4.79  ***  Black Protestant   Recent immigrants: 0  94  20  (base)     Recent immigrants: 0–5  4  28  4.50  **   Recent immigrants: >5  2  56  10.85  *  White Mainline Protestant   Recent immigrants: 0  85  15  (base)     Recent immigrants: 0–5  11  59  7.30  ***   Recent immigrants: >5  4  38  3.09  †  Non-Christian   Recent immigrants: 0  56  29  (base)     Recent immigrants: 0–5  34  29  1.55     Recent immigrants: >5  8  17  1.21    Religious tradition  Congregations by category (%)a  Congregations within category with STM trip (%)b  Marginal effect of immigrants on STM trip (odds ratio)  p  Roman Catholic   Recent immigrants: 0  39  22  (base)     Recent immigrants: 0–5  39  16  1.30     Recent immigrants: >5  22  24  1.47    White Evangelical Protestant   Recent immigrants: 0  81  29  (base)     Recent immigrants: 0–5  13  50  3.79  ***   Recent immigrants: >5  5  57  4.79  ***  Black Protestant   Recent immigrants: 0  94  20  (base)     Recent immigrants: 0–5  4  28  4.50  **   Recent immigrants: >5  2  56  10.85  *  White Mainline Protestant   Recent immigrants: 0  85  15  (base)     Recent immigrants: 0–5  11  59  7.30  ***   Recent immigrants: >5  4  38  3.09  †  Non-Christian   Recent immigrants: 0  56  29  (base)     Recent immigrants: 0–5  34  29  1.55     Recent immigrants: >5  8  17  1.21    Source: 2012 National Congregations Study. Imputed data (m = 20). aWithin each religious tradition, column may not sum to 100% due to rounding. bCell values show rate within immigrant category of a religious tradition. †p < .10; * p < .05, **p < .01; ***p < .001, two-tailed tests. View Large Table 4. Effect of the Interaction of Religious Tradition and Immigrant Presence on Foreign STM Travel Religious tradition  Congregations by category (%)a  Congregations within category with STM trip (%)b  Marginal effect of immigrants on STM trip (odds ratio)  p  Roman Catholic   Recent immigrants: 0  39  22  (base)     Recent immigrants: 0–5  39  16  1.30     Recent immigrants: >5  22  24  1.47    White Evangelical Protestant   Recent immigrants: 0  81  29  (base)     Recent immigrants: 0–5  13  50  3.79  ***   Recent immigrants: >5  5  57  4.79  ***  Black Protestant   Recent immigrants: 0  94  20  (base)     Recent immigrants: 0–5  4  28  4.50  **   Recent immigrants: >5  2  56  10.85  *  White Mainline Protestant   Recent immigrants: 0  85  15  (base)     Recent immigrants: 0–5  11  59  7.30  ***   Recent immigrants: >5  4  38  3.09  †  Non-Christian   Recent immigrants: 0  56  29  (base)     Recent immigrants: 0–5  34  29  1.55     Recent immigrants: >5  8  17  1.21    Religious tradition  Congregations by category (%)a  Congregations within category with STM trip (%)b  Marginal effect of immigrants on STM trip (odds ratio)  p  Roman Catholic   Recent immigrants: 0  39  22  (base)     Recent immigrants: 0–5  39  16  1.30     Recent immigrants: >5  22  24  1.47    White Evangelical Protestant   Recent immigrants: 0  81  29  (base)     Recent immigrants: 0–5  13  50  3.79  ***   Recent immigrants: >5  5  57  4.79  ***  Black Protestant   Recent immigrants: 0  94  20  (base)     Recent immigrants: 0–5  4  28  4.50  **   Recent immigrants: >5  2  56  10.85  *  White Mainline Protestant   Recent immigrants: 0  85  15  (base)     Recent immigrants: 0–5  11  59  7.30  ***   Recent immigrants: >5  4  38  3.09  †  Non-Christian   Recent immigrants: 0  56  29  (base)     Recent immigrants: 0–5  34  29  1.55     Recent immigrants: >5  8  17  1.21    Source: 2012 National Congregations Study. Imputed data (m = 20). aWithin each religious tradition, column may not sum to 100% due to rounding. bCell values show rate within immigrant category of a religious tradition. †p < .10; * p < .05, **p < .01; ***p < .001, two-tailed tests. View Large The rates of STM trip production vary markedly by immigrant presence category within each religious tradition. The evidence in column three suggests that STM travel production in each of the three Protestant traditions is substantially higher for congregations where immigrants are present. For example, 29% of White Evangelical Protestant congregations without recent immigrants report STM trip production, but 50% of White Evangelical Protestant congregations with immigrant presence of up to 5% report it. The effect of immigrant presence appears even greater among White Mainline Protestant congregations. Fifteen percent of White Mainline Protestant congregations without immigrants report STM trip production, but those congregations with immigrant presence of up to 5% report STM trip production at a rate nearly four times higher (59%). Black Protestant congregations, too, have higher STM travel rates with higher immigrant presence, especially when immigrant levels are above 5%. To understand whether these patterns hold given controls, we examined the interaction effect of religious tradition and immigrant presence. Due to the difficulty of examining interaction effects in logistic regression models (Ai and Norton 2003), we present both multiplicative and additive results of interaction (Buis 2010). The multiplicative approach examines the ratio of change in STM trip production relative to the baseline odds. When we included the interaction of religious tradition by immigrant category in the full regression model, no interacted pair (e.g., Catholic × 5%+ immigrants) was significant. Nor was the test of the joint significance of all pairs significant (p = .27). This multiplicative approach, however, does not show how the differences between religious traditions in their baseline odds of STM trip production combine with changes in level of immigrant presence. The additive approach, using marginal effects, shows the difference in the expected odds of STM trip production within a religious tradition as immigrant presence changes (Buis 2010). This approach provides a more accurate look at “real world” values of the interacted variables. The last column in Table 4 displays the marginal effect of the unique values of religious tradition and immigrant category on the odds of STM trip production (Williams 2012). After controlling for all other factors, congregations have increased odds of STM trip production when compared to congregations without immigrants, no matter the religious tradition. To illustrate this overall pattern, Figure 1 displays the log odds of STM trip production for congregations, by religious tradition, at the three levels of immigrant presence. FIGURE 1. View largeDownload slide Interaction of Religious Tradition and Immigrant Presence. FIGURE 1. View largeDownload slide Interaction of Religious Tradition and Immigrant Presence. Despite this overall trend, as the final column in Table 4 also indicates, the effect of difference in immigrant presence is only significant for congregations in the three Protestant religious traditions, significantly increasing their odds of STM production. Catholic congregations and non-Christian congregations with immigrants clearly produce STM travel, but the presence of immigrants does not appear to significantly affect the likelihood of this activity relative to congregations without immigrants. DISCUSSION Relatively little research has focused on the factors that produce STM travel despite voluminous research on the civic and political activities of congregations inside the United States. Our results suggest that STM travel has become part of the congregational civic repertoire in the United States, practiced by over a quarter of congregations annually. We find some evidence that three factors used to explain other types of congregational civic activity—resources, congregant demographics, and religious tradition—may influence STM travel production. Our research also demonstrates new insight into STM travel by linking immigrant-related dynamics to congregation-based transnationalism. As we expected, organizational resources matter. Youth ministry role specialization is associated with STM trip production. STM travel has long been construed as a way to transform youth religious identity, which is further confirmed by a congregation’s formal support of a position to do just that (Trinitapoli and Vaisey 2009). While our remaining resource variables fell just outside the conventional level of statistical significance, we suspect that refined measures, especially of congregational finances or of fundraising, would confirm the salience of resource factors. Demographic factors matter for STM trip production as well. Our results for educational level suggest that STM travel is partially due to the biographical availability and global experiential focus associated with higher education backgrounds. One unexpected demographic finding was the non-association of teenagers with congregational STM trip production. We suspect that this null result is a measurement weakness rather than a valid statement that teenagers are not associated with STM travel. There are two reasons for this conclusion. First, the NCS attendee measures are the strongest regarding adults, so the teenager measure may be weak. Second, the significance of the youth minister role suggests that congregational targeting of teenagers for STM travel likely exists. One conclusion to draw from these results is that some factors, like resources and congregant educational level, which are associated with externally focused congregational activities in general are associated with STM travel as well (Chaves and Tsitsos 2001a; Todd and Houston 2013). The cultural capital and cosmopolitan orientation of highly educated congregants, as well as organizational resources, appear to orient a congregation toward overseas civic engagement. Congregations already engaged with their local communities, represented by the significant coefficient for social services programs in Table 3, are also more likely to be transnationally engaged. We note, however, two differences of STM travel production when compared to external congregational activities in other research. First, local theological differences are not salient for differentiating STM travel production. Whereas theological leanings are associated with many types of congregational external engagement, this is not the case for STM travel. STM travel, as a mode of civic engagement oriented toward “the globe,” draws congregations from all theological stripes. And, yet, these theological differences matter for what congregations do during travel and why they do it (Adler and Offutt 2018). A second difference for STM travel is the role of wealthy congregants. Whereas congregations with high proportions of wealthy individuals appear to have less social service activity (Chaves and Tsitsos 2001b), in the case of STM, our results suggest that the presence of high-income individuals may foster activity (though this was on the margins of statistical significance). This is a topic in need of further research to understand how wealth inequality among individuals and congregations influences transnational engagement, possibly recreating class-based privileges. A core finding is the patterning of STM trip production by religious tradition. Protestant congregations are more likely to produce STM travel, controlling for all other factors. This difference cannot simply be attributed to the Protestant connotation of “STM,” as the dependent variable that we used avoided mission-centric language. The civic-centric language of our dependent variable shows the continued salience of religious tradition in differentiating international civic activity. The formalized foreign aid services of Roman Catholicism, and its weak engagement levels at the congregation level, likely contribute to that tradition’s lower STM rates. So might the large presence of immigrants, which we address below. Given the heterogeneity of the non-Christian category, the explanation for its relatively weak impact on STM travel is less clear. For example, 52% of Jewish congregations (about one-quarter of congregations within the NCS’ non-Christian category) produced STM travel, about twice the rate of other non-Christian congregations. The dynamics of non-Christian congregations require better research to understand religious transnational engagement. Even though bivariate rates of STM travel varied among Protestant religious traditions, our multivariate analysis suggested that controlling for other factors removed these intra-Protestant differences. This is most surprising in the case of Black Protestant congregations, which have been characterized as more focused on local, domestic engagement (Ammerman 2005; Wuthnow 2009). We see three plausible explanations for this finding. First, previously reported Protestant differences on international activity may be due to methodological decisions that focused on highly resourced congregations (Lynn 2016) or that used mission-centric language (Wuthnow 2009). Second, controlling for resources may be especially important for understanding a resource-intensive activity like STM within this tradition. For example, past research has suggested that lower levels of resource-intensive political activity by Black Protestant congregations is due to resource inequality; when resources are accounted for, Black Protestant congregations are at least equal to other congregations in the activity (Brown 2006). Third, Black Protestant congregations may simply engage in STM travel more than scholars have noticed. Large missionary structures are less extensive in Black Protestantism (Ammerman 2005), yet large numbers of congregations affiliated with Black Protestant denominations exist outside the United States with which U.S. congregations could partner (Lincoln and Mamiya 1990). This portion of the global Christian religious field seems to be a blind spot in current sociological research on transnational civic engagement (though see Bakker 2014). One surprising result arises from the interaction of religious tradition and immigrant presence. Those religious traditions that have relatively fewer congregations with immigrants (e.g. Protestant) are also those traditions in which congregations appear to be significantly influenced in their STM trip production by the presence of immigrants. We consider this pattern as part of our theorization of the connection between STM travel and immigration in transnational religion. The Immigrant Effect and Civic Remittances For over two decades, studies of religion have shown strong evidence of the transnational character of religion produced by post-1965 immigrants. Congregations have often been portrayed as organizations crucial to domestic integration, even while immigrants themselves sustained thick transnational lives. This portrayal left the transnational civic activities of congregation out of focus in two ways. First, the portrayal focused on idea-typical “immigrant congregations” despite the fact that immigrants are to be found in congregations more generally. Second, for actual “immigrant congregations,” the role of congregational support in how immigrants aid and shape their sending communities has likely been larger than reported (Ebaugh and Chafetz 2002; Menjívar 1999). Our research has highlighted the congregational role in transnational religion as a way to understand how immigrants use or prompt organizations for transnational civic engagement, like STM travel. Our results provide evidence for an immigrant effect: the influence that immigrants can have, as immigrants, on the activity of a U.S. congregation. We hypothesized that relatively low levels of recent immigrants would differentiate congregations regarding STM production. Nearly one-third (30%) of STM travel in 2012 came from congregations with recent immigrants. Twenty-two percent of STM travel in that year came from congregations with low levels (0 to 5%) of their adult members as recent immigrants. After controlling for other factors, including a congregation’s service orientation to immigrants, our results confirmed this hypothesis. Even though we did not expect it, our results also show evidence that higher levels of recent immigrants (>5%) influence trip production.13 This effect evokes patterns familiar to researchers of immigrant religion more generally. For example, in their review of immigrant religion in the United States, Yang and Ebaugh (2001:284) wrote, “immigrants in the United States … possess rich resources, including material wealth, advanced technologies, [and] organizational skills….” To this list of what immigrants possess, we would add membership in U.S. religious congregations. Congregations can provide an organized platform for immigrants to influence foreign communities, distinct from immigrants’ individual actions or hometown associations (Vertovec 2004). We can only speculate about how recent immigrants influence STM travel production. It is possible that recent immigrants could be drawn to congregations that lack immigrants but have reputations for international engagement, making our results spurious. We would be surprised if the occurrence was frequent enough to negate our findings. We think two other pathways are more likely. The first pathway includes those congregations that support a continued stream of immigration into the United States. In this pathway, STM travel becomes part of a formalized, organized transnational relationship, with newly arrived immigrants sustaining existing transnational connections, as well as providing motivation to continue to be transnationally engaged (Hansing 2014). A second pathway are those congregations that respond to immigrants’ new presence. This includes congregations with no or little historical connection to settling earlier immigration waves, but who now have recent immigrants as congregants. In this case, immigrants may expand the geographic and social imagination of a congregation. For individual congregants, immigrants may frame and personalize international need (Schnable 2015b). From the organizational view, personal connection to a specific international locale may ground abstract congregational visions of helping the world (Wuthnow 2009), while also decreasing the start-up costs of initiating STM travel. This pattern was alluded to by Wuthnow and Offutt (2008:218), who observed that “short-term trips are generally facilitated by preexisting transnational ties.” The power of small numbers of people to activate congregational social service and political activities is well known, so STM travel should join that list (Chaves 2004). For either of these pathways, we suggest seeing the result of this immigrant effect as a civic remittance. Levitt (2001) expanded the idea of economic remittances to the noneconomic dimension of immigrant transnational life. Social remittances, she explained, are delineated and personal, grounded in transnational interaction and capable of constructing new identities (Levitt 1998). The concept of civic remittances extends Levitt’s work, providing a way to understand the sizeable, targeted, but also unequal transfer of resources which she has referred to as the “transnationalizing [of] community development” (Levitt 2001). STM travel is civic through its normative vision of partnership, its locally organized relationships, its reliance on personal interactions, and its symbolic mechanisms to manage material inequality (Adler and Offutt 2018; Probasco 2016). Like other remittances, STM travel is a delineable activity with clear material outcomes that is based on personal relationships and facilitates continued transnational interactions. In the context of organization-based aid, the concept of remittance reflects the agency that immigrants themselves can have in prompting organizational action (Bakker 2014), instead of just being a cause for aid done by others in their presence (Schnable 2015b). Our research does provide evidence that the immigrant effect and ensuing civic remittance activity varies by religious tradition. Roman Catholic and non-Christian congregations with immigrants do engage in STM travel and thus civic remittance activity; however, they do not do so at significantly higher rates than congregations in their tradition with no immigrants. This evidence echoes other studies which have documented distinct styles of remittance activity by religious groups in different religious traditions (Ebaugh and Chafetz 2002; Kurien 2014; Menjívar 1999). Some of the difference we detect may be due to differences in immigrants in these religious traditions. For example, undocumented immigrants to the United States have historically come from majority-Catholic countries (Matovina 2012; Mooney 2009). Catholic congregations in the United States may be the destinations for these immigrants, but immigrants’ presence may not prompt increased STM activity due to the travel restrictions related to legal documentation and related economic difficulties.14 Such congregations are simply limited in their ability to engage in this particular form of civic remittance. It is also possible that the vision of religious citizenship among non-Protestant congregations in a historically Protestant country affects congregations’ orientation toward congregation-based civic remittance activity. Indeed, some evidence from Catholic parishes with immigrants suggests a focus away from strong transnational connection toward local life in the United States and broad immigrant identity (Cherry 2014; Menjívar 1999; Mooney 2009). One way to interpret the large, significant difference in STM travel production by Protestant congregations with immigrants compared to those without immigrants is to consider the meaning of immigrants in those traditions. Protestant congregations with any recent immigrants stand out, as less than 20% of congregations in any of the Protestant traditions have recent immigrants present. The transnational identity and focus of these congregations may be especially salient, as they exist in a religious context in which immigrant presence produces noticeable demographic difference. They may have a different stream of immigrants to begin with (Connor 2012) and stronger connection to specific religious communities in foreign countries (Hagan 2008), both of which could encourage STM.15 Our analysis also accounted for other immigrant-related dimensions of congregations beyond immigrant presence. For example, we found that foreign-born clergy were not associated with STM travel production. The heterogeneity in the “foreign-born clergy” category likely produced this result, as some may be longtime U.S. residents without foreign connections and some may be from countries unlikely to attract STM travel (Lynn 2016; Priest et al. 2006). Finally, we hypothesized that congregations with externally focused services to immigrants would be more likely to travel. While our formal expectation was wrong, our attention to congregational patterns of serving immigrants as distinct from the presence of recent immigrants revealed that congregations with internal services alone were more likely to produce STM travel. Since our data source provided no way of measuring the proportion of a congregations’ members that are long-term U.S. residents, or even second-generation immigrants, it is possible that this measure of internal services represents continued immigrant identity in a congregation even if recent immigration is minimal. This internal orientation may be a sign of social isolation due to immigrant needs, but also a sign of bonding around an identity that still extends transnationally. Other research literature has documented a similar dynamic of settled immigrant groups focused abroad (Kurien 2014; Menjívar 1999). For example, a long-standing Cuban immigrant parish in Miami had numerous internally focused ministries built around immigrant identity and one externally focused program: a “long-distance sister relationship” to the Cuban diocese through visits with humanitarian aid and money (Hansing 2014). This finding may qualify other literature which argues that transnational dynamics atrophy as the second generation emerges in a congregation (Chafetz and Ebaugh 2002; Levitt 2009). It may not take the presence of recent, first-generation immigrants to continue certain kinds of transnational connection, like STM travel. STM travel’s well-legitimated presence and its organizational basis may even make it a more durable transnational connection that outlasts first-generation personal relationships to sending communities. There are a number of weaknesses to consider in our research. First, better resource variables regarding staff size or budget categories would help to specify how resources matter in STM travel, which they likely do. Second, there is a universe of transnationally focused religious organizations in the United States that engender STM travel demand (Schnable 2015a). Their affiliations and relationships shape what congregations might do. Third, we have no information about where STM trips went. While we follow the logic of studies that show that congregations with the immigrant effect send STM travel to those immigrants’ countries, we cannot verify this with our data. Fourth, we have some information about a congregation’s international focus, for example, a measure of visiting clergy, but we do not have a longitudinal view to show how that focus developed and whether STM travel is the endpoint of that focus. A final weakness in our research is that we have no information for the size, number, or duration of STM trips, thereby missing heterogeneity within the congregation population produced by extremely active congregations. CONCLUSION Experienced annually by millions of people living in the United States, STM travel is a congregational activity whose conditions of production have been poorly understood. Alongside organizational programs such as worship, religious education, and local civic engagement, STM travel has clearly entered the repertoire of religious activity produced by U.S. religious congregations. Our research has clarified the frequency of STM travel among congregations and demonstrated the organizational factors that lead to differential trip production. While large relief and development organizations visibly dominate the portrait of religious transnational civic engagement, our research shows how a decentralized form of civic aid is produced at the congregational level, the latest iteration in a lineage of voluntary social engagement by American religious groups (Ammerman 2005). STM travel has been recognized as an organizational mechanism to accomplish transnational influence. Here, we have shown that this influence should be seen to include transnational influences to begin with. By connecting evidence and concepts from literatures that have tended to remain separate, we have shown that the transnational flow of immigration is connected to the flow of STM travel. We have argued that this immigrant effect likely occurs across the general population of congregations, not just among “immigrant congregations.” Immigrant religious communities in United States “are forming the worldwide organizational center of their respective religions in the United States” (Chafetz and Ebaugh 2002:284) and scholars have argued that we should attend to the religious processes that lead to civic engagement by immigrants (Levitt 2008). By tracing their connection to STM travel, we have shown how immigrants can use or prompt their U.S. congregations to be transnationally engaged, thereby influencing both their sending communities and U.S. religious organizations themselves. ACKNOWLEDGMENTS Previous versions of this article were presented at the meetings of the Society for the Scientific Study of Religion. We thank Roger Finke, Stephen Offutt, and two anonymous reviewers for comments. Footnotes 1 There are some cases where the pattern is reversed, especially for immigrant groups affiliated with foreign-based religious groups or denominations (Kurien 2013). 2 The data were downloaded from the Association of Religion Data Archives, www.TheARDA.com. 3 If the GSS response rate is included, the response rate is 52%. 4 In the end, this recoding only marginally influenced coefficient size of Black Protestant religious tradition in regression models below, but not statistical significance. 5 We also tested a version that split the third category into “moderate” and “high,” based on one definition of “immigrant congregations” as those with more than 20% of congregants characterized as recent immigrants (Numrich and Kniss 2007). When used in the main regression analyses, congregations with high levels of recent immigrants (20%+) were not significantly different in producing STM travel compared to congregations with no immigrants, a result in line with theory. However, since less than 2% of congregations in the NCS sample fall into the “high” category, we could not use it for interaction effect analyses. We note that combining the “moderate” and “high” categories do not change the main results. 6 The number of imputed datasets is based on M ≥ 100 × FMI, where FMI is the fraction of missing information for a variable and M is the number of imputed datasets. See White et al. (2011). 7 We used the Stata ado file wgttest by Ben Jann to run this test. 8 This is 2% higher than the rate reported in the 2006 NCS, though the difference is not significant. Using the NCS’ attendee-based survey weight, we note that 41% of attendees in an average congregation in 2012 were in one that reported an STM trip in the previous year. 9 Both White Mainline Protestant (p < .10) and Black Protestant (p < .01) congregations were significantly different from White Evangelical Protestant congregations when all variables except immigration-related variables were included (model not shown). This suggests the importance of accounting for this dimension of STM travel production. 10 We conducted a sensitivity analysis of “token” levels of recent immigrants by setting the level of recent immigrants to zero for any congregation that reported less than 1% of regular adults as recent immigrants. There was no salient change in the results. 11 We also ran Model 6 with immigrant presence variables removed and, separately, with immigrant service variables removed. In both cases, the variables had smaller p-values and increased coefficient size. This suggests some overlap in these concepts, though variance inflation factors are below 2 for these variables. 12 We thank two reviewers for suggesting closer attention to immigration-related differences between religious traditions. 13 See footnote 5 for evidence regarding high levels of recent immigrants. 14 We thank a reviewer for this important interpretation. 15 The NCS has information about the country origins of recent immigrants. As an example of how religious tradition may shape STM, of Black Protestant congregations with recent immigrants, 79% reported that the recent immigrants came from the Caribbean or Africa. REFERENCES Adler, Gary J., and Stephen Offutt. 2018. “ The Gift Economy of Direct Transnational Civic Action: How Reciprocity and Inequality Are Managed in Religious ‘Partnerships’.” Journal for the Scientific Study of Religion. Forthcoming . Ai, Chunrong, and Edward C. Norton. 2003. “ Interaction Terms in Logit and Probit Models.” Economics Letters  80: 123– 29. Google Scholar CrossRef Search ADS   Ammerman, Nancy Tatom. 2005. Pillars of Faith: American Congregations and Their Partners . Berkeley: University of California Press. Bakker, Janel Kragt. 2014. Sister Churches: American Congregations and Their Partners Abroad . New York: Oxford University Press. Bane, Mary Jo. 2005. “ The Catholic Puzzle: Parishes and Civic Life.” In Taking Faith Seriously , edited by Mary Jo Bane, Brent Coffin, and Richard Higgins, 63– 93. Cambridge, MA: Harvard University Press. Barnes, Sandra L. 2005. “ Black Church Culture and Community Action.” Social Forces  84: 967– 94. Barnes, Sandra L. 2011. “ Black Church Sponsorship of Economic Programs: A Test of Survival and Liberation Strategies.” Review of Religious Research  53: 23– 40. Google Scholar CrossRef Search ADS   Beyerlein, Kraig, and Mark Chaves. 2003. “ The Political Activities of Religious Congregations in the United States.” Journal for the Scientific Study of Religion  42: 17. Google Scholar CrossRef Search ADS   Beyerlein, Kraig, Gary Adler, and Jennifer Trinitapoli. 2011. “ The Effect of Religious Mission Trips on Youth Civic Participation.” Journal for the Scientific Study of Religion  50: 780– 95. Google Scholar CrossRef Search ADS   Brown, R. Khari. 2006. “ Racial Differences in Congregation-Based Political Activism.” Social Forces  84: 1581– 604. Google Scholar CrossRef Search ADS   Buis, Maarten L. 2010. “ Stata Tip 87: Interpretation of Interactions in Non-Linear Models.” The Stata Journal  10: 305– 08. Cadge, Wendy. 2008. “ De Facto Congregationalism and the Religious Organizations of Post-1965 Immigrants to the United States: A Revised Approach.” Journal of the American Academy of Religion  76: 344– 74. Google Scholar CrossRef Search ADS   Cadge, Wendy, and Elaine Howard Ecklund. 2007. “ Immigration and Religion.” Annual Review of Sociology  33: 359– 79. Google Scholar CrossRef Search ADS   Calderisi, Robert. 2013. Earthly Mission: The Catholic Church and World Development . Hartford, CT: Yale University Press. Case, Jay R. 2006. “ And Ever the Twain Shall Meet: The Holiness Missionary Movement and the Birth of World Pentecostalism, 1870––1920.” Religion and American Culture: A Journal of Interpretation  16: 125– 60. Google Scholar CrossRef Search ADS   Chafetz, Janet Saltzman, and Helen Rose Ebaugh. 2002. “ The Variety of Transnational Religious Networks.” Religion Across Borders: Transnational Immigrant Networks : 165– 91. Chaves, Mark. 2004. Congregations in America . Cambridge, MA: Harvard University Press. Chaves, Mark, and Shawna L. Anderson. 2014. “ Changing American Congregations: Findings from the Third Wave of the National Congregations Study.” Journal for the Scientific Study of Religion  53: 676– 86. Google Scholar CrossRef Search ADS   Chaves, Mark, and William Tsitsos. 2001a. “ Congregations and Social Services: What They Do, How They Do It, and with Whom.” Nonprofit and Voluntary Sector Quarterly  30: 660– 83. Google Scholar CrossRef Search ADS   ———. 2001b. “ Congregations and Social Services: What They Do, How They Do It, and with Whom.” Nonprofit and Voluntary Sector Quarterly  30: 660– 83. CrossRef Search ADS   Cherry, Stephen M. 2014. Faith, Family, and Filipino American Community Life . New Brunswick, NJ: Rutgers University Press. Connor, Phillip. 2012. “ International Migration and Religious Selection.” Journal for the Scientific Study of Religion  51: 184– 94. Google Scholar CrossRef Search ADS   ———. 2014. Immigrant Faith: Patterns of Immigrant Religion in the United States, Canada, and Western Europe . New York: New York University Press. Cornelius, David. 2002. “ A Brief Historical Survey of African-American Involvement in International Missions.” In African-American Experience in World Mission: A Call Beyond Community , edited by V. J. Walston and R. J. Stevens, 48– 53. Pasadena, CA: William Carey Library. DuMouchel, William H., and Greg J. Duncan. 1983. “ Using Sample Survey Weights in Multiple Regression Analyses of Stratified Samples.” Journal of the American Statistical Association  78: 535– 43. Google Scholar CrossRef Search ADS   Ebaugh, Helen Rose. 2004. “ Religion across Borders: Transnational Religious Ties.” Asian Journal of Social Science  32: 216– 31. Google Scholar CrossRef Search ADS   Ebaugh, Helen Rose, and Janet Saltzman Chafetz. 2002. Religion Across Borders: Transnational Immigrant Networks . Lanham, MD: Altamira Press. Ebaugh, Helen Rose, and Paula Pipes. 2001. “ Immigration Congregations as Social Service Providers: Are They Safety Nets for Welfare Reform?” In Religion and Social Policy , edited by P. D. Nesbitt. AltaMira Press. Frenk, Steven M., Shawna L. Anderson, Mark Chaves, and Nancy Martin. 2011. “ Assessing the Validity of Key Informant Reports About Congregations’ Social Composition.” Sociology of Religion  72: 78– 90. Google Scholar CrossRef Search ADS   Hagan, Jacqueline Maria. 2008. Migration Miracle: Faith, Hope, and Meaning on the Undocumented Journey . Cambridge, MA: Harvard University Press. Hansing, Katrin. 2014. “ Unidos en la Fe: Transnational Civic Social Engagement between Two Cuban Catholic Parishes.” In Churches and Charity in the Immigrant City: Religion, Immigration, and Civic Engagement in Miami , edited by A. Stepick, S. J. Mahler, and T. Rey. New Brunswick, NJ: Rutgers University Press. Hefferan, Tara. 2007. Twinning Faith and Development: Catholic Parish Partnering in the U.S. and Haiti . Bloomfield, CT: Kumarian Press. Hirschman, Charles. 2004. “ The Role of Religion in the Origins and Adaptation of Immigrant Groups in the United States.” International Migration Review  38: 1206– 33. Google Scholar CrossRef Search ADS   Howell, Brian. 2010. “Roots of the Short-Term Missionary, 1960–1985”: Building Church Leaders. http://www.buildingchurchleaders.com/articles/2006/rootsmissionary.html. ———. 2012. Short-Term Mission: An Ethnography of Christian Travel Narrative and Experience . Downers Grove, IL: IVP Academic. Jacobs, Sylvia M. 1980. “ Black American Missionaries in Africa: A Selected Bibliography.” A Current Bibliography on African Affairs  13: 166– 72. Google Scholar CrossRef Search ADS   Kelner, Shaul. 2010. Tours That Bind: Diaspora, Pilgrimage, and Israeli Birthright Tourism . New York: New York University Press. Google Scholar CrossRef Search ADS   Killingray, David. 2003. “ The Black Atlantic Missionary Movement and Africa, 1780s–1920s.” Journal of Religion in Africa  33: 3– 31. Google Scholar CrossRef Search ADS   Kinney, Nancy T. 2015. “ Structure, Context, and Ideological Dissonance in Transnational Religious Networks.” Voluntas  26: 382– 407. Google Scholar CrossRef Search ADS   Kivisto, Peter. 2016. “ International Migration.” In Handbook of Religion and Society , edited by D. Yamane, 547– 68. New York, NY: Springer. Google Scholar CrossRef Search ADS   Kurien, Prema. 2013. “ Religion, Social Incorporation, and Civic Engagement: Second-Generation Indian American Christians.” Review of Religious Research  55: 81– 104. Google Scholar CrossRef Search ADS   ———. 2014. “ The Impact of International Migration on Home Churches: The Mar Thoma Syrian Christian Church in India.” Journal for the Scientific Study of Religion  53: 109– 29. CrossRef Search ADS   Levitt, Peggy. 1998. “ Local-Level Global Religion: The Case of U.S.-Dominican Migration.” Journal for the Scientific Study of Religion  37: 74– 89. ———. 2001. Transnational Villagers . Berkeley, CA: UC Press. ———. 2007. God Needs No Passport: Immigrants and the Changing American Religious Landscape . New York, NY: The New Press. ———. 2008. “ Religion as a Path to Civic Engagement.” Ethnic and Racial Studies  31: 766– 91. CrossRef Search ADS   ———. 2009. “ Roots and Routes: Understanding the Lives of the Second Generation Transnationally.” Journal of Ethnic and Migration Studies  35: 1225– 42. CrossRef Search ADS   Ley, David. 2008. “ The Immigrant Church as an Urban Service Hub.” Urban Studies  45: 2057– 74. Google Scholar CrossRef Search ADS   Lincoln, C. Eric, and Lawrence H. Mamiya. 1990. The Black Church in the African American Experience . Durham, NC: Duke University Press. Google Scholar CrossRef Search ADS   Lynn, Monty L. 2016. “ Congregational Aid: North American Protestant Engagement in International Relief and Development.” The Journal of Development Studies  52: 965– 85. Google Scholar CrossRef Search ADS   Matovina, Timothy. 2012. Latino Catholicism: Transformation in American’s Largest Church . Princeton, NJ: Princeton University Press. McAlister, Melani. 2008. “ What Is Your Heart For?: Affect and Internationalism in the Evangelical Public Sphere.” American Literary History  20: 870– 95. Google Scholar CrossRef Search ADS   McCleary, Rachel M., and Robert J. Barro. 2008. “ Private Voluntary Organizations Engaged in International Assistance, 1939–2004.” Nonprofit and Voluntary Sector Quarterly  37: 512– 36. Google Scholar CrossRef Search ADS   Menjívar, Cecilia. 1999. “ Religious Institutions and Transnationalism: A Case Study of Catholic and Evangelical Salvadoran Immigrants.” International Journal of Politics, Culture, and Society  12: 589– 612. Google Scholar CrossRef Search ADS   ———. 2010. “ Immigrants, Immigration, and Sociology: Reflecting on the State of the Discipline.” Sociological Inquiry  80: 3– 27. CrossRef Search ADS   Mooney, Margarita. 2009. Faith Makes Us Live: Surviving and Thriving in the Haitian Diaspora . Berkeley, CA: University of California Press. Numrich, Paul, and Fred Kniss. 2007. Sacred Assemblies and Civic Engagement: How Religion Matters for America’s Newest Immigrants . New Brunswick, NJ: Rutgers University Press. Offutt, Stephen. 2011. “ The Role of Short-Term Mission Teams in the New Centers of Global Christianity.” Journal for the Scientific Study of Religion  50: 796– 811. Offutt, Stephen. 2015. New Centers of Global Evangelicalism in Latin America and Africa . New York: Cambridge University Press. Google Scholar CrossRef Search ADS   Offutt, Stephen, and Grant Miller. 2016. “Transnationalism.” In Handbook of Religion and Society , edited by D. Yamane, 525– 45. Switzerland: Springer International Publishing. Google Scholar CrossRef Search ADS   Pew Research Center. 2014. Religious Landscape Study . Pew Research Center: Washington, DC. Priest, Robert J., Terry Dischinger, Steve Rasmussen, and Christopher M. Brown. 2006. “ Researching the Short-Term Mission Movement.” Missiology: An International Review  34: 431– 51. Google Scholar CrossRef Search ADS   Probasco, LiErin. 2013. “ Giving Time, Not Money: Long-Term Impacts of Short-Term Mission Trips.” Missiology: An International Review  41: 202– 24. Google Scholar CrossRef Search ADS   ———. 2016. “ Prayer, Patronage, and Personal Agency in Nicaraguan Accounts of Receiving International Aid.” Journal for the Scientific Study of Religion  55: 233– 49. CrossRef Search ADS   Roll, Jarod. 2010. “ Garveyism and the Eschatology of African Redemption in the Rural South, 1920–1936.” Religion and American Culture: A Journal of Interpretation  20: 27– 56. Google Scholar CrossRef Search ADS   Schnable, Allison. 2015a. “ New American Relief and Development Organizations: Voluntarizing Global Aid.” Social Problems  62: 309– 29. Google Scholar CrossRef Search ADS   ———. 2015b. “ Religion and Giving for International Aid: Evidence from a Survey of U.S. Church Members.” Sociology of Religion  76: 72– 94. CrossRef Search ADS   Smith, Christian. 2005. Soul Searching: The Religious and Spiritual Lives of American Teenagers. New York, NY: Oxford University Press. Smith, Christian. 1996. Resisting Reagan: The U.S. Central America Peace Movement . Chicago, IL: University of Chicago. Google Scholar CrossRef Search ADS   Snee, Helene. 2013. “ Framing the Other: Cosmopolitanism and the Representation of Difference in Overseas Gap Year Narratives.” The British Journal of Sociology  64: 142– 62. Google Scholar CrossRef Search ADS PubMed  Steensland, Brian, Jerry Z. Park, Mark D. Regnerus, Lynn D. Robinson, W. Bradford Wilcox, and Robert D. Woodberry. 2000. “ The Measure of American Religion: Toward Improving the State of the Art.” Social Forces  79: 291– 318. Google Scholar CrossRef Search ADS   Sutherland, Jim. 2003. “ How Black Is the Harvester? A Profile of the African-American Missionary Force, the Extent of the Problem.” In African-American Experience in World Mission: A Call Beyond Community , edited by V. J. Walston and R. J. Stevens, 135– 39. Pasadena, CA: William Carey Library. Thaut, Laura C. 2009. “The Role of Faith in Christian Faith-Based Humanitarian Agencies: Constructing the Taxonomy.” Voluntas 20:319–50. Todd, Nathan R., and Jaclyn D. Houston. 2013. “ Examining Patterns of Political, Social Service, and Collaborative Involvement of Religious Congregations: A Latent Class and Transition Analysis.” American Journal of Community Psychology  51: 442– 38. Google Scholar CrossRef Search ADS   Trinitapoli, Jenny, and Stephen Vaisey. 2009. “ The Transformative Role of Religious Experience: The Case of Short Term Missions.” Social Forces  88: 121– 46. Google Scholar CrossRef Search ADS   Ver Beek, Kurt Alan. 2006. “ The Impact of Short-Term Missions: A Case Study of House Construction in Honduras after Hurricane Mitch.” Missiology: An International Review  34: 477– 96. Google Scholar CrossRef Search ADS   Vertovec, Steven. 2004. “ Migrant Transnationalism and Modes of Transformation.” The International Migration Review  38: 970– 1001. Google Scholar CrossRef Search ADS   Warner, R. Stephen. 2007. “ The Role of Religion in the Process of Segmented Assimilation.” The Annals of the American Academy of Political and Social Science  612: 100– 15. Google Scholar CrossRef Search ADS   Warner, R. Stephen, and Judith G. Wittner. 1998. Gatherings in Diaspora: Religious Communities and the New Immigration . Philadelphia, PA: Temple University Press. White, Ian R., Patrick Royston, and Angela M. Wood. 2011. “ Multiple Imputation Using Chained Equations: Issues and Guidance for Practice.” Statistics in Medicine  30: 377– 99. Google Scholar CrossRef Search ADS PubMed  Williams, Richard. 2012. “ Using the Margins Command to Estimate and Interpret Adjusted Predictions and Marginal Effects.” The Stata Journal  12: 308– 31. Woodberry, Robert D., Jerry Z. Park, Lyman A. Kellstedt, Mark D. Regnerus, and Brian Steensland. 2012. “ The Measure of American Religious Traditions: Theoretical and Measurement Considerations.” Social Forces  91: 65– 73. Google Scholar CrossRef Search ADS   Wuthnow, Robert. 2009. Boundless Faith: The Global Outreach of American Churches . Los Angeles: University of California Press. Google Scholar CrossRef Search ADS   Wuthnow, Robert, and Stephen Offutt. 2008. “ Transnational Religious Connections.” Sociology of Religion  69: 187– 209. Google Scholar CrossRef Search ADS   Yang, Fenggang, and Helen Rose Ebaugh. 2001. “ Transformations in New Immigrant Religions and Their Global Implications.” American Sociological Review  66: 269– 88. Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of the Association for the Sociology of Religion. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Sociology of Religion Oxford University Press

The Immigrant Effect: Short-Term Mission Travel as Transnational Civic Remittance

Loading next page...
 
/lp/ou_press/the-immigrant-effect-short-term-mission-travel-as-transnational-civic-J91xH45NMM
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Association for the Sociology of Religion. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
1069-4404
eISSN
1759-8818
D.O.I.
10.1093/socrel/srx060
Publisher site
See Article on Publisher Site

Abstract

Abstract Short-term mission (STM) travel is a popular religious and civic practice done by religious congregations, but the local conditions that facilitate its production are poorly understood. We analyze organizational factors behind STM travel, with special focus on the role of recent immigrants within congregations. We use data from the third wave of the National Congregations Study. Our results show large differences by religious tradition, as well as the influence of foreign clergy, youth ministry, college-educated members, recent immigrants, and immigrant service orientation. We identify an immigrant effect, theorizing how immigrant presence and identity influence U.S. congregations’ transnational engagement, especially within religious traditions with relatively low levels of recent immigrants. By connecting research on congregational civic engagement with that on transnational immigrant religion, we argue that about 30% of STM travel is a form of civic remittance in which recent immigrants and their U.S. congregations aid foreign communities. Short-term mission (STM) travel by U.S.-based congregations to international locations has received markedly increased scholarly attention in recent years (Adler and Offutt 2018; Bakker 2014; Beyerlein et al. 2011; Howell 2012; Priest et al. 2006; Wuthnow 2009). This large movement of people and resources facilitated by congregations has been presented as a flow of civic, religious transnationalism, primarily moving in the opposite direction of another well-known transnational flow, immigration (Adler and Offutt 2018; Offutt 2015; Offutt and Miller 2016; Priest et al. 2006; Wuthnow and Offutt 2008). Yet, no research has systematically examined the factors that differentiate why some U.S. congregations produce such distinctive transnational civic engagement and others do not. In this article, we argue that this general lacunae, in turn, has led scholars to miss the role that immigrants have in prompting transnational civic engagement by some U.S. congregations. Instead of two disconnected flows passing each other by, we argue that immigration and STM travel are two related aspects of religious transnationalism. In the past two decades, a body of transnational immigration research has shown the range of personal ties that immigrants retain to sending communities and has demonstrated the role of U.S. religious congregations in shaping immigrant assimilation (Cadge and Ecklund 2007; Ebaugh and Chafetz 2002; Hirschman 2004; Kurien 2013, 2014; Levitt 2001, 2007; Menjívar 2010; Warner 2007; Warner and Wittner 1998; Yang and Ebaugh 2001). Within this growing literature, some studies have pointed to the limited civic engagement propensities of immigrant congregations, noting that their patterns of informal, localized services tend to bond immigrants together (Cadge 2008; Ebaugh and Pipes 2001; Hirschman 2004; Levitt 2008; Ley 2008). At the same time, other studies have demonstrated the sizeable engagements that immigrants have with their sending communities, particularly through social and economic remittances (Kurien 2014; Levitt 1998; Menjívar 1999; Vertovec 2004). While these two sets of findings reveal much about immigrants and congregational activity, together they minimize the role of congregations in facilitating immigrants’ formalized transnational engagement, while also neglecting ways that immigrants may motivate transnational engagement among congregations more generally. We translate concepts from research on congregational civic engagement to analyze why congregations direct substantial organizational focus and resources to STM travel. Our results show significant, sizeable differences in STM trip production by a number of factors, including the presence of recent immigrants and organizational styles of serving immigrants. We theorize that these results point to an immigrant effect: the power of immigrants’ transnational connections, identities, and motivations to shape U.S. congregations’ transnational civic activity (Schnable 2015b). For nearly one-third (30%) of congregations, then, STM travel is an extension of a transnational process already set in motion by immigrants themselves. Given this, we suggest that STM travel and other transnational civic activities by U.S. congregations could be understood as a form of civic remittance when immigrants are involved. Similar to other remittances, civic remittances are a resource that immigrants can use to influence the material, cultural, and religious life of their home communities (Levitt 2001). This research contributes to an understanding of congregations as producers of civic action beyond their locales, adding to knowledge about the political, social, and cultural impact of religious organizations on public life. In particular, our research addresses a problem identified by Numrich and Kniss (2007:220), that “civic engagement patterns have garnered less attention from scholars of immigrant congregations than from scholars of nonimmigrant congregations.” Part of our contribution is to expand research attention about immigrants and congregational civic engagement, in both the “immigrant congregation” setting and beyond. We also expand knowledge about how religious transnationalism works by showing the connection between two transnational flows of people that have tended to be analyzed separately (though see Schnable 2015b). In a recent review of religious transnationalism, Offutt and Miller (2016:536) argued that transnational activities like STM travel are “a genuinely new wrinkle in how [U.S.] congregations operate and [they] should force sociologists to rethink what we know about the cultures and environments of local religious organizations.” Our research sheds light on this “new wrinkle,” showing how religious transnationalism might influence local congregations through immigrants, but also how local U.S. congregations in turn help construct religious transnationalism. BACKGROUND OF STM TRAVEL STM travel by U.S. residents through congregations was nearly nonexistent in 1950s (Howell 2012; Wuthnow 2009). The expansion of mass travel infrastructure, the emergence of diversified transnational religious organizations (Bakker 2014; McCleary and Barro 2008), and shifts in global religious demographics (Wuthnow 2009) increased the ease of STM travel, while changes in religious discourse about foreign others (Hefferan 2007; Howell 2012) and the increased valuation of identity-shaping religious practices among religious Americans (Bakker 2014; Trinitapoli and Vaisey 2009) led to a drastic increase in STM popularity. By the mid-2000s, STM travel engaged 1.6 million U.S. church-going adults each year (Wuthnow 2009). This is likely an undercount of the actual total of congregation-based traveling religious persons since it does not include adults that attend worship services less regularly or those under the age of 18 to whom trips are often targeted (Priest et al. 2006; Probasco 2013). Indeed, using retrospective data, Wuthnow (2009) showed that STM travel by teenagers has increased five-fold since 1980s. As further evidence of STM’s popularity and familiarity, the National Study of Youth and Religion (NSYR) indicates that in 2002, 29% of U.S. teenagers between the ages of 13 and 17 reported having gone on a religious mission trip of some type (Trinitapoli and Vaisey 2009). STM travel is differentiated from other modes of religious travel by a reliance on local U.S. religious organizations to mobilize traveling groups (Adler and Offutt 2018; Priest et al. 2006); by interactions with foreign communities through local religious organizations native to foreign communities (Bakker 2014; Wuthnow 2009); and by the usage of partnership language to articulate a form of moral relationship (Adler and Offutt 2018; Bakker 2014; Wuthnow 2009). Ammerman (2005) characterizes this direct, local-to-local pattern as “pervasive” among American congregations, with denominational, parachurch, and international aid organizations helping to broker connections between local organizations and foreign communities (Kinney 2015; Wuthnow 2009). One common feature of STM travel is the delivery of aid, whereby “the majority of trips involve some kind of service project in an impoverished community” (Probasco 2013:204), as well as finanical capital, educational materials, and/or medical supplies to facilitate these projects (Adler and Offutt 2018). Alongside the delivery of aid, STM trips may also involve “learning about cultural and social justice issues, and[/or] engaging in evangelization efforts among local populations” (Beyerlein et al. 2011:781). What this STM activity accomplishes for foreign organizations and communities—beyond charitable contributions and the ideal of religious partnership—is a topic of continued debate and depends on differences within the broad category (Adler and Offutt 2018; Hefferan 2007; Offutt 2011; Priest et al 2006; Ver Beek 2006; Wuthnow 2009). Our purpose is to focus on the organizational characteristics that may influence congregational STM travel to begin with. In the following section, we introduce four such factors. FACTORS INFLUENCING CONGREGATION-BASED STM TRAVEL Religious Tradition Religious tradition categorizes congregations according to shared historical formation and social ideas (Steensland et al. 2000; Woodberry et al. 2012). Religious tradition has been shown to influence a range of congregation-based social service activity and political engagement (Barnes 2011; Beyerlein and Chaves 2003). Religious tradition matters for STM production because traditions combine distinct orientations to “the world,” preferred modes of action for transforming “the world,” and different organizational structures that facilitate international engagement. Basic correlations provide evidence that religious tradition influences STM trip production. The NSYR suggests variation in traveler participation by religious tradition, with Latter-Day Saint, Mainline Protestant, and Conservative Protestant youth more likely to travel (Trinitapoli and Vaisey 2009). Beyerlein et al. (2011) report in a footnote using 2006 National Congregations Study (NCS) data that, without controls, Evangelical Protestant congregations reported the highest annual rate of STM travel. Similarly, using representative data from congregations in Arizona, Adler and Offutt (2018) report that Evangelical Protestant congregations were the most likely to travel within the previous year as well as over a 5-year period. Here, we briefly theorize how religious tradition may structure STM travel variation. After the mid-20th century, “many Mainline Protestant denominations were out of the overseas missions business entirely…” (Ammerman 2005:293), which meant less opportunity for foreign connections. The Mainline Protestant style of foreign engagement shifted away from emphasizing the export of religious belief to the provision of foreign relief. This relative disengagement of personal encounter by Mainline Protestants suggests that congregations in the Mainline Protestant tradition may not produce STM travel with the highest frequency (Bakker 2014). However, because Mainline Protestant denominations and organizations exist throughout the developing world (Lynn 2016; McCleary and Barro 2008), and because congregant interest in international affairs is still high (Ammerman 2005; Schnable 2015b), we expect that Mainline Protestant congregations will produce short-term travel at a moderate level compared to other traditions. Also in the mid-20th century, the growing Evangelical Protestant tradition massively invested in foreign missionary fields, part of a renewed proselytization strategy (Ammerman 2005; Wuthnow 2009). Numerous well-funded, U.S.-based organizations developed to facilitate foreign engagement by Evangelical Protestant congregations (Thaut 2009). Today, Evangelical Protestant congregations support international work at relatively high rates through both denominational and parachurch structures (Ammerman 2005), even though individuals are significantly less likely to donate to international aid (Schnable 2015b). As Howell (2010) recounts in a partial history, the STM idea was embraced by Evangelical Protestant religious groups as a method of recruiting future missionaries (though that specific purpose generally failed), reflecting an individualist aid logic (Schnable 2015b). With the demographic center of Christianity shifting to the Global South, U.S. Evangelical Protestant religious groups increasingly have religious international peers (co-religionists) to connect with (Bakker 2014). McAlister (2008) characterizes this contemporary global focus among U.S. evangelicals as an “enchanted internationalism” in which foreign evangelicals are seen as partners in mission activity, but also spiritual authorities given their social sufferings. We expect Evangelical Protestant congregations to organize STM travel at the highest rates compared to other traditions. Catholic congregations in the mid-20th century United States shifted attention to overseas relief, particularly as dioceses and religious orders mobilized a missionary movement in Central and South America (Smith 1996). This movement was supported by prioritization of international development in Roman Catholic social encyclicals and the increased centrality of U.S.-based Catholic organizations and their monies in international development (Calderisi 2013). Catholic international organizations (e.g., Catholic Relief Services) and religious orders facilitated massive flows of financial and organizational support to the developing world, especially helping to build the capacities of foreign native Catholic organizations (Calderisi 2013). Local Catholic parishes have been an important financial source in this movement. As a result of this structure and legacy, Wuthnow (2009) reports that Catholic congregations are leaders in the support of international hunger relief and refugee work, while Schnable (2015b) reports that individual Catholics are relatively more likely to donate to international aid and support Catholic international efforts. However, parishes appear to transmit financial support for foreign activity through professionalized organizations (Ammerman 2005) and Catholic congregations are characterized by deflated rates of civic involvement by members (Bane 2005). We expect Catholic congregations to have a comparatively low rate of STM travel. Beginning in the mid-19th century, Black Protestant denominations and missionary societies actively sponsored foreign missionaries, primarily to Africa and the Caribbean (Cornelius 2002; Killingray 2003; Lincoln and Mamiya 1990). By World War I, however, Black Protestant denominations, which had been instruments of foreign-focused involvement for congregations, scaled down foreign-oriented missionary work (Ammerman 2005). This withdrawal was partially in response to the domestic pressures of African American organizational life under Jim Crow and partially due to colonial authorities in Africa seeking to prevent ideals of racial equality or pan-African solidarity (Jacobs 1980; Killingray 2003). Despite declines in missionary activity, “most Afro-Americans learned of Africa through their churches or missionaries who had been stationed in Africa” (Jacobs 1980:167). And, as a result, tens of thousands of congregations affiliated with U.S.-based Black Protestant denominations were formed in other countries, especially on the continent of Africa (Lincoln and Mamiya 1990). One exception to the pattern of post-World War I decline among Black Protestant groups was the growth of global missions activity by Pentecostal and Holiness groups beginning around 1900 (Case 2006). During the interwar years, the Holiness focus on redeeming Africa even laid the groundwork for a major domestic political movement like Garveyism (Roll 2010). There is some evidence that the Holiness tradition within Black Protestant Christianity even increased foreign mission activity after World War II (Cornelius 2002). Still, up until the present day, the percent of African American missionaries among the total Christian missionary population never came close to matching the percentage of African Americans in the U.S. adult population (Killingray 2003). Recent research on Black Protestant congregations suggests that, despite a strong public commitment to evangelism, their external activities still tilt relatively toward involvement in domestic activities that are directly related to overcoming the political and economic inequality of Black communities (Ammerman 2005; Barnes 2005). While there has been renewed emphasis on cooperative “foreign missions” activity between Black Protestant denominations, as well as some evidence of embracing STM (Sutherland 2003), we expect Black Protestant congregations to have a comparatively low rate of STM travel. Among non-Christian congregations, the activities and dynamics associated with “STM travel” likely occur in different styles. For minority status religious groups in the United States, such travel may incorporate visits to religious homelands alongside aid provision. For example, travel is widely used for Jewish identity construction among U.S.-based Jewish groups like Birthright Israel (Kelner 2010). Because of a less extensive institutional structure, we expect relatively lower rates of travel among non-Christian congregations in general. One specific aspect of any religious tradition that may foster transnational interaction is the congregational hosting of foreign clergypersons. These clergypersons may be missionaries, aid workers, or heads of foreign congregations (Wuthnow 2009), present in the U.S. for education, awareness raising, or fundraising (Offutt 2015). Similar to missionary speakers, they can cultivate knowledge and motivation for international aid (Schnable 2015b). Their presence varies by religious tradition and may help account for differences in prompting transnational engagement. To summarize, we expect that rates of STM travel will vary by religious tradition with the greatest likelihood of travel among Evangelical Protestant congregations, followed by Mainline Protestant, Catholic, Black Protestant, and Non-Christian congregations (in that order). We expect that congregations that host foreign clergypersons will be more likely to produce STM travel. Resources STM travel requires an extensive set of resources to be mobilized. For example, Wuthnow (2009:180) reports that the average short-term trip costs “at least $1,000 per person….” Three types of resources are likely to influence STM travel by congregations. Congregational financial funds are a main resource for STM travel, particularly because of their size and fungible character. Ammerman (2005) reports that congregations with small budgets are disproportionately uninvolved with foreign engagement activity. We expect that congregations with greater organizational income will be more likely to produce STM travel. While congregations themselves may provide funds for STM travel, case studies of individual STM trips show that the costs of travel are sometimes treated as external to organizational budgets, with would-be travelers required to fundraise from family, friends, and other congregants (Howell 2012; Wuthnow 2009). This would make STM travel especially available to congregations with wealthier members. We expect that congregations with greater levels of high-income persons will be more likely to produce STM travel because of greater extra-congregational financial resources. Increased congregational staff could also facilitate trip production since international travel requires extensive planning (Wuthnow 2009). Since STM travel is especially popular among youth (Trinitapoli and Vaisey 2009), a youth minister or official youth leader would likely facilitate trip production (Wuthnow 2009). We expect that congregations with more staff members as well as those with role specialization in a youth minister will be more likely to produce STM travel. Congregants’ Age and Education According to descriptions of STM trips, congregants’ age and educational profiles may influence congregational trip production (Ammerman 2005; Howell 2012; Probasco 2013; Wuthnow 2009). Teenagers have biographical availability at numerous times during the year and are considered to be at a prime developmental age for such a religious experience (Smith 2005). Young adults are also relatively biographically available and increasingly have familiarity with STM travel from their teenage years (Howell 2012; Ver Beek 2006; Wuthnow 2009). Wuthnow (2009) reports that young adults are twice as likely as middle-aged adults to go on such a trip. We expect that congregations with higher levels of teenagers as well as higher proportions of young adults will be more likely to produce STM travel. A second demographic characteristic, educational level, is correlated with trip participation, such that college education doubles the likelihood of an individual going on a STM trip (Wuthnow 2009). Educational level may indicate an orientation toward global cosmopolitanism or a desire to learn about global others (Snee 2013). We expect that congregations with higher-educated congregants will be more likely to produce STM travel. Immigrants in Congregations and Elements of Immigrant Congregations While previous research has suggested the connection of the preceding factors to STM travel, we know of no research that fully theorizes the connection of immigration and STM travel. We provide that here. Immigrant groups arriving in the United States since 1965 have been more religiously and ethnically diverse (Kivisto 2016; Warner and Wittner 1998; Yang and Ebaugh 2001). As many scholars have noted (Ebaugh 2004; Warner and Wittner 1998), the frequency, density, and diversity of transnational interactions of post-1965 immigrants to the United States are distinct. Writing in the late 1990s, Menjívar (1999:592) noted that “many of today’s immigrants can easily remain active in their homeland communities through travel and/or other means, such as remittances, telephone and video conferencing, and continued streams of immigrants….” Levitt’s (1998, 2001) landmark study in the Boston area helped to define the transnational, religious dimension of recent immigration by conceptualizing what she called social remittances: cultural elements of life, such as norms and practices, that could be carried by individuals, especially for intentional projects of creating transnational community. Extending this dynamic, Kurien (2014) has shown the sizeable “direct effect” immigrants can have on their sending communities, particularly in how they transform organizations in their homelands. Surprisingly, the small amount of research on transnational civic engagement by U.S. congregations that includes information about immigrants is not focused on ideal-typical “immigrant congregations.” For example, Alison Schnable (2015b) showed the self-reported giving of international aid by regular churchgoers is likely influenced by the presence of immigrants in a congregation. She suggested that “[i]mmigrant members of congregations can be sources of information about global problems and can open broader channels of collaboration…” (Schnable 2015b: 91). Bakker’s (2014) study of “twinning” relationships between United States and foreign congregations shows both of these possibilities. Though Bakker was not directly interested in this pattern, four of the twelve congregational cases in her research, across three different religious traditions, showed evidence of immigrant members playing a key role in establishing a “twinning” relationship. For example, a suburban Catholic Church comprised of upper-middle-class families twinned with a parish in Haiti due to an immigrant with continued Haitian relationships. A suburban, middle-class Mainline Protestant church began its twinning relationship through the efforts of a Liberian refugee family. Summarizing his cases of transnational engagement, Wuthnow (2009:144) noted that some “churches initiate direct programs in particular locations because an immigrant from that country becomes a member of their congregation.” From this evidence, we expect congregations with any amount of immigrants to be more likely than congregations with no immigrants to produce STM travel. Further, we expect that a specific type of immigrant—foreign-born clergypersons—will make congregations more likely to produce STM travel (Bakker 2014; Offutt 2015; Schnable 2015b). These hypotheses are about immigrants within congregations in general, but a close read of research literature about “immigrant congregations” reveals that many of those facilitate transnational civic activity by their members as well. For example, Levitt (1998:78) noted the presence of transnational “adopt-a-parish-type relationships” that offered “financial and material support” abroad. Menjívar (1999:604) described a congregation that “coordinated a project in El Salvador to help orphan children” as well as sponsored “missionary groups” that visited El Salvador. Ebaugh (2004) reported that immigrant congregations often supported their sending communities. Vietnamese Buddhist and Catholic congregations sent aid to “churches and temples in Vietnam” (Ebaugh 2004:223). The types of aid that these congregations sent abroad were distinct: targeted toward specific location and organizations, but not “religiously relevant” (Ebaugh 2004:175). Aid included items like financial capital, technology, technological expertise, and books. In each of these examples, the reported transnational activity is strikingly similar to STM travel: the movement of persons and resources from U.S.-based congregations to foreign religious organizations for localized aid beyond strictly “religious” activity. One conclusion from this evidence would be that “immigrant congregations” are more likely to produce transnational STM travel. However, there is debate about what characterizes an “immigrant congregation.” For example, Cadge (2008) has critiqued the category of immigrant congregation as collapsing heterogeneity. Chafetz and Ebaugh (2002) have advocated moving away from an ideal-typical approach to immigrant-focused congregations. Even moreso, Cadge and Ecklund (2007) suggest including various organizational characteristics as independent variables to adequately address the diversity of processes within the category of “immigrant congregation.” To this end, we focus on two dimensions of immigrants and congregations: immigrant generation and congregational style of serving immigrants. Research on immigrants, congregations, and civic engagement has often tied these two dimensions together. For example, when first-generation immigrants dominate a congregation, research suggests formal organized civic engagement may be weak. Ley (2008) characterized first-generation immigrant congregations as “hubs” that were not civically engaged, instead providing informal services like clergy counseling or job referrals to immigrant congregants. Ebaugh and Pipe’s (2001) analysis of first-generation immigrant religious groups in the Houston area concluded that few had organized social services, instead providing support through informal networks. In this line of interpretation, first-generation immigrants produce an internally focused, resource-poor congregation that avoids external engagement. However, we expect immigrant generation to influence the activity of STM travel in a more complex way. As hypothesized above, their presence in a congregation increases the possibility of congregation-based STM travel. There may be a tipping point, though, at which a high proportion of first-generation immigrants in a congregation might hinder transnational engagement and thus depress STM travel. For example, Ebaugh et al. (Chafetz and Ebaugh 2002; Ebaugh 2004) report that congregations defined by the heavy presence of first-generation immigrants often receive foreign support in early years, instead of sending it.1 These are the immigrant hub congregations that may be solely focused on domestic settlement and assimilation work. The case research suggests that the presence of the second generation—the children of original immigrants—or other nonimmigrant groups within a congregation, influences congregations to reach beyond the needs of the congregational community (Ebaugh 2004; Ebaugh and Pipes 2001; Ley 2008). In other words, a congregation with recent immigrants, but that is not dominated by such immigrants, should be more likely to produce STM travel. In the one study to examine immigrant generation and congregational STM travel activity, albeit without multivariate controls, Wuthnow (2009) found that high levels of first-generation immigrants were unrelated to STM travel, but that low levels of first-generation immigrants were significantly, positively associated with STM travel. Given this, we expect that congregations with any recent immigrants, when compared to congregations with no such immigrants, will be more likely to produce STM travel. We also expect that congregations with a higher proportion of recent immigrants will not be more likely to do so. The second dimension of congregations related to immigrants that we expect to influence STM production is the style of congregational service to immigrants, independent of the length of residency of immigrants or the presence of multiple generations of immigrants. Case studies suggest that congregations with an internal, informal type of focus may limit external activities, even transnational activities (Ebaugh and Pipes 2001; Menjívar 1999). We expect that congregations that serve immigrants solely through informal, internally focused services will be less likely to produce transnational STM travel because of an organizational identity focused on the immediate, localized needs of immigrants. By contrast, we expect congregations with formal, externally focused services to immigrants to be more likely to produce transnational STM, an example of a bridging identity attentive to social needs and community beyond the congregation (Kurien 2013). We expect congregations with no services to immigrants to be the least likely to produce STM travel, as they have no prioritized focus on the transnational identities and relationships of immigrants. DATA, MEASURES, AND METHOD We use data from the 2012 NCS (Chaves and Anderson 2014).2 The 2012 NCS provides an extensive set of variables relevant to STM travel and immigrants. The NCS is a representative sample of congregations constructed through respondents to the 2012 General Social Survey. Congregations were identified through a hypernetwork sample in which GSS respondents that reported attending religious services at least once a year were then asked to report the name and location of their religious congregation. The resulting set of congregations was contacted to participate in a key informant survey. Key informant surveys, including the NCS, have been validated as an accurate way to collect congregational information, particularly information related to directly observable characteristics instead of attitudes (Frenk et al. 2011). The response rate for the 2012 NCS is 73%.3 Dependent Variable We use a dichotomous variable that indicates whether in the previous 12 months a congregation had a group “to travel to another country to provide assistance to people in need” (0 = No, 1 = Yes). A particular strength of this question is its concentration on aid-focused travel—a common element of STM travel—without Christian-centric “mission” terminology. Independent Variables Religious tradition hypotheses are based off of Steensland et al. (2000, Woodberry et al. 2012). Religious tradition categories include White Evangelical Protestant, Catholic, White Mainline Protestant, Black Protestant, and non-Christian. The final category is heterogeneous, including Jewish, Muslim, and other non-Christian groups. The default NCS scheme categorizes a congregation as Black Protestant if it is affiliated with a traditionally Black denomination/movement or if it is affiliated with a White Evangelical or White Mainline denomination and has a membership that is 80% or more African American/Black. Out of concern that default coding merges congregations from the original Black denominations that may have distinct foreign activity patterns with Black majority congregations in other denominations, we recoded the latter congregations into their appropriate religious tradition categories.4 We measure foreign clergy speakers by whether a congregation indicated it “hosted a clergyperson or preacher who lives in another country” in the previous 12 months (0 = No, 1 = Yes). Our resource hypotheses use three variables. The first is financial resources, from the question, “What is the total amount of money your congregation received in income from all sources during your most recent fiscal year?” Answers in whole dollars were divided by $10,000. The second is the presence of high-income individuals. We use the percentage of regularly participating adults that live in households with incomes higher than $140,000 a year. The third is a dichotomous variable (0 = No, 1 = Yes) from the question, “Do you have a youth minister or other leader specially designated to coordinate activities for youth?” Hypotheses for membership demographics use three measures. For the measure of the presence of teenagers, we use the number of regularly participating teenagers “between 13 and 19 years old” in the congregation. For the measure of young adults, we use “What percent of the regular adult participants would you say are under 35 years old?” For a measure of college-educated adults, we use “About what percentage [of regular adult participants] would you say have four-year degrees or more?” Our hypotheses related to immigrant characteristics use three measures. To indicate whether a clergy member is foreign born, we reverse-coded the question, “Was the [head clergyperson] born in the United States?” to indicate foreign birth (0 = No, 1 = Yes). For a measure of immigrant generation, we draw from the question, “Of regular adult participants in your congregation, what percent would you have come to the United States within the past five years?” This measure indicates the proportion of the congregation comprised of first-generation individuals that recently immigrated. From this, we created a three-category measure. Our base category includes congregations with zero recent immigrants. Based on Wuthnow’s (2009) evidence about “low numbers” of immigrants, our next category indicates congregations with up to 5% recent immigrants. We created a final category that represents a congregation with a moderate to high portion of first-generation immigrants (5% ≤ 100%).5 Our final immigrant-related measure is built from three variables: whether in a congregation a “group met to offer services for immigrants, such as legal assistance, translation, or job assistance” (0 = No, 1 = Yes), whether the congregation had a group to teach English (0 = No, 1 = Yes), and whether the congregation answered an open-ended question about social service and human service programs by indicating a program for immigrants, migrants, or refugees (0 = No, 1 = Yes). The first question is an indicator of informal service provision oriented toward individual immigrant cases; the second is an indicator of internal service that aids assimilation; and the third indicates an organized, formal program that targets those outside the congregation. Congregations were coded as “internal” servers if they indicated “yes” to either or both of the first questions, but “no” to the third question; “external” servers if they indicated “yes” to the third question; and “nonservers” if they indicated “no” to all. These questions are not focused toward “recent” immigrants, so they may measure services provided to any immigrant groups, including first-generation immigrants with long residency as well as second-generation immigrants. Control Variables We control for congregational size using the number of regular adult attenders (divided by 100). Theological orientation is based on a set of dummy variables indicating the theological leanings of a congregation: “liberal,” “in the middle,” or “conservative,” which serves as the base category. To control for denominational influence, a dummy variable shows whether a congregation is affiliated with a denomination (0 = No, 1 = Yes). To control for general congregational external engagement, we indicate whether a congregation has any reported social service program (0 = No, 1 = Yes). We also control for percent of members that are over the age of 60, percent that are white, and percent that are female. We control for location in two ways. The first is by the inclusion of dummy variables gauging urban, suburban, or rural (reference category) location. The second is the inclusion of dummy variables indicating regional location, with Northeast/Atlantic as the reference category. Analytic Method The data are missing at random (MAR), with the income variable having the most missing values at 17% of cases. To retain as much case information as possible, missing values were imputed in Stata 14 using chained equations (White et al. 2011), which allow for models appropriate to diverse variable types (categorical, continuous, ordinal). The imputation model included all variables used in regression analyses in addition to variables associated with salient aspects of congregational difference (denominational affiliation, moral boundaries, context). Of 1,331 cases in the 2012 NCS, we were able to reliably impute missing data for all variables in 1,280 cases. All results are presented using 20 imputed data sets.6 We present univariate and bivariate statistics using a congregation-level weight that adjusts for sample selection probability due to size. Next, we use logistic regression to model the effect of variables on the likelihood of STM travel production. Our regression analysis is presented with no weighting, having passed the test from DuMouchel and Duncan (1983).7 RESULTS Table 1 presents all variables used in the analysis. In 2012, 27% of U.S. congregations reported STM travel.8 Table 1 Univariate Descriptive Statistics   Mean  SD  Foreign STM travel  .27  .45  Religious tradition   Roman Catholic  .06     White Evangelical Protestant  .48     Black Protestant  .18     White Mainline Protestant  .22     Non-Christian  .07     Hosted foreign clergy  .30  .46  Resources   % Adults in high-income households  6.93  12.64   Income (×$10000)  53.58  212.80   Youth minister/leader  .54  .50   No. of full-time staff  3.02  11.75  Demographics   % Adults with BA degree  32.18  27.13   % Young adults  24.79  18.24   No. of teenagers  22.31  71.56  Immigrant-related   % Recent immigrants: 0  .81     % Recent immigrants: 0 to 5  .14     % Recent immigrants: > 5  .05     No services for immigrants  .89     Internal service for immigrants  .01     External service for immigrants  .10     Foreign-born head clergy  .09  .29  Controls   No. of regular attenders (×100)  1.18  3.23   Has denomination  .76  .42   Has a social service program  .82  .38   Conservative theology  .63     Moderate theology  .25     Liberal theology  .12     % Senior adults  36.62  24.82   % White  62.46  42.47   % Female  61.98  12.85   Urban location  .50     Suburban location  .18     Rural location  .32     Mid-Atlantic/Northeast  .12     Midwest  .23     South  .51     Mountain/Pacific  .14      Mean  SD  Foreign STM travel  .27  .45  Religious tradition   Roman Catholic  .06     White Evangelical Protestant  .48     Black Protestant  .18     White Mainline Protestant  .22     Non-Christian  .07     Hosted foreign clergy  .30  .46  Resources   % Adults in high-income households  6.93  12.64   Income (×$10000)  53.58  212.80   Youth minister/leader  .54  .50   No. of full-time staff  3.02  11.75  Demographics   % Adults with BA degree  32.18  27.13   % Young adults  24.79  18.24   No. of teenagers  22.31  71.56  Immigrant-related   % Recent immigrants: 0  .81     % Recent immigrants: 0 to 5  .14     % Recent immigrants: > 5  .05     No services for immigrants  .89     Internal service for immigrants  .01     External service for immigrants  .10     Foreign-born head clergy  .09  .29  Controls   No. of regular attenders (×100)  1.18  3.23   Has denomination  .76  .42   Has a social service program  .82  .38   Conservative theology  .63     Moderate theology  .25     Liberal theology  .12     % Senior adults  36.62  24.82   % White  62.46  42.47   % Female  61.98  12.85   Urban location  .50     Suburban location  .18     Rural location  .32     Mid-Atlantic/Northeast  .12     Midwest  .23     South  .51     Mountain/Pacific  .14    Source: 2012 National Congregations Study. Imputed data (m = 20); weighted. View Large Table 1 Univariate Descriptive Statistics   Mean  SD  Foreign STM travel  .27  .45  Religious tradition   Roman Catholic  .06     White Evangelical Protestant  .48     Black Protestant  .18     White Mainline Protestant  .22     Non-Christian  .07     Hosted foreign clergy  .30  .46  Resources   % Adults in high-income households  6.93  12.64   Income (×$10000)  53.58  212.80   Youth minister/leader  .54  .50   No. of full-time staff  3.02  11.75  Demographics   % Adults with BA degree  32.18  27.13   % Young adults  24.79  18.24   No. of teenagers  22.31  71.56  Immigrant-related   % Recent immigrants: 0  .81     % Recent immigrants: 0 to 5  .14     % Recent immigrants: > 5  .05     No services for immigrants  .89     Internal service for immigrants  .01     External service for immigrants  .10     Foreign-born head clergy  .09  .29  Controls   No. of regular attenders (×100)  1.18  3.23   Has denomination  .76  .42   Has a social service program  .82  .38   Conservative theology  .63     Moderate theology  .25     Liberal theology  .12     % Senior adults  36.62  24.82   % White  62.46  42.47   % Female  61.98  12.85   Urban location  .50     Suburban location  .18     Rural location  .32     Mid-Atlantic/Northeast  .12     Midwest  .23     South  .51     Mountain/Pacific  .14      Mean  SD  Foreign STM travel  .27  .45  Religious tradition   Roman Catholic  .06     White Evangelical Protestant  .48     Black Protestant  .18     White Mainline Protestant  .22     Non-Christian  .07     Hosted foreign clergy  .30  .46  Resources   % Adults in high-income households  6.93  12.64   Income (×$10000)  53.58  212.80   Youth minister/leader  .54  .50   No. of full-time staff  3.02  11.75  Demographics   % Adults with BA degree  32.18  27.13   % Young adults  24.79  18.24   No. of teenagers  22.31  71.56  Immigrant-related   % Recent immigrants: 0  .81     % Recent immigrants: 0 to 5  .14     % Recent immigrants: > 5  .05     No services for immigrants  .89     Internal service for immigrants  .01     External service for immigrants  .10     Foreign-born head clergy  .09  .29  Controls   No. of regular attenders (×100)  1.18  3.23   Has denomination  .76  .42   Has a social service program  .82  .38   Conservative theology  .63     Moderate theology  .25     Liberal theology  .12     % Senior adults  36.62  24.82   % White  62.46  42.47   % Female  61.98  12.85   Urban location  .50     Suburban location  .18     Rural location  .32     Mid-Atlantic/Northeast  .12     Midwest  .23     South  .51     Mountain/Pacific  .14    Source: 2012 National Congregations Study. Imputed data (m = 20); weighted. View Large Table 2 presents the bivariate relationships of key independent variables with STM travel. Most of the factors we hypothesized are significantly related to STM travel at the bivariate level. Congregations with immigrants, especially relatively low levels of immigrants, are more likely to produce STM travel. In fact, these congregations represent 30% of congregations that produce STM travel, nearly twice the STM travel rate of congregations that do not have immigrants. Table 2. Religious Tradition, Resource, Demographic, and Immigrant-Related Variables by Foreign STM Travel in 2012   STM trip?  pa  No  Yes  Religious tradition   Roman Catholic  0.81  0.19  ***   White Evangelical Protestant  0.67  0.33   Black Protestant  0.79  0.21   White Mainline Protestant  0.79  0.21   Non-Christian  0.69  0.31   Hosted foreign clergy  0.22  0.5  ***  Resources   % Adults in high-income households  5.56  10.58  ***   Income (× $10,000)  32.49  109.99  ***   Youth minister/leader  0.49  0.66  ***   No. of full-time staff  1.59  6.82  ***  Demographics   % Adults with BA degree  30.47  36.82  ***   % Young adults  22.24  31.53  ***   No. of teenagers  18.94  31.08  ***  Immigrant-related   Recent immigrants    % Recent immigrants: 0  0.77  0.23  ***    % Recent immigrants: 0 to 5  0.58  0.42    % Recent immigrants: > 5  0.55  0.45   Services to immigrants    No services for immigrants  0.53  0.47  ***    Internal services for immigrants  0.76  0.24    External services for immigrants  0.75  0.25   Foreign-born head clergy  0.09  0.1      STM trip?  pa  No  Yes  Religious tradition   Roman Catholic  0.81  0.19  ***   White Evangelical Protestant  0.67  0.33   Black Protestant  0.79  0.21   White Mainline Protestant  0.79  0.21   Non-Christian  0.69  0.31   Hosted foreign clergy  0.22  0.5  ***  Resources   % Adults in high-income households  5.56  10.58  ***   Income (× $10,000)  32.49  109.99  ***   Youth minister/leader  0.49  0.66  ***   No. of full-time staff  1.59  6.82  ***  Demographics   % Adults with BA degree  30.47  36.82  ***   % Young adults  22.24  31.53  ***   No. of teenagers  18.94  31.08  ***  Immigrant-related   Recent immigrants    % Recent immigrants: 0  0.77  0.23  ***    % Recent immigrants: 0 to 5  0.58  0.42    % Recent immigrants: > 5  0.55  0.45   Services to immigrants    No services for immigrants  0.53  0.47  ***    Internal services for immigrants  0.76  0.24    External services for immigrants  0.75  0.25   Foreign-born head clergy  0.09  0.1    Source: 2012 National Congregations Study. Imputed data (m = 20); weighted. aWald test. †p < .10; *p < .05; **p < .01; ***p < .001, two-tailed tests. View Large Table 2. Religious Tradition, Resource, Demographic, and Immigrant-Related Variables by Foreign STM Travel in 2012   STM trip?  pa  No  Yes  Religious tradition   Roman Catholic  0.81  0.19  ***   White Evangelical Protestant  0.67  0.33   Black Protestant  0.79  0.21   White Mainline Protestant  0.79  0.21   Non-Christian  0.69  0.31   Hosted foreign clergy  0.22  0.5  ***  Resources   % Adults in high-income households  5.56  10.58  ***   Income (× $10,000)  32.49  109.99  ***   Youth minister/leader  0.49  0.66  ***   No. of full-time staff  1.59  6.82  ***  Demographics   % Adults with BA degree  30.47  36.82  ***   % Young adults  22.24  31.53  ***   No. of teenagers  18.94  31.08  ***  Immigrant-related   Recent immigrants    % Recent immigrants: 0  0.77  0.23  ***    % Recent immigrants: 0 to 5  0.58  0.42    % Recent immigrants: > 5  0.55  0.45   Services to immigrants    No services for immigrants  0.53  0.47  ***    Internal services for immigrants  0.76  0.24    External services for immigrants  0.75  0.25   Foreign-born head clergy  0.09  0.1      STM trip?  pa  No  Yes  Religious tradition   Roman Catholic  0.81  0.19  ***   White Evangelical Protestant  0.67  0.33   Black Protestant  0.79  0.21   White Mainline Protestant  0.79  0.21   Non-Christian  0.69  0.31   Hosted foreign clergy  0.22  0.5  ***  Resources   % Adults in high-income households  5.56  10.58  ***   Income (× $10,000)  32.49  109.99  ***   Youth minister/leader  0.49  0.66  ***   No. of full-time staff  1.59  6.82  ***  Demographics   % Adults with BA degree  30.47  36.82  ***   % Young adults  22.24  31.53  ***   No. of teenagers  18.94  31.08  ***  Immigrant-related   Recent immigrants    % Recent immigrants: 0  0.77  0.23  ***    % Recent immigrants: 0 to 5  0.58  0.42    % Recent immigrants: > 5  0.55  0.45   Services to immigrants    No services for immigrants  0.53  0.47  ***    Internal services for immigrants  0.76  0.24    External services for immigrants  0.75  0.25   Foreign-born head clergy  0.09  0.1    Source: 2012 National Congregations Study. Imputed data (m = 20); weighted. aWald test. †p < .10; *p < .05; **p < .01; ***p < .001, two-tailed tests. View Large In Table 3, we introduce sets of theoretical variables based on our hypotheses. Each set of theoretical variables in Models 1 through 5 passed an F-test that they were not jointly equal to zero. Since coefficients across logistic regression models with different specifications are not comparable, we focus our interpretation on Model 6, which includes all variables. Model 6 suggests some evidence for our hypotheses regarding religious tradition, hosting foreign clergy, youth ministers, education level of congregants, recent immigrants, and immigrant services. Table 3 Logistic Regression of Foreign STM Travel on Religious Tradition, Resources, Demographics, Immigrant-Related Variables, and Controls   Model 1  Model 2  Model 3  Model 4  Model 5  Model 6  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  Religious tradition   Roman Catholic (base = WEP)  −1.58  ***                  −1.92  ***    (0.16)                    (0.23)     Black Protestant (base = WEP)  −.92  ***                  −.40      (0.21)                    (0.32)     White liberal/moderate (base = WEP)  −.54  **                  −.37      (0.17)                    (0.22)     Non-Christian (base = WEP)  −1.36  ***                  −1.75  ***    (0.36)                    (0.45)     Hosted foreign clergy  1.21  ***                  .84  ***  (0.13)                    (0.15)    Resources   Income (×$10,000)      .00  *              .00          (0.00)                (0.00)     % Adults in high-income      .01  *              .01          (0.00)                (0.01)     No. of full-time staff      .02  **              .01          (0.01)                (0.01)     Youth minister/leader      .74  ***              .43  *        (0.15)                (0.17)    Demographics   % Adults with BA degree          .01  ***          .01  †          (0.00)            (0.00)     % Young adults          .02  ***          .00            (0.00)            (0.00)     No. of teenagers          .00  **          .00            (0.00)            (0.00)     % Recent immigrants: 0 to 5 (base = 0%)              .75  ***      .48  **              (0.13)        (0.18)     % Recent immigrants: >5 (base = 0%)              .51  **      .57  †              −.18        (0.28)     Foreign-born head clergy              −.27        .19                (0.19)        (0.23)     Internal service immigrants (base = none)                  .64  ***  .51  **                  (0.13)    (0.19)     External service immigrants (base = none)                  .42    −.17                    (0.30)    (0.38)    Controls   No. of adults                      .00                        (0.00)     Has denomination                      −.01                        (0.20)     Mod theology (base = conservative)                      −.21                        (0.17)     Lib theology (base = conservative)                      −.08                        (0.24)     Has a social service program                      .76  *                      (0.30)     % Age 60 and older                      −.01  **                      (0.00)     % White                      .00                        (0.00)     % Female                      −.01                        (0.01)     Urban location                      −.08                        (0.22)     Suburban location                      .21                        (0.25)     Midwest                      −.44  †                      (0.23)     South                      −.45  *                      (0.22)     Mountain/Pacific                      −.35                        (0.25)    Constant  −.39  ***  −1.35  ***  −1.51  ***  −.72  ***  −.60  ***  −1.12  †  (0.11)    (0.14)    (0.16)    (0.08)    (0.07)    (0.67)    N  1,322    1,284    1,306    1,314    1,323    1,280    Pseudo-R2  .1 0    .07    .04    .02    .01    .20      Model 1  Model 2  Model 3  Model 4  Model 5  Model 6  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  Religious tradition   Roman Catholic (base = WEP)  −1.58  ***                  −1.92  ***    (0.16)                    (0.23)     Black Protestant (base = WEP)  −.92  ***                  −.40      (0.21)                    (0.32)     White liberal/moderate (base = WEP)  −.54  **                  −.37      (0.17)                    (0.22)     Non-Christian (base = WEP)  −1.36  ***                  −1.75  ***    (0.36)                    (0.45)     Hosted foreign clergy  1.21  ***                  .84  ***  (0.13)                    (0.15)    Resources   Income (×$10,000)      .00  *              .00          (0.00)                (0.00)     % Adults in high-income      .01  *              .01          (0.00)                (0.01)     No. of full-time staff      .02  **              .01          (0.01)                (0.01)     Youth minister/leader      .74  ***              .43  *        (0.15)                (0.17)    Demographics   % Adults with BA degree          .01  ***          .01  †          (0.00)            (0.00)     % Young adults          .02  ***          .00            (0.00)            (0.00)     No. of teenagers          .00  **          .00            (0.00)            (0.00)     % Recent immigrants: 0 to 5 (base = 0%)              .75  ***      .48  **              (0.13)        (0.18)     % Recent immigrants: >5 (base = 0%)              .51  **      .57  †              −.18        (0.28)     Foreign-born head clergy              −.27        .19                (0.19)        (0.23)     Internal service immigrants (base = none)                  .64  ***  .51  **                  (0.13)    (0.19)     External service immigrants (base = none)                  .42    −.17                    (0.30)    (0.38)    Controls   No. of adults                      .00                        (0.00)     Has denomination                      −.01                        (0.20)     Mod theology (base = conservative)                      −.21                        (0.17)     Lib theology (base = conservative)                      −.08                        (0.24)     Has a social service program                      .76  *                      (0.30)     % Age 60 and older                      −.01  **                      (0.00)     % White                      .00                        (0.00)     % Female                      −.01                        (0.01)     Urban location                      −.08                        (0.22)     Suburban location                      .21                        (0.25)     Midwest                      −.44  †                      (0.23)     South                      −.45  *                      (0.22)     Mountain/Pacific                      −.35                        (0.25)    Constant  −.39  ***  −1.35  ***  −1.51  ***  −.72  ***  −.60  ***  −1.12  †  (0.11)    (0.14)    (0.16)    (0.08)    (0.07)    (0.67)    N  1,322    1,284    1,306    1,314    1,323    1,280    Pseudo-R2  .1 0    .07    .04    .02    .01    .20    Source: 2012 National Congregations Study. Imputed data (m = 20); unweighted. Coefficient estimates are unstandardized logits. †p < .10; *p < .05; ** p < .01; *** p < .001. View Large Table 3 Logistic Regression of Foreign STM Travel on Religious Tradition, Resources, Demographics, Immigrant-Related Variables, and Controls   Model 1  Model 2  Model 3  Model 4  Model 5  Model 6  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  Religious tradition   Roman Catholic (base = WEP)  −1.58  ***                  −1.92  ***    (0.16)                    (0.23)     Black Protestant (base = WEP)  −.92  ***                  −.40      (0.21)                    (0.32)     White liberal/moderate (base = WEP)  −.54  **                  −.37      (0.17)                    (0.22)     Non-Christian (base = WEP)  −1.36  ***                  −1.75  ***    (0.36)                    (0.45)     Hosted foreign clergy  1.21  ***                  .84  ***  (0.13)                    (0.15)    Resources   Income (×$10,000)      .00  *              .00          (0.00)                (0.00)     % Adults in high-income      .01  *              .01          (0.00)                (0.01)     No. of full-time staff      .02  **              .01          (0.01)                (0.01)     Youth minister/leader      .74  ***              .43  *        (0.15)                (0.17)    Demographics   % Adults with BA degree          .01  ***          .01  †          (0.00)            (0.00)     % Young adults          .02  ***          .00            (0.00)            (0.00)     No. of teenagers          .00  **          .00            (0.00)            (0.00)     % Recent immigrants: 0 to 5 (base = 0%)              .75  ***      .48  **              (0.13)        (0.18)     % Recent immigrants: >5 (base = 0%)              .51  **      .57  †              −.18        (0.28)     Foreign-born head clergy              −.27        .19                (0.19)        (0.23)     Internal service immigrants (base = none)                  .64  ***  .51  **                  (0.13)    (0.19)     External service immigrants (base = none)                  .42    −.17                    (0.30)    (0.38)    Controls   No. of adults                      .00                        (0.00)     Has denomination                      −.01                        (0.20)     Mod theology (base = conservative)                      −.21                        (0.17)     Lib theology (base = conservative)                      −.08                        (0.24)     Has a social service program                      .76  *                      (0.30)     % Age 60 and older                      −.01  **                      (0.00)     % White                      .00                        (0.00)     % Female                      −.01                        (0.01)     Urban location                      −.08                        (0.22)     Suburban location                      .21                        (0.25)     Midwest                      −.44  †                      (0.23)     South                      −.45  *                      (0.22)     Mountain/Pacific                      −.35                        (0.25)    Constant  −.39  ***  −1.35  ***  −1.51  ***  −.72  ***  −.60  ***  −1.12  †  (0.11)    (0.14)    (0.16)    (0.08)    (0.07)    (0.67)    N  1,322    1,284    1,306    1,314    1,323    1,280    Pseudo-R2  .1 0    .07    .04    .02    .01    .20      Model 1  Model 2  Model 3  Model 4  Model 5  Model 6  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  b/SE  p  Religious tradition   Roman Catholic (base = WEP)  −1.58  ***                  −1.92  ***    (0.16)                    (0.23)     Black Protestant (base = WEP)  −.92  ***                  −.40      (0.21)                    (0.32)     White liberal/moderate (base = WEP)  −.54  **                  −.37      (0.17)                    (0.22)     Non-Christian (base = WEP)  −1.36  ***                  −1.75  ***    (0.36)                    (0.45)     Hosted foreign clergy  1.21  ***                  .84  ***  (0.13)                    (0.15)    Resources   Income (×$10,000)      .00  *              .00          (0.00)                (0.00)     % Adults in high-income      .01  *              .01          (0.00)                (0.01)     No. of full-time staff      .02  **              .01          (0.01)                (0.01)     Youth minister/leader      .74  ***              .43  *        (0.15)                (0.17)    Demographics   % Adults with BA degree          .01  ***          .01  †          (0.00)            (0.00)     % Young adults          .02  ***          .00            (0.00)            (0.00)     No. of teenagers          .00  **          .00            (0.00)            (0.00)     % Recent immigrants: 0 to 5 (base = 0%)              .75  ***      .48  **              (0.13)        (0.18)     % Recent immigrants: >5 (base = 0%)              .51  **      .57  †              −.18        (0.28)     Foreign-born head clergy              −.27        .19                (0.19)        (0.23)     Internal service immigrants (base = none)                  .64  ***  .51  **                  (0.13)    (0.19)     External service immigrants (base = none)                  .42    −.17                    (0.30)    (0.38)    Controls   No. of adults                      .00                        (0.00)     Has denomination                      −.01                        (0.20)     Mod theology (base = conservative)                      −.21                        (0.17)     Lib theology (base = conservative)                      −.08                        (0.24)     Has a social service program                      .76  *                      (0.30)     % Age 60 and older                      −.01  **                      (0.00)     % White                      .00                        (0.00)     % Female                      −.01                        (0.01)     Urban location                      −.08                        (0.22)     Suburban location                      .21                        (0.25)     Midwest                      −.44  †                      (0.23)     South                      −.45  *                      (0.22)     Mountain/Pacific                      −.35                        (0.25)    Constant  −.39  ***  −1.35  ***  −1.51  ***  −.72  ***  −.60  ***  −1.12  †  (0.11)    (0.14)    (0.16)    (0.08)    (0.07)    (0.67)    N  1,322    1,284    1,306    1,314    1,323    1,280    Pseudo-R2  .1 0    .07    .04    .02    .01    .20    Source: 2012 National Congregations Study. Imputed data (m = 20); unweighted. Coefficient estimates are unstandardized logits. †p < .10; *p < .05; ** p < .01; *** p < .001. View Large When compared to White Evangelical Protestant congregations, those in the Roman Catholic and non-Christian traditions are significantly less likely to produce STM travel. For example, the odds of a Roman Catholic congregation producing a STM trip were 85% lower (exp[−1.92] = .15) than those of a White Evangelical Protestant congregation, controlling for all other factors. The odds of a non-Christian congregation producing a STM trip were 83% lower (exp[−1.75] = .17) than those of a White Evangelical Protestant congregation. With relevant controls included, there is no significant difference in STM travel production among Protestant congregations, with Protestant congregations all more likely than non-Protestant congregations to produce STM travel.9 By examining pairwise contrasts, we confirmed this relative ordering, as well as no significant difference between Roman Catholic and non-Christian congregations. Overall, this ordering differs from our hypotheses by the absence of intra-Protestant variation and by the absence of difference between Roman Catholic and non-Christian congregations. Congregations that hosted a foreign clergyperson in the previous 12 months, no matter the religious tradition, were 2.3 times more likely to produce STM travel. Table 3 shows some limited evidence in support of our hypotheses regarding resource variables. The odds of a congregation producing a STM trip were 50% higher (exp[.43]) = 1.5) for congregations with a youth minister compared to those without (p < .05). We find some evidence for the effect of educational level (p < .10). Each one percent increase in adults with a college degree increased the odds of a congregational STM trip by about 0.7% (exp[.007] = .007). The resource variables of income, high-income households, and full-time staff fell just outside the cutoff of p <.10 for two-tailed tests, meaning they would be significant at p <.10 for one-tailed tests. Thus, the statistical evidence for them in this study is weak, but suggests their relevance. There is strong evidence for our hypotheses regarding the presence of immigrants. Congregations with a low level of recent, first-generation immigrants were more likely than congregations with no immigrants to produce an STM travel trip (p <.01).10 The odds of a congregation with a low level of recent immigrants producing STM travel were 60% higher (exp[.48] = 1.6) than those of a congregation with no recent immigrants. We note, too, that congregations with a higher level of recent immigrants (5%+) were also more likely than congregations with no recent immigrants to produce STM travel (p < .05). We had expected that congregations reporting external services for immigrants would be more likely to produce STM travel. However, the results show that they are not significantly different from congregations with no immigrant-serving programs at all. By contrast, congregations that reported only internal services for immigrants had 66% higher odds (exp[.512] = 1.67) than congregations with no immigrant-serving programs to produce STM travel (p < .01).11 Our results offer no support for age hypotheses or foreign-born clergy. Given the significance of both religious tradition and immigrant presence for trip production, we examined whether the effect of religious tradition on trip production was moderated by the presence of immigrants. Beyond the separate significance of these variables, we see three reasons why religious tradition might be moderated by immigrant presence.12 First, immigrant presence varies by religious tradition (Connor 2014). For example, about 28% of adult Catholics in the United States are immigrants, while immigrants make up less than 10% each of adult White Evangelical Protestants, White Mainline Protestants, and Black Protestants (Pew Forum 2014). Second, recent immigrants continue to influence all religious traditions (Connor 2012; Warner and Wittner 1998; Yang and Ebaugh 2001). Third, case studies suggest that the way that congregations connect immigrants to transnational communities, if at all, varies by religious tradition (Ebaugh and Chafetz 2002; Menjívar 2010; Mooney 2009). Table 4 examines this relationship, with the first column showing each religious tradition. The second column shows the distribution of congregations within a religious tradition across the three immigrant presence categories. Congregations in the Roman Catholic category and the non-Christian category demonstrate the highest presence of recent immigrants. The third column shows the rate of STM trip production within each immigrant category in a religious tradition. Table 4. Effect of the Interaction of Religious Tradition and Immigrant Presence on Foreign STM Travel Religious tradition  Congregations by category (%)a  Congregations within category with STM trip (%)b  Marginal effect of immigrants on STM trip (odds ratio)  p  Roman Catholic   Recent immigrants: 0  39  22  (base)     Recent immigrants: 0–5  39  16  1.30     Recent immigrants: >5  22  24  1.47    White Evangelical Protestant   Recent immigrants: 0  81  29  (base)     Recent immigrants: 0–5  13  50  3.79  ***   Recent immigrants: >5  5  57  4.79  ***  Black Protestant   Recent immigrants: 0  94  20  (base)     Recent immigrants: 0–5  4  28  4.50  **   Recent immigrants: >5  2  56  10.85  *  White Mainline Protestant   Recent immigrants: 0  85  15  (base)     Recent immigrants: 0–5  11  59  7.30  ***   Recent immigrants: >5  4  38  3.09  †  Non-Christian   Recent immigrants: 0  56  29  (base)     Recent immigrants: 0–5  34  29  1.55     Recent immigrants: >5  8  17  1.21    Religious tradition  Congregations by category (%)a  Congregations within category with STM trip (%)b  Marginal effect of immigrants on STM trip (odds ratio)  p  Roman Catholic   Recent immigrants: 0  39  22  (base)     Recent immigrants: 0–5  39  16  1.30     Recent immigrants: >5  22  24  1.47    White Evangelical Protestant   Recent immigrants: 0  81  29  (base)     Recent immigrants: 0–5  13  50  3.79  ***   Recent immigrants: >5  5  57  4.79  ***  Black Protestant   Recent immigrants: 0  94  20  (base)     Recent immigrants: 0–5  4  28  4.50  **   Recent immigrants: >5  2  56  10.85  *  White Mainline Protestant   Recent immigrants: 0  85  15  (base)     Recent immigrants: 0–5  11  59  7.30  ***   Recent immigrants: >5  4  38  3.09  †  Non-Christian   Recent immigrants: 0  56  29  (base)     Recent immigrants: 0–5  34  29  1.55     Recent immigrants: >5  8  17  1.21    Source: 2012 National Congregations Study. Imputed data (m = 20). aWithin each religious tradition, column may not sum to 100% due to rounding. bCell values show rate within immigrant category of a religious tradition. †p < .10; * p < .05, **p < .01; ***p < .001, two-tailed tests. View Large Table 4. Effect of the Interaction of Religious Tradition and Immigrant Presence on Foreign STM Travel Religious tradition  Congregations by category (%)a  Congregations within category with STM trip (%)b  Marginal effect of immigrants on STM trip (odds ratio)  p  Roman Catholic   Recent immigrants: 0  39  22  (base)     Recent immigrants: 0–5  39  16  1.30     Recent immigrants: >5  22  24  1.47    White Evangelical Protestant   Recent immigrants: 0  81  29  (base)     Recent immigrants: 0–5  13  50  3.79  ***   Recent immigrants: >5  5  57  4.79  ***  Black Protestant   Recent immigrants: 0  94  20  (base)     Recent immigrants: 0–5  4  28  4.50  **   Recent immigrants: >5  2  56  10.85  *  White Mainline Protestant   Recent immigrants: 0  85  15  (base)     Recent immigrants: 0–5  11  59  7.30  ***   Recent immigrants: >5  4  38  3.09  †  Non-Christian   Recent immigrants: 0  56  29  (base)     Recent immigrants: 0–5  34  29  1.55     Recent immigrants: >5  8  17  1.21    Religious tradition  Congregations by category (%)a  Congregations within category with STM trip (%)b  Marginal effect of immigrants on STM trip (odds ratio)  p  Roman Catholic   Recent immigrants: 0  39  22  (base)     Recent immigrants: 0–5  39  16  1.30     Recent immigrants: >5  22  24  1.47    White Evangelical Protestant   Recent immigrants: 0  81  29  (base)     Recent immigrants: 0–5  13  50  3.79  ***   Recent immigrants: >5  5  57  4.79  ***  Black Protestant   Recent immigrants: 0  94  20  (base)     Recent immigrants: 0–5  4  28  4.50  **   Recent immigrants: >5  2  56  10.85  *  White Mainline Protestant   Recent immigrants: 0  85  15  (base)     Recent immigrants: 0–5  11  59  7.30  ***   Recent immigrants: >5  4  38  3.09  †  Non-Christian   Recent immigrants: 0  56  29  (base)     Recent immigrants: 0–5  34  29  1.55     Recent immigrants: >5  8  17  1.21    Source: 2012 National Congregations Study. Imputed data (m = 20). aWithin each religious tradition, column may not sum to 100% due to rounding. bCell values show rate within immigrant category of a religious tradition. †p < .10; * p < .05, **p < .01; ***p < .001, two-tailed tests. View Large The rates of STM trip production vary markedly by immigrant presence category within each religious tradition. The evidence in column three suggests that STM travel production in each of the three Protestant traditions is substantially higher for congregations where immigrants are present. For example, 29% of White Evangelical Protestant congregations without recent immigrants report STM trip production, but 50% of White Evangelical Protestant congregations with immigrant presence of up to 5% report it. The effect of immigrant presence appears even greater among White Mainline Protestant congregations. Fifteen percent of White Mainline Protestant congregations without immigrants report STM trip production, but those congregations with immigrant presence of up to 5% report STM trip production at a rate nearly four times higher (59%). Black Protestant congregations, too, have higher STM travel rates with higher immigrant presence, especially when immigrant levels are above 5%. To understand whether these patterns hold given controls, we examined the interaction effect of religious tradition and immigrant presence. Due to the difficulty of examining interaction effects in logistic regression models (Ai and Norton 2003), we present both multiplicative and additive results of interaction (Buis 2010). The multiplicative approach examines the ratio of change in STM trip production relative to the baseline odds. When we included the interaction of religious tradition by immigrant category in the full regression model, no interacted pair (e.g., Catholic × 5%+ immigrants) was significant. Nor was the test of the joint significance of all pairs significant (p = .27). This multiplicative approach, however, does not show how the differences between religious traditions in their baseline odds of STM trip production combine with changes in level of immigrant presence. The additive approach, using marginal effects, shows the difference in the expected odds of STM trip production within a religious tradition as immigrant presence changes (Buis 2010). This approach provides a more accurate look at “real world” values of the interacted variables. The last column in Table 4 displays the marginal effect of the unique values of religious tradition and immigrant category on the odds of STM trip production (Williams 2012). After controlling for all other factors, congregations have increased odds of STM trip production when compared to congregations without immigrants, no matter the religious tradition. To illustrate this overall pattern, Figure 1 displays the log odds of STM trip production for congregations, by religious tradition, at the three levels of immigrant presence. FIGURE 1. View largeDownload slide Interaction of Religious Tradition and Immigrant Presence. FIGURE 1. View largeDownload slide Interaction of Religious Tradition and Immigrant Presence. Despite this overall trend, as the final column in Table 4 also indicates, the effect of difference in immigrant presence is only significant for congregations in the three Protestant religious traditions, significantly increasing their odds of STM production. Catholic congregations and non-Christian congregations with immigrants clearly produce STM travel, but the presence of immigrants does not appear to significantly affect the likelihood of this activity relative to congregations without immigrants. DISCUSSION Relatively little research has focused on the factors that produce STM travel despite voluminous research on the civic and political activities of congregations inside the United States. Our results suggest that STM travel has become part of the congregational civic repertoire in the United States, practiced by over a quarter of congregations annually. We find some evidence that three factors used to explain other types of congregational civic activity—resources, congregant demographics, and religious tradition—may influence STM travel production. Our research also demonstrates new insight into STM travel by linking immigrant-related dynamics to congregation-based transnationalism. As we expected, organizational resources matter. Youth ministry role specialization is associated with STM trip production. STM travel has long been construed as a way to transform youth religious identity, which is further confirmed by a congregation’s formal support of a position to do just that (Trinitapoli and Vaisey 2009). While our remaining resource variables fell just outside the conventional level of statistical significance, we suspect that refined measures, especially of congregational finances or of fundraising, would confirm the salience of resource factors. Demographic factors matter for STM trip production as well. Our results for educational level suggest that STM travel is partially due to the biographical availability and global experiential focus associated with higher education backgrounds. One unexpected demographic finding was the non-association of teenagers with congregational STM trip production. We suspect that this null result is a measurement weakness rather than a valid statement that teenagers are not associated with STM travel. There are two reasons for this conclusion. First, the NCS attendee measures are the strongest regarding adults, so the teenager measure may be weak. Second, the significance of the youth minister role suggests that congregational targeting of teenagers for STM travel likely exists. One conclusion to draw from these results is that some factors, like resources and congregant educational level, which are associated with externally focused congregational activities in general are associated with STM travel as well (Chaves and Tsitsos 2001a; Todd and Houston 2013). The cultural capital and cosmopolitan orientation of highly educated congregants, as well as organizational resources, appear to orient a congregation toward overseas civic engagement. Congregations already engaged with their local communities, represented by the significant coefficient for social services programs in Table 3, are also more likely to be transnationally engaged. We note, however, two differences of STM travel production when compared to external congregational activities in other research. First, local theological differences are not salient for differentiating STM travel production. Whereas theological leanings are associated with many types of congregational external engagement, this is not the case for STM travel. STM travel, as a mode of civic engagement oriented toward “the globe,” draws congregations from all theological stripes. And, yet, these theological differences matter for what congregations do during travel and why they do it (Adler and Offutt 2018). A second difference for STM travel is the role of wealthy congregants. Whereas congregations with high proportions of wealthy individuals appear to have less social service activity (Chaves and Tsitsos 2001b), in the case of STM, our results suggest that the presence of high-income individuals may foster activity (though this was on the margins of statistical significance). This is a topic in need of further research to understand how wealth inequality among individuals and congregations influences transnational engagement, possibly recreating class-based privileges. A core finding is the patterning of STM trip production by religious tradition. Protestant congregations are more likely to produce STM travel, controlling for all other factors. This difference cannot simply be attributed to the Protestant connotation of “STM,” as the dependent variable that we used avoided mission-centric language. The civic-centric language of our dependent variable shows the continued salience of religious tradition in differentiating international civic activity. The formalized foreign aid services of Roman Catholicism, and its weak engagement levels at the congregation level, likely contribute to that tradition’s lower STM rates. So might the large presence of immigrants, which we address below. Given the heterogeneity of the non-Christian category, the explanation for its relatively weak impact on STM travel is less clear. For example, 52% of Jewish congregations (about one-quarter of congregations within the NCS’ non-Christian category) produced STM travel, about twice the rate of other non-Christian congregations. The dynamics of non-Christian congregations require better research to understand religious transnational engagement. Even though bivariate rates of STM travel varied among Protestant religious traditions, our multivariate analysis suggested that controlling for other factors removed these intra-Protestant differences. This is most surprising in the case of Black Protestant congregations, which have been characterized as more focused on local, domestic engagement (Ammerman 2005; Wuthnow 2009). We see three plausible explanations for this finding. First, previously reported Protestant differences on international activity may be due to methodological decisions that focused on highly resourced congregations (Lynn 2016) or that used mission-centric language (Wuthnow 2009). Second, controlling for resources may be especially important for understanding a resource-intensive activity like STM within this tradition. For example, past research has suggested that lower levels of resource-intensive political activity by Black Protestant congregations is due to resource inequality; when resources are accounted for, Black Protestant congregations are at least equal to other congregations in the activity (Brown 2006). Third, Black Protestant congregations may simply engage in STM travel more than scholars have noticed. Large missionary structures are less extensive in Black Protestantism (Ammerman 2005), yet large numbers of congregations affiliated with Black Protestant denominations exist outside the United States with which U.S. congregations could partner (Lincoln and Mamiya 1990). This portion of the global Christian religious field seems to be a blind spot in current sociological research on transnational civic engagement (though see Bakker 2014). One surprising result arises from the interaction of religious tradition and immigrant presence. Those religious traditions that have relatively fewer congregations with immigrants (e.g. Protestant) are also those traditions in which congregations appear to be significantly influenced in their STM trip production by the presence of immigrants. We consider this pattern as part of our theorization of the connection between STM travel and immigration in transnational religion. The Immigrant Effect and Civic Remittances For over two decades, studies of religion have shown strong evidence of the transnational character of religion produced by post-1965 immigrants. Congregations have often been portrayed as organizations crucial to domestic integration, even while immigrants themselves sustained thick transnational lives. This portrayal left the transnational civic activities of congregation out of focus in two ways. First, the portrayal focused on idea-typical “immigrant congregations” despite the fact that immigrants are to be found in congregations more generally. Second, for actual “immigrant congregations,” the role of congregational support in how immigrants aid and shape their sending communities has likely been larger than reported (Ebaugh and Chafetz 2002; Menjívar 1999). Our research has highlighted the congregational role in transnational religion as a way to understand how immigrants use or prompt organizations for transnational civic engagement, like STM travel. Our results provide evidence for an immigrant effect: the influence that immigrants can have, as immigrants, on the activity of a U.S. congregation. We hypothesized that relatively low levels of recent immigrants would differentiate congregations regarding STM production. Nearly one-third (30%) of STM travel in 2012 came from congregations with recent immigrants. Twenty-two percent of STM travel in that year came from congregations with low levels (0 to 5%) of their adult members as recent immigrants. After controlling for other factors, including a congregation’s service orientation to immigrants, our results confirmed this hypothesis. Even though we did not expect it, our results also show evidence that higher levels of recent immigrants (>5%) influence trip production.13 This effect evokes patterns familiar to researchers of immigrant religion more generally. For example, in their review of immigrant religion in the United States, Yang and Ebaugh (2001:284) wrote, “immigrants in the United States … possess rich resources, including material wealth, advanced technologies, [and] organizational skills….” To this list of what immigrants possess, we would add membership in U.S. religious congregations. Congregations can provide an organized platform for immigrants to influence foreign communities, distinct from immigrants’ individual actions or hometown associations (Vertovec 2004). We can only speculate about how recent immigrants influence STM travel production. It is possible that recent immigrants could be drawn to congregations that lack immigrants but have reputations for international engagement, making our results spurious. We would be surprised if the occurrence was frequent enough to negate our findings. We think two other pathways are more likely. The first pathway includes those congregations that support a continued stream of immigration into the United States. In this pathway, STM travel becomes part of a formalized, organized transnational relationship, with newly arrived immigrants sustaining existing transnational connections, as well as providing motivation to continue to be transnationally engaged (Hansing 2014). A second pathway are those congregations that respond to immigrants’ new presence. This includes congregations with no or little historical connection to settling earlier immigration waves, but who now have recent immigrants as congregants. In this case, immigrants may expand the geographic and social imagination of a congregation. For individual congregants, immigrants may frame and personalize international need (Schnable 2015b). From the organizational view, personal connection to a specific international locale may ground abstract congregational visions of helping the world (Wuthnow 2009), while also decreasing the start-up costs of initiating STM travel. This pattern was alluded to by Wuthnow and Offutt (2008:218), who observed that “short-term trips are generally facilitated by preexisting transnational ties.” The power of small numbers of people to activate congregational social service and political activities is well known, so STM travel should join that list (Chaves 2004). For either of these pathways, we suggest seeing the result of this immigrant effect as a civic remittance. Levitt (2001) expanded the idea of economic remittances to the noneconomic dimension of immigrant transnational life. Social remittances, she explained, are delineated and personal, grounded in transnational interaction and capable of constructing new identities (Levitt 1998). The concept of civic remittances extends Levitt’s work, providing a way to understand the sizeable, targeted, but also unequal transfer of resources which she has referred to as the “transnationalizing [of] community development” (Levitt 2001). STM travel is civic through its normative vision of partnership, its locally organized relationships, its reliance on personal interactions, and its symbolic mechanisms to manage material inequality (Adler and Offutt 2018; Probasco 2016). Like other remittances, STM travel is a delineable activity with clear material outcomes that is based on personal relationships and facilitates continued transnational interactions. In the context of organization-based aid, the concept of remittance reflects the agency that immigrants themselves can have in prompting organizational action (Bakker 2014), instead of just being a cause for aid done by others in their presence (Schnable 2015b). Our research does provide evidence that the immigrant effect and ensuing civic remittance activity varies by religious tradition. Roman Catholic and non-Christian congregations with immigrants do engage in STM travel and thus civic remittance activity; however, they do not do so at significantly higher rates than congregations in their tradition with no immigrants. This evidence echoes other studies which have documented distinct styles of remittance activity by religious groups in different religious traditions (Ebaugh and Chafetz 2002; Kurien 2014; Menjívar 1999). Some of the difference we detect may be due to differences in immigrants in these religious traditions. For example, undocumented immigrants to the United States have historically come from majority-Catholic countries (Matovina 2012; Mooney 2009). Catholic congregations in the United States may be the destinations for these immigrants, but immigrants’ presence may not prompt increased STM activity due to the travel restrictions related to legal documentation and related economic difficulties.14 Such congregations are simply limited in their ability to engage in this particular form of civic remittance. It is also possible that the vision of religious citizenship among non-Protestant congregations in a historically Protestant country affects congregations’ orientation toward congregation-based civic remittance activity. Indeed, some evidence from Catholic parishes with immigrants suggests a focus away from strong transnational connection toward local life in the United States and broad immigrant identity (Cherry 2014; Menjívar 1999; Mooney 2009). One way to interpret the large, significant difference in STM travel production by Protestant congregations with immigrants compared to those without immigrants is to consider the meaning of immigrants in those traditions. Protestant congregations with any recent immigrants stand out, as less than 20% of congregations in any of the Protestant traditions have recent immigrants present. The transnational identity and focus of these congregations may be especially salient, as they exist in a religious context in which immigrant presence produces noticeable demographic difference. They may have a different stream of immigrants to begin with (Connor 2012) and stronger connection to specific religious communities in foreign countries (Hagan 2008), both of which could encourage STM.15 Our analysis also accounted for other immigrant-related dimensions of congregations beyond immigrant presence. For example, we found that foreign-born clergy were not associated with STM travel production. The heterogeneity in the “foreign-born clergy” category likely produced this result, as some may be longtime U.S. residents without foreign connections and some may be from countries unlikely to attract STM travel (Lynn 2016; Priest et al. 2006). Finally, we hypothesized that congregations with externally focused services to immigrants would be more likely to travel. While our formal expectation was wrong, our attention to congregational patterns of serving immigrants as distinct from the presence of recent immigrants revealed that congregations with internal services alone were more likely to produce STM travel. Since our data source provided no way of measuring the proportion of a congregations’ members that are long-term U.S. residents, or even second-generation immigrants, it is possible that this measure of internal services represents continued immigrant identity in a congregation even if recent immigration is minimal. This internal orientation may be a sign of social isolation due to immigrant needs, but also a sign of bonding around an identity that still extends transnationally. Other research literature has documented a similar dynamic of settled immigrant groups focused abroad (Kurien 2014; Menjívar 1999). For example, a long-standing Cuban immigrant parish in Miami had numerous internally focused ministries built around immigrant identity and one externally focused program: a “long-distance sister relationship” to the Cuban diocese through visits with humanitarian aid and money (Hansing 2014). This finding may qualify other literature which argues that transnational dynamics atrophy as the second generation emerges in a congregation (Chafetz and Ebaugh 2002; Levitt 2009). It may not take the presence of recent, first-generation immigrants to continue certain kinds of transnational connection, like STM travel. STM travel’s well-legitimated presence and its organizational basis may even make it a more durable transnational connection that outlasts first-generation personal relationships to sending communities. There are a number of weaknesses to consider in our research. First, better resource variables regarding staff size or budget categories would help to specify how resources matter in STM travel, which they likely do. Second, there is a universe of transnationally focused religious organizations in the United States that engender STM travel demand (Schnable 2015a). Their affiliations and relationships shape what congregations might do. Third, we have no information about where STM trips went. While we follow the logic of studies that show that congregations with the immigrant effect send STM travel to those immigrants’ countries, we cannot verify this with our data. Fourth, we have some information about a congregation’s international focus, for example, a measure of visiting clergy, but we do not have a longitudinal view to show how that focus developed and whether STM travel is the endpoint of that focus. A final weakness in our research is that we have no information for the size, number, or duration of STM trips, thereby missing heterogeneity within the congregation population produced by extremely active congregations. CONCLUSION Experienced annually by millions of people living in the United States, STM travel is a congregational activity whose conditions of production have been poorly understood. Alongside organizational programs such as worship, religious education, and local civic engagement, STM travel has clearly entered the repertoire of religious activity produced by U.S. religious congregations. Our research has clarified the frequency of STM travel among congregations and demonstrated the organizational factors that lead to differential trip production. While large relief and development organizations visibly dominate the portrait of religious transnational civic engagement, our research shows how a decentralized form of civic aid is produced at the congregational level, the latest iteration in a lineage of voluntary social engagement by American religious groups (Ammerman 2005). STM travel has been recognized as an organizational mechanism to accomplish transnational influence. Here, we have shown that this influence should be seen to include transnational influences to begin with. By connecting evidence and concepts from literatures that have tended to remain separate, we have shown that the transnational flow of immigration is connected to the flow of STM travel. We have argued that this immigrant effect likely occurs across the general population of congregations, not just among “immigrant congregations.” Immigrant religious communities in United States “are forming the worldwide organizational center of their respective religions in the United States” (Chafetz and Ebaugh 2002:284) and scholars have argued that we should attend to the religious processes that lead to civic engagement by immigrants (Levitt 2008). By tracing their connection to STM travel, we have shown how immigrants can use or prompt their U.S. congregations to be transnationally engaged, thereby influencing both their sending communities and U.S. religious organizations themselves. ACKNOWLEDGMENTS Previous versions of this article were presented at the meetings of the Society for the Scientific Study of Religion. We thank Roger Finke, Stephen Offutt, and two anonymous reviewers for comments. Footnotes 1 There are some cases where the pattern is reversed, especially for immigrant groups affiliated with foreign-based religious groups or denominations (Kurien 2013). 2 The data were downloaded from the Association of Religion Data Archives, www.TheARDA.com. 3 If the GSS response rate is included, the response rate is 52%. 4 In the end, this recoding only marginally influenced coefficient size of Black Protestant religious tradition in regression models below, but not statistical significance. 5 We also tested a version that split the third category into “moderate” and “high,” based on one definition of “immigrant congregations” as those with more than 20% of congregants characterized as recent immigrants (Numrich and Kniss 2007). When used in the main regression analyses, congregations with high levels of recent immigrants (20%+) were not significantly different in producing STM travel compared to congregations with no immigrants, a result in line with theory. However, since less than 2% of congregations in the NCS sample fall into the “high” category, we could not use it for interaction effect analyses. We note that combining the “moderate” and “high” categories do not change the main results. 6 The number of imputed datasets is based on M ≥ 100 × FMI, where FMI is the fraction of missing information for a variable and M is the number of imputed datasets. See White et al. (2011). 7 We used the Stata ado file wgttest by Ben Jann to run this test. 8 This is 2% higher than the rate reported in the 2006 NCS, though the difference is not significant. Using the NCS’ attendee-based survey weight, we note that 41% of attendees in an average congregation in 2012 were in one that reported an STM trip in the previous year. 9 Both White Mainline Protestant (p < .10) and Black Protestant (p < .01) congregations were significantly different from White Evangelical Protestant congregations when all variables except immigration-related variables were included (model not shown). This suggests the importance of accounting for this dimension of STM travel production. 10 We conducted a sensitivity analysis of “token” levels of recent immigrants by setting the level of recent immigrants to zero for any congregation that reported less than 1% of regular adults as recent immigrants. There was no salient change in the results. 11 We also ran Model 6 with immigrant presence variables removed and, separately, with immigrant service variables removed. In both cases, the variables had smaller p-values and increased coefficient size. This suggests some overlap in these concepts, though variance inflation factors are below 2 for these variables. 12 We thank two reviewers for suggesting closer attention to immigration-related differences between religious traditions. 13 See footnote 5 for evidence regarding high levels of recent immigrants. 14 We thank a reviewer for this important interpretation. 15 The NCS has information about the country origins of recent immigrants. As an example of how religious tradition may shape STM, of Black Protestant congregations with recent immigrants, 79% reported that the recent immigrants came from the Caribbean or Africa. REFERENCES Adler, Gary J., and Stephen Offutt. 2018. “ The Gift Economy of Direct Transnational Civic Action: How Reciprocity and Inequality Are Managed in Religious ‘Partnerships’.” Journal for the Scientific Study of Religion. Forthcoming . Ai, Chunrong, and Edward C. Norton. 2003. “ Interaction Terms in Logit and Probit Models.” Economics Letters  80: 123– 29. Google Scholar CrossRef Search ADS   Ammerman, Nancy Tatom. 2005. Pillars of Faith: American Congregations and Their Partners . Berkeley: University of California Press. Bakker, Janel Kragt. 2014. Sister Churches: American Congregations and Their Partners Abroad . New York: Oxford University Press. Bane, Mary Jo. 2005. “ The Catholic Puzzle: Parishes and Civic Life.” In Taking Faith Seriously , edited by Mary Jo Bane, Brent Coffin, and Richard Higgins, 63– 93. Cambridge, MA: Harvard University Press. Barnes, Sandra L. 2005. “ Black Church Culture and Community Action.” Social Forces  84: 967– 94. Barnes, Sandra L. 2011. “ Black Church Sponsorship of Economic Programs: A Test of Survival and Liberation Strategies.” Review of Religious Research  53: 23– 40. Google Scholar CrossRef Search ADS   Beyerlein, Kraig, and Mark Chaves. 2003. “ The Political Activities of Religious Congregations in the United States.” Journal for the Scientific Study of Religion  42: 17. Google Scholar CrossRef Search ADS   Beyerlein, Kraig, Gary Adler, and Jennifer Trinitapoli. 2011. “ The Effect of Religious Mission Trips on Youth Civic Participation.” Journal for the Scientific Study of Religion  50: 780– 95. Google Scholar CrossRef Search ADS   Brown, R. Khari. 2006. “ Racial Differences in Congregation-Based Political Activism.” Social Forces  84: 1581– 604. Google Scholar CrossRef Search ADS   Buis, Maarten L. 2010. “ Stata Tip 87: Interpretation of Interactions in Non-Linear Models.” The Stata Journal  10: 305– 08. Cadge, Wendy. 2008. “ De Facto Congregationalism and the Religious Organizations of Post-1965 Immigrants to the United States: A Revised Approach.” Journal of the American Academy of Religion  76: 344– 74. Google Scholar CrossRef Search ADS   Cadge, Wendy, and Elaine Howard Ecklund. 2007. “ Immigration and Religion.” Annual Review of Sociology  33: 359– 79. Google Scholar CrossRef Search ADS   Calderisi, Robert. 2013. Earthly Mission: The Catholic Church and World Development . Hartford, CT: Yale University Press. Case, Jay R. 2006. “ And Ever the Twain Shall Meet: The Holiness Missionary Movement and the Birth of World Pentecostalism, 1870––1920.” Religion and American Culture: A Journal of Interpretation  16: 125– 60. Google Scholar CrossRef Search ADS   Chafetz, Janet Saltzman, and Helen Rose Ebaugh. 2002. “ The Variety of Transnational Religious Networks.” Religion Across Borders: Transnational Immigrant Networks : 165– 91. Chaves, Mark. 2004. Congregations in America . Cambridge, MA: Harvard University Press. Chaves, Mark, and Shawna L. Anderson. 2014. “ Changing American Congregations: Findings from the Third Wave of the National Congregations Study.” Journal for the Scientific Study of Religion  53: 676– 86. Google Scholar CrossRef Search ADS   Chaves, Mark, and William Tsitsos. 2001a. “ Congregations and Social Services: What They Do, How They Do It, and with Whom.” Nonprofit and Voluntary Sector Quarterly  30: 660– 83. Google Scholar CrossRef Search ADS   ———. 2001b. “ Congregations and Social Services: What They Do, How They Do It, and with Whom.” Nonprofit and Voluntary Sector Quarterly  30: 660– 83. CrossRef Search ADS   Cherry, Stephen M. 2014. Faith, Family, and Filipino American Community Life . New Brunswick, NJ: Rutgers University Press. Connor, Phillip. 2012. “ International Migration and Religious Selection.” Journal for the Scientific Study of Religion  51: 184– 94. Google Scholar CrossRef Search ADS   ———. 2014. Immigrant Faith: Patterns of Immigrant Religion in the United States, Canada, and Western Europe . New York: New York University Press. Cornelius, David. 2002. “ A Brief Historical Survey of African-American Involvement in International Missions.” In African-American Experience in World Mission: A Call Beyond Community , edited by V. J. Walston and R. J. Stevens, 48– 53. Pasadena, CA: William Carey Library. DuMouchel, William H., and Greg J. Duncan. 1983. “ Using Sample Survey Weights in Multiple Regression Analyses of Stratified Samples.” Journal of the American Statistical Association  78: 535– 43. Google Scholar CrossRef Search ADS   Ebaugh, Helen Rose. 2004. “ Religion across Borders: Transnational Religious Ties.” Asian Journal of Social Science  32: 216– 31. Google Scholar CrossRef Search ADS   Ebaugh, Helen Rose, and Janet Saltzman Chafetz. 2002. Religion Across Borders: Transnational Immigrant Networks . Lanham, MD: Altamira Press. Ebaugh, Helen Rose, and Paula Pipes. 2001. “ Immigration Congregations as Social Service Providers: Are They Safety Nets for Welfare Reform?” In Religion and Social Policy , edited by P. D. Nesbitt. AltaMira Press. Frenk, Steven M., Shawna L. Anderson, Mark Chaves, and Nancy Martin. 2011. “ Assessing the Validity of Key Informant Reports About Congregations’ Social Composition.” Sociology of Religion  72: 78– 90. Google Scholar CrossRef Search ADS   Hagan, Jacqueline Maria. 2008. Migration Miracle: Faith, Hope, and Meaning on the Undocumented Journey . Cambridge, MA: Harvard University Press. Hansing, Katrin. 2014. “ Unidos en la Fe: Transnational Civic Social Engagement between Two Cuban Catholic Parishes.” In Churches and Charity in the Immigrant City: Religion, Immigration, and Civic Engagement in Miami , edited by A. Stepick, S. J. Mahler, and T. Rey. New Brunswick, NJ: Rutgers University Press. Hefferan, Tara. 2007. Twinning Faith and Development: Catholic Parish Partnering in the U.S. and Haiti . Bloomfield, CT: Kumarian Press. Hirschman, Charles. 2004. “ The Role of Religion in the Origins and Adaptation of Immigrant Groups in the United States.” International Migration Review  38: 1206– 33. Google Scholar CrossRef Search ADS   Howell, Brian. 2010. “Roots of the Short-Term Missionary, 1960–1985”: Building Church Leaders. http://www.buildingchurchleaders.com/articles/2006/rootsmissionary.html. ———. 2012. Short-Term Mission: An Ethnography of Christian Travel Narrative and Experience . Downers Grove, IL: IVP Academic. Jacobs, Sylvia M. 1980. “ Black American Missionaries in Africa: A Selected Bibliography.” A Current Bibliography on African Affairs  13: 166– 72. Google Scholar CrossRef Search ADS   Kelner, Shaul. 2010. Tours That Bind: Diaspora, Pilgrimage, and Israeli Birthright Tourism . New York: New York University Press. Google Scholar CrossRef Search ADS   Killingray, David. 2003. “ The Black Atlantic Missionary Movement and Africa, 1780s–1920s.” Journal of Religion in Africa  33: 3– 31. Google Scholar CrossRef Search ADS   Kinney, Nancy T. 2015. “ Structure, Context, and Ideological Dissonance in Transnational Religious Networks.” Voluntas  26: 382– 407. Google Scholar CrossRef Search ADS   Kivisto, Peter. 2016. “ International Migration.” In Handbook of Religion and Society , edited by D. Yamane, 547– 68. New York, NY: Springer. Google Scholar CrossRef Search ADS   Kurien, Prema. 2013. “ Religion, Social Incorporation, and Civic Engagement: Second-Generation Indian American Christians.” Review of Religious Research  55: 81– 104. Google Scholar CrossRef Search ADS   ———. 2014. “ The Impact of International Migration on Home Churches: The Mar Thoma Syrian Christian Church in India.” Journal for the Scientific Study of Religion  53: 109– 29. CrossRef Search ADS   Levitt, Peggy. 1998. “ Local-Level Global Religion: The Case of U.S.-Dominican Migration.” Journal for the Scientific Study of Religion  37: 74– 89. ———. 2001. Transnational Villagers . Berkeley, CA: UC Press. ———. 2007. God Needs No Passport: Immigrants and the Changing American Religious Landscape . New York, NY: The New Press. ———. 2008. “ Religion as a Path to Civic Engagement.” Ethnic and Racial Studies  31: 766– 91. CrossRef Search ADS   ———. 2009. “ Roots and Routes: Understanding the Lives of the Second Generation Transnationally.” Journal of Ethnic and Migration Studies  35: 1225– 42. CrossRef Search ADS   Ley, David. 2008. “ The Immigrant Church as an Urban Service Hub.” Urban Studies  45: 2057– 74. Google Scholar CrossRef Search ADS   Lincoln, C. Eric, and Lawrence H. Mamiya. 1990. The Black Church in the African American Experience . Durham, NC: Duke University Press. Google Scholar CrossRef Search ADS   Lynn, Monty L. 2016. “ Congregational Aid: North American Protestant Engagement in International Relief and Development.” The Journal of Development Studies  52: 965– 85. Google Scholar CrossRef Search ADS   Matovina, Timothy. 2012. Latino Catholicism: Transformation in American’s Largest Church . Princeton, NJ: Princeton University Press. McAlister, Melani. 2008. “ What Is Your Heart For?: Affect and Internationalism in the Evangelical Public Sphere.” American Literary History  20: 870– 95. Google Scholar CrossRef Search ADS   McCleary, Rachel M., and Robert J. Barro. 2008. “ Private Voluntary Organizations Engaged in International Assistance, 1939–2004.” Nonprofit and Voluntary Sector Quarterly  37: 512– 36. Google Scholar CrossRef Search ADS   Menjívar, Cecilia. 1999. “ Religious Institutions and Transnationalism: A Case Study of Catholic and Evangelical Salvadoran Immigrants.” International Journal of Politics, Culture, and Society  12: 589– 612. Google Scholar CrossRef Search ADS   ———. 2010. “ Immigrants, Immigration, and Sociology: Reflecting on the State of the Discipline.” Sociological Inquiry  80: 3– 27. CrossRef Search ADS   Mooney, Margarita. 2009. Faith Makes Us Live: Surviving and Thriving in the Haitian Diaspora . Berkeley, CA: University of California Press. Numrich, Paul, and Fred Kniss. 2007. Sacred Assemblies and Civic Engagement: How Religion Matters for America’s Newest Immigrants . New Brunswick, NJ: Rutgers University Press. Offutt, Stephen. 2011. “ The Role of Short-Term Mission Teams in the New Centers of Global Christianity.” Journal for the Scientific Study of Religion  50: 796– 811. Offutt, Stephen. 2015. New Centers of Global Evangelicalism in Latin America and Africa . New York: Cambridge University Press. Google Scholar CrossRef Search ADS   Offutt, Stephen, and Grant Miller. 2016. “Transnationalism.” In Handbook of Religion and Society , edited by D. Yamane, 525– 45. Switzerland: Springer International Publishing. Google Scholar CrossRef Search ADS   Pew Research Center. 2014. Religious Landscape Study . Pew Research Center: Washington, DC. Priest, Robert J., Terry Dischinger, Steve Rasmussen, and Christopher M. Brown. 2006. “ Researching the Short-Term Mission Movement.” Missiology: An International Review  34: 431– 51. Google Scholar CrossRef Search ADS   Probasco, LiErin. 2013. “ Giving Time, Not Money: Long-Term Impacts of Short-Term Mission Trips.” Missiology: An International Review  41: 202– 24. Google Scholar CrossRef Search ADS   ———. 2016. “ Prayer, Patronage, and Personal Agency in Nicaraguan Accounts of Receiving International Aid.” Journal for the Scientific Study of Religion  55: 233– 49. CrossRef Search ADS   Roll, Jarod. 2010. “ Garveyism and the Eschatology of African Redemption in the Rural South, 1920–1936.” Religion and American Culture: A Journal of Interpretation  20: 27– 56. Google Scholar CrossRef Search ADS   Schnable, Allison. 2015a. “ New American Relief and Development Organizations: Voluntarizing Global Aid.” Social Problems  62: 309– 29. Google Scholar CrossRef Search ADS   ———. 2015b. “ Religion and Giving for International Aid: Evidence from a Survey of U.S. Church Members.” Sociology of Religion  76: 72– 94. CrossRef Search ADS   Smith, Christian. 2005. Soul Searching: The Religious and Spiritual Lives of American Teenagers. New York, NY: Oxford University Press. Smith, Christian. 1996. Resisting Reagan: The U.S. Central America Peace Movement . Chicago, IL: University of Chicago. Google Scholar CrossRef Search ADS   Snee, Helene. 2013. “ Framing the Other: Cosmopolitanism and the Representation of Difference in Overseas Gap Year Narratives.” The British Journal of Sociology  64: 142– 62. Google Scholar CrossRef Search ADS PubMed  Steensland, Brian, Jerry Z. Park, Mark D. Regnerus, Lynn D. Robinson, W. Bradford Wilcox, and Robert D. Woodberry. 2000. “ The Measure of American Religion: Toward Improving the State of the Art.” Social Forces  79: 291– 318. Google Scholar CrossRef Search ADS   Sutherland, Jim. 2003. “ How Black Is the Harvester? A Profile of the African-American Missionary Force, the Extent of the Problem.” In African-American Experience in World Mission: A Call Beyond Community , edited by V. J. Walston and R. J. Stevens, 135– 39. Pasadena, CA: William Carey Library. Thaut, Laura C. 2009. “The Role of Faith in Christian Faith-Based Humanitarian Agencies: Constructing the Taxonomy.” Voluntas 20:319–50. Todd, Nathan R., and Jaclyn D. Houston. 2013. “ Examining Patterns of Political, Social Service, and Collaborative Involvement of Religious Congregations: A Latent Class and Transition Analysis.” American Journal of Community Psychology  51: 442– 38. Google Scholar CrossRef Search ADS   Trinitapoli, Jenny, and Stephen Vaisey. 2009. “ The Transformative Role of Religious Experience: The Case of Short Term Missions.” Social Forces  88: 121– 46. Google Scholar CrossRef Search ADS   Ver Beek, Kurt Alan. 2006. “ The Impact of Short-Term Missions: A Case Study of House Construction in Honduras after Hurricane Mitch.” Missiology: An International Review  34: 477– 96. Google Scholar CrossRef Search ADS   Vertovec, Steven. 2004. “ Migrant Transnationalism and Modes of Transformation.” The International Migration Review  38: 970– 1001. Google Scholar CrossRef Search ADS   Warner, R. Stephen. 2007. “ The Role of Religion in the Process of Segmented Assimilation.” The Annals of the American Academy of Political and Social Science  612: 100– 15. Google Scholar CrossRef Search ADS   Warner, R. Stephen, and Judith G. Wittner. 1998. Gatherings in Diaspora: Religious Communities and the New Immigration . Philadelphia, PA: Temple University Press. White, Ian R., Patrick Royston, and Angela M. Wood. 2011. “ Multiple Imputation Using Chained Equations: Issues and Guidance for Practice.” Statistics in Medicine  30: 377– 99. Google Scholar CrossRef Search ADS PubMed  Williams, Richard. 2012. “ Using the Margins Command to Estimate and Interpret Adjusted Predictions and Marginal Effects.” The Stata Journal  12: 308– 31. Woodberry, Robert D., Jerry Z. Park, Lyman A. Kellstedt, Mark D. Regnerus, and Brian Steensland. 2012. “ The Measure of American Religious Traditions: Theoretical and Measurement Considerations.” Social Forces  91: 65– 73. Google Scholar CrossRef Search ADS   Wuthnow, Robert. 2009. Boundless Faith: The Global Outreach of American Churches . Los Angeles: University of California Press. Google Scholar CrossRef Search ADS   Wuthnow, Robert, and Stephen Offutt. 2008. “ Transnational Religious Connections.” Sociology of Religion  69: 187– 209. Google Scholar CrossRef Search ADS   Yang, Fenggang, and Helen Rose Ebaugh. 2001. “ Transformations in New Immigrant Religions and Their Global Implications.” American Sociological Review  66: 269– 88. Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of the Association for the Sociology of Religion. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

Sociology of ReligionOxford University Press

Published: Feb 17, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off