LITERATURE REVIEW AND THEORETICAL FRAMEWORK

A long and rich tradition on immigrant incorporation in sociology seeks to model how immigrants and their descendants fare once they arrive in the United States (Alba and Nee Reference Alba and Nee2003; Bean et al. Reference Bean, Brown and Bachmeier2015; Kasinitz et al. Reference Kasinitz, Mollenkopf, Waters and Holdaway2008; Portes and Rumbaut Reference Portes and Rumbaut2001; Zhou et al. Reference Zhou, Lee, Vallejo, Tafoya-Estrada and Xiong2008). Considerable work in this area has focused broadly on the “contexts of reception” greeting immigrants, such as the ways in which governmental policies, labor markets, and social reception by either co-ethnic or mainstream host communities influence how immigrants eventually become incorporated into American society over time (Portes and Rumbaut Reference Portes and Rumbaut2014). Much of this literature now recognizes that receptivity toward immigrants largely depends on existing racialized dynamics in the host society; moreover, it readily acknowledges that anti-Black racism is a central factor shaping U.S. policies and institutions, which in turn shapes opportunities and obstacles new immigrants encounter when they arrive (Foner and Alba Reference Foner and Alba2010; Kasinitz et al. Reference Kasinitz, Mollenkopf, Waters and Holdaway2008; Waters and Kasinitz Reference Waters, Kasinitz, Foner and Simon2015). This legacy of racial inequality intersects with geography and economy to stratify and segregate communities and neighborhoods, giving immigrants different “segments” of Americans with whom to live and interact (Masuoka and Junn Reference Masuoka and Junn2013; Portes and Rumbaut Reference Portes and Rumbaut2001, Reference Portes and Rumbaut2014).

This conceptualization situates public opinion toward immigrants as one among several key factors shaping processes of immigrant incorporation. Numerous studies have examined public opinion toward immigrants and immigration in the U.S. context (e.g., Fetzer Reference Fetzer2000; Muste Reference Muste2013; NASEM Reference Waters and Gerstein Pineau2015; Segovia and Defever Reference Segovia and Defever2010; Suro Reference Suro2009). This research has focused primarily on Whites’ attitudes, and suggests that their exclusionary attitudes toward immigrants can be explained in large part by dominant group concerns about the potential negative impacts of rising immigration (e.g., Ayers et al. Reference Ayers, Richard Hofstetter, Schnakenberg and Kolody2009; Ben-Nun Bloom et al. Reference Ben-Nun Bloom, Arikan and Lahav2015; Burns and Gimpel Reference Burns and Gimpel2000; Citrin et al. Reference Citrin, Green, Muste and Wong1997; Fussell Reference Fussell2014; Hainmueller and Hopkins 2014; Haynes et al. Reference Haynes, Merolla and Karthick Ramakrishnan2016; Hood and Morris Reference Hood and Morris2000; Hopkins Reference Hopkins2010; Huber and Espenshade Reference Huber and Espenshade1997; Masuoka and Junn Reference Masuoka and Junn2013; Newman et al. Reference Newman, Hartman and Taber2012; Scheve and Slaughter Reference Scheve and Slaughter2001; Stein et al. Reference Stein, Shirley Post and Rinden2000).

Fewer studies have examined attitudes toward immigrants among more than one U.S.- born racial group, but those that do suggest it is important to evaluate U.S.-born attitudes toward immigrants in light of the long history of White racial domination over African Americans (Bonilla-Silva Reference Bonilla-Silva1997; Desmond and Emirbayer, Reference Desmond and Emirbayer2009; Golash-Boza Reference Golash-Boza2016; Jung Reference Jung2009). For example, Lawrence D. Bobo and Vincent L. Hutchings (Reference Bobo and Hutchings1996) emphasize that the attitudes of one racial group toward another are fundamentally shaped by its own historical treatment and current positioning within White-dominated U.S. society (also see Gay Reference Gay2006; Hutchings and Wong 2004; Oliver and Wong Reference Oliver and Wong2003; Wilkinson Reference Wilkinson2015). Using survey data from Los Angeles in the 1990s, Bobo and Hutchings (Reference Bobo and Hutchings1996) found that Blacks were most likely (and Whites least likely) to see other groups as competitive threats, and that racial alienation—the feeling that one’s racial group is powerless and disenfranchised—was a key predictor of perceiving other groups as threats. Similarly, with national survey data from the 2000s, Natalie Masuoka and Jane Junn (Reference Masuoka and Junn2013) demonstrate how different status positions of Whites, Blacks, Asians, and Latinos in the U.S. racial hierarchy act like a “dispersive prism,” refracting public opinion to produce distinct views of immigrants and preferences for immigration policy among each one. They found that a strong sense of linked fate to one’s racial group was associated with more restrictive attitudes toward immigration among Whites, but less restrictive attitudes among racial minorities; these authors conclude that Whites attempt to preserve their status at the top of the U.S. racial hierarchy when evaluating immigration, whereas minority groups are more attuned to their own marginalization. Fussell (Reference Fussell2014) agrees, arguing that Blacks’ unique historical experience vis-à-vis Whites shapes their opinions toward immigration in ways that are distinct from White public opinion (also see Browne et al. Reference Browne, Deckard and Rodriguez2016; Browne et al. Reference Browne, Reingold and Kronberg2018; Diamond Reference Diamond1998; Nteta Reference Nteta2013; Wilkinson and Bingham Reference Wilkinson and Bingham2016; Williams Reference Williams2016; Williams and Hannon Reference Williams and Hannon2016). At the same time, other research shows that, like Whites, Blacks may still feel economically threatened when immigrant newcomers in their local environment show greater economic advancement and/or substantial increase in size (Browne et al. Reference Browne, Reingold and Kronberg2018; Fussell Reference Fussell2014; Gay Reference Gay2006; McClain et al. Reference McClain, Lyle, Carter, De Francesco Soto, Lackey, Cotton, Nunnally, Scotto, Grynasviski and Alan Kendrick2007; Telles et al. Reference Telles, Sawyer and Rivera-Salgado2011).

Qualitative scholars also find evidence that racial inequality reinforces ongoing racial tensions among Blacks and Whites, which can produce different sentiments and responses toward immigrant newcomers among the two host groups (Fennelly Reference Fennelly and Massey2008; Jiménez Reference Jiménez2016; Marrow Reference Marrow2011; McDermott Reference McDermott, Telles, Sawyer and Rivera-Salgado2011; Waters Reference Waters1999). Overall, this literature largely agrees with its quantitative counterpart that Whites often resist immigrants out of a sense of racialized group position and entitlement, whereas Blacks’ resistance to immigrants, when it arises, comes out of an “historical context of racial exclusion and exploitation” vis-à-vis Whites (Smith Reference Smith, Ansley and Shefner2009; see also Ribas Reference Ribas2016; Stuesse Reference Stuesse, Odem and Lacy2009). Thus, differences in status between U.S.-born Whites and Blacks, which are derived from past and present patterns of economic and political inequality, should be taken into account when interpreting attitudes among the U.S.-born toward immigrants (Brown and Jones Reference Brown and Jones2015; Jung Reference Jung2009; Kasinitz et al. Reference Kasinitz, Mollenkopf, Waters and Holdaway2008).

Receptivity toward immigrants among the U.S.-born is also likely to depend on the status positions that distinct immigrant groups typically occupy within the U.S. economic and racial hierarchy. More often than not, Mexican immigrants in the United States today are perceived as a “low economic status” group (Massey Reference Massey2007; Perlmann 2005; Bean et al. Reference Bean, Brown and Bachmeier2015; Telles and Ortiz Reference Telles and Ortiz2008), and American public opinion on immigration has focused “squarely on the plight of poor, unauthorized Mexican migrants” (Jiménez Reference Jiménez2010, p. 51). By contrast, South Asian Indians are viewed as a “high economic status” group, joining other East Asian immigrant groups who are often perceived as straddling the line between “model minorities” and “perpetual outsiders” in U.S. society (Dhingra Reference Dhingra2012; Mishra Reference Mishra2016; Nee and Holbrow Reference Nee and Holbrow2013; Tuan Reference Tuan1998; Zhou and Lee 2015; also see Chou and Feagin Reference Chou and Feagin2008; Kim Reference Kim1999; Xiu and Lee Reference Xu and Lee2013 on the racial triangulation of Asian Americans). Corresponding to common conceptions of these differential status positions, Indian immigrants are the most highly educated and one of the most professionally-employed of all immigrant groups in the United States today, whereas Mexicans are among the least educated and lowest-earning ones (Chakravorty et al. Reference Chakravorty, Kapur and Singh2017; Garip Reference Garip2017; NASEM Reference Waters and Gerstein Pineau2015; Portes and Rumbaut Reference Portes and Rumbaut2014). It is therefore reasonable to expect that Whites and Blacks, who occupy the traditional dominant and subordinate status positions in the U.S. racial hierarchy respectively, might likewise vary in receptivity toward such differentially-positioned immigrant groups.

RELATIONAL DYNAMICS BETWEEN THE U.S.-BORN AND IMMIGRANTS

While the extant scholarship on U.S. public opinion usefully clarifies how attitudes about immigration depend on the status positions different racial groups hold in society, it does so in a relatively abstract way. Still missing is greater insight into how relational intra-group dynamics between U.S.-born racial groups—how Whites and Blacks come into contact and interact with one another on a day-to-day basis—help shape each group’s receptivity toward immigrants. Our view is complementary to that of Ryan D. Enos (Reference Enos2017) who distinguishes between the perceptual level of how groups perceive and rank each other in space and status, and the experiential level of how groups interpret the contact they do have with one another within that larger context. Indeed, prior work in a range of settings has revealed multiple ways that contact between racial groups in daily interactions can shape and reshape people’s attitudes not only toward each other, but also toward other groups. Monica McDermott (Reference McDermott, Telles, Sawyer and Rivera-Salgado2011), for example, documents how strained interactions between upper-middle-class Blacks and Whites for control and visibility in cultural institutions in Greenville, South Carolina, help explain part of the former’s negative orientations toward Latino newcomers (also see Deeb-Sossa Reference Deeb-Sossa2013 for similar findings in a community health clinic setting in North Carolina). Conversely, Vanesa Ribas (Reference Ribas2016) demonstrates how workplace relations with White employers and supervisors whom Blacks may perceive as exploitative contribute to Blacks’ positioning of undocumented Latino coworkers as also (or even more) marginalized in a food processing plant employing low-skilled laborers. Among Whites, Helen B. Marrow (Reference Marrow2011), Roger Waldinger (Reference Waldinger1997), and Mary C. Waters (Reference Waters1999) all show that interpretations of strained or uncomfortable interactions with U.S.-born Blacks are connected to Whites’ explicit preferences for immigrants, both Latino and Black Caribbean, thereby upholding racial inequality.

As in studies of public opinion, many of these empirical studies of relational dynamics involving immigrants focus more strongly on tensions and hostility, than on cooperation or friendliness (Jones-Correa Reference Jones-Correa, Telles, Sawyer and Rivera-Salgado2011a; Lee 2001). This is somewhat surprising, as governmental and institutional policies can signal inclusion and receptivity, not merely exclusion and hostility, at a broader level (Bloemraad Reference Bloemraad2006; de Graauw Reference de Graauw2016; Fisher Williamson 2018; Huang and Liu Reference Huang and Liu2017; Huo et al. Reference Huo, Dovidio, Jiménez and Schildkraut2018; Williams Reference Williams2015). Emerging research is also increasingly granting attention to immigrants’ relational experiences of inclusion, and not just exclusion, with the U.S.-born within such institutions, which range from nonprofit organizations to law enforcement to healthcare clinics to social services offices (e.g., Calvo et al. Reference Calvo, Jablonska-Bayro and Waters2017; Gast and Okamoto 2014; Horton Reference Horton2004; Marrow Reference Marrow2011, Reference Marrow2012). Moreover, research in social psychology is also increasingly exploring ways in which members of dominant groups might be receptive to difference and willing to foster inclusion (Tropp and Mallett Reference Tropp and Mallett2011; Vollhardt et al. Reference Vollhardt, Migacheva, Tropp and de Rivera2009). But because few research studies have adopted this kind of relational lens to study the topic of immigration attitudes, we still have limited knowledge about how ongoing relations between racial groups in the host society may inform either inclusionary or exclusionary attitudes toward new immigrant arrivals. In the present research, we seek to pay greater attention to the influence of interracial contact between U.S.-born Whites and Blacks on both groups’ receptivity toward different immigrant groups residing in their metropolitan areas.

INTERGROUP CONTACT AND SECONDARY TRANSFER EFFECTS

In so doing, we draw on social psychological perspectives on intergroup contact, which, in our view, has not been fully conversant with the scholarship on immigrant incorporation, group position, and racial triangulation cited above. Traditionally, contact studies in psychology have examined the effects of direct contact, or how individuals’ face-to-face interactions with members of other racial and ethnic groups can produce more inclusive attitudes toward those groups (see Pettigrew and Tropp Reference Pettigrew and Tropp2011; Tropp and Page-Gould Reference Tropp, Page-Gould, Mikulincer, Shaver, Dovidio and Simpson2015 for reviews).^{Footnote 1} Yet new generations of contact research have also begun to focus on indirect contact effects—that is, how contact between members of two groups can have farther-reaching effects, beyond merely affecting each group’s attitudes toward the other. A growing body of work in this area focuses on what is known as the secondary transfer effect (STE). Whereas direct contact with a member of an outgroup shapes attitudes toward members of that same outgroup, a secondary transfer effect (STE) occurs when contact with a member of one outgroup shapes attitudes toward members of an outgroup not directly involved in the contact interaction (Pettigrew Reference Pettigrew2009).

While much of this research on secondary transfer effects, like our own, is cross-sectional, other longitudinal and experimental studies have now demonstrated that having individual-level contact with a member of one outgroup—whether ethnic, racial, or religious—can indeed have a range of consequences for one’s attitudes toward members of other, uninvolved groups (Hindriks et al. Reference Hindriks, Verkuyten and Coenders2014; Lolliot et al. Reference Lolliot, Schmid, Hewstone, Ramiah, Tausch, Swart, Hodson and Hewstone2013; Schmid et al. Reference Schmid, Hewstone, Küpper, Zick and Wagner2012; Schmid et al. Reference Schmid, Hewstone and Tausch2014; Shook et al. Reference Shook, Hopkins and Koech2016). Further, the longitudinal studies among them confirm that STEs cannot be fully explained by participant selection into contact with outgroups (see Tausch et al. Reference Tausch, Hewstone, Kenworthy, Psaltis, Schmid, Popan, Cairns and Hughes2010), which helps to allay one of the key criticisms leveled at cross-sectional analyses of contact effects (Enos Reference Enos2017; Powers and Ellison Reference Powers and Ellison1995). Thus, existing social psychological research bolsters confidence that secondary transfer effects can provide a useful framework with which to address the present study’s central question of how U.S.-born Whites’ and Blacks’ contact with one another might correspond with their attitudes toward new immigrants, even after controlling for Whites’ and Blacks’ direct contact with immigrants themselves.

Although most STE research has focused on arenas other than immigration, studies in Europe have begun to explore their effects in this area. For example, Paul Hindriks and colleagues (2014) show that when Turks and Moroccans in the Netherlands have more frequent contact with each other, they report less social distance toward Surinamese and Antilleans as well (see also Brylkaa et al. Reference Brylkaa, Jasinskaja-Lahti and Mähönen2016). Only a handful of studies have focused on secondary transfer effects related to immigration in the United States, and those that do typically examine STEs only with undergraduate student samples, or without differentiating by the economic status of immigrant outgroup to whom the STEs are theorized to extend, even as they have made strides in including respondents of different racial and ethnic backgrounds (e.g., Bowman and Griffin Reference Bowman and Griffin2012; Tausch et al. Reference Tausch, Hewstone, Kenworthy, Psaltis, Schmid, Popan, Cairns and Hughes2010; Van Laar et al. 2005).

Most relevant to the present research, Nicole Tausch and colleagues (2010, Study 3), asked 199 White and 76 Black U.S. college students to report their friendships with and attitudes toward Hispanics (the primary outgroup), and they were also randomly assigned to report their attitudes toward either Vietnamese or Indians as recent immigrant arrivals (two secondary outgroups). The authors observed significant STE effects, showing that students who reported greater numbers of Hispanic friends also reported more positive attitudes toward a secondary immigrant outgroup (Vietnamese or Indian), even when controlling for direct contact with the secondary outgroup (as measured by the numbers of Vietnamese or Indian immigrant friends they had). Moreover, comparable STE effects were observed among White and Black students in relation to Vietnamese and Indian immigrants, and even after controlling for the tendency to respond in socially desirable ways. However, we note that the Tausch et al. study took place in just one U.S. setting (Texas) drawing only on a college student sample. Further, it only investigated the relationship between Whites’ and Blacks’ contact with Hispanics and their attitudes toward two different groups of Asian immigrants, not also on how Whites’ and Blacks’ contact with each other might yield unique insights. On the whole, how STE dynamics might play out in field settings, with samples of adults and closely specified U.S.-born and immigrant groups, has yet to be explored.

SECONDARY TRANSFER EFFECTS IN RELATION TO GROUP POSITION

It is somewhat surprising that Tausch et al. (Reference Tausch, Hewstone, Kenworthy, Psaltis, Schmid, Popan, Cairns and Hughes2010) did not observe significant differences in STE effects among White and Black students, given other research in social psychology showing that contact effects are often stronger among members of dominant racial groups (e.g., Whites) and weaker among members of historically disadvantaged racial groups (e.g., Blacks) (Tropp and Pettigrew Reference Tropp and Pettigrew2005a; also see Binder et al. Reference Binder, Zagefka, Brown, Funke, Kessler, Mummendey, Maquil, Demoulin and Leyens2009). Here too, the results could be due to the authors’ use of a college student sample whose overall similarities may have masked possible differences associated with racial status. A larger study drawing on representative samples of respondents who vary in racial status and on a range of demographic characteristics would offer the opportunity to conduct a more robust analysis of STE effects among U.S.-born racial groups in relation to immigrants.

In sum, the existing literature on secondary transfer effects provides a useful foundation for extending prior research in sociology and political science that examines U.S.-born groups’ attitudes toward immigrants in other ways. This approach offers a novel way for immigration and race researchers to model and analyze how Whites’ and Blacks’ contact experiences with each other, in addition to each’s own demographic characteristics or perceptions of threat from immigrants, may shape their attitudes toward immigrants. While studies of secondary transfer effects have yet to be integrated into mainstream research on immigrant incorporation, this approach easily complements other novel work in sociology and political science, which has begun to explore how the familial, friendship, coworker, neighbor, and/or acquaintance contact networks Whites and Blacks directly have with immigrants predicts the former’s immigration attitudes, via primary transfer effects (e.g., Berg Reference Berg2009; Pearson-Merkowitz et al. Reference Pearson-Merkowitz, Filindra and Dyck2016; Wilkinson Reference Wilkinson2015). Merging the approaches in these literatures will enhance our understanding of how interracial contact among the U.S.-born helps shape White and Black attitudes toward immigrants in the contemporary multi-ethnic context of the United States.

HYPOTHESES

This study’s central question, then, is how contact between U.S.-born Whites and Blacks is meaningfully associated with each group’s receptivity toward immigrant newcomers. In line with the extensive literature on intergroup contact, we generally expect that greater contact between Whites and Blacks should correspond with more positive intergroup attitudes among both groups toward each other. Moreover, drawing upon the extant literature on secondary transfer effects, we expect that greater contact between Whites and Blacks should also correspond with greater receptivity toward immigrant groups, even after controlling for Whites’ and Blacks’ direct contact experiences with the immigrant groups themselves [H1].

At the same time, given that contact between U.S.-born Whites and Blacks still occurs within a context of high racial and economic segregation and inequality, we expect that STE processes will likely function differently for members of the two U.S.-born groups. As a numerical minority group in the United States, Blacks tend to have greater contact with Whites than vice versa (Butler and Wilson Reference Butler and Wilson1978; Sigelman and Welch Reference Sigelman and Welch1993). Additionally, as an historically disadvantaged group, Blacks are also more likely to report having negative contact experiences with Whites, as compared to the contact experiences that Whites report having with Blacks (Pew Research Center 2016; Stephan et al. Reference Stephan, Boniecki, Ybarra, Bettencourt, Ervin, Jackson, McNatt and Lausanne Renfro2002), and such negative intergroup experiences have the potential to hinder the positive effects of contact on Blacks’ racial attitudes toward other groups (Tropp Reference Tropp2007). These patterns may help to explain why contact effects are often less pronounced among subordinate racial groups relative to the effects typically observed among dominant racial groups (Binder et al. Reference Binder, Zagefka, Brown, Funke, Kessler, Mummendey, Maquil, Demoulin and Leyens2009; Tropp and Pettigrew Reference Tropp and Pettigrew2005a). Correspondingly, we expect that secondary transfer effects growing from group members’ contact experiences will be less pronounced among U.S.-born Blacks than among U.S.-born Whites [H2].

Though existing contact research provides less guidance about how contact between U.S.-born Blacks and Whites might shape receptivity toward immigrants who are differentially positioned in American society, there is some reason to believe that such secondary transfer effects may correspond with where immigrant groups are perceived to fall within existing U.S. racial and economic hierarchies. For instance, some scholars have proposed that secondary transfer effects should be particularly likely when the uninvolved outgroup is somehow perceived to be similar to the outgroup with whom direct contact is taking place (see Harwood et al. Reference Harwood, Paolini, Joyce, Rubin and Arroyo2011; Pettigrew Reference Pettigrew2009; Tausch et al. Reference Tausch, Hewstone, Kenworthy, Psaltis, Schmid, Popan, Cairns and Hughes2010). Following a similar line of reasoning, we might predict that contact between U.S.-born Whites and Blacks will be especially likely to transfer out toward immigrant groups when there is some correspondence between the racial and economic positioning of an immigrant outgroup and the U.S.-born outgroup with whom contact is taking place. More specifically, then, we expect that among U.S.-born Whites, greater contact with U.S.-born Blacks should be more likely to transfer out to attitudes toward immigrant outgroups that are perceived to be lower in socioeconomic status and positioned as nearer to Blacks (e.g., Mexican immigrants; see Masuoka and Junn Reference Masuoka and Junn2013; Portes and Rumbaut Reference Portes and Rumbaut2014) [H3a]; by contrast, we expect that among U.S.-born Blacks, greater contact with U.S.-born Whites should be more likely to transfer out to attitudes toward immigrant outgroups that are perceived to be higher in socioeconomic status and positioned as nearer to Whites (e.g., Indian immigrants; also see Bonilla Silva 1997; Lee and Fiske Reference Lee and Fiske2006; Lee and Zhou Reference Lee and Zhou2015; Nee and Holbrow Reference Nee and Holbrow2013) [H3b].

DATA AND METHODS

To test these hypotheses, we draw on the Study of Immigrants and Non-Immigrants in Atlanta and Philadelphia (SINAP), an original representative survey conducted in 2013 that includes responses from 503 U.S-born Whites and 502 U.S.-born Blacks living in metropolitan Philadelphia and Atlanta. We selected these locations and groups for both theoretical and demographic reasons. Since Black-White relations have figured prominently in the literature on intergroup contact (Pettigrew and Tropp Reference Pettigrew and Tropp2011), we sought to carry out our research in places with a significant presence and history of Black-White contact and relations, as well as large contemporary immigrant populations. Using population estimates from the 2008 and 2010 American Community Survey (ACS), we selected the Philadelphia and Atlanta metropolitan areas as our two research sites because they are (a) comparable to each other in population size; (b) both have long-standing histories of White-Black relations that have profoundly shaped their regional politics and social interactions, and that now serve as context into which new immigrants from around the world arrive; and (c) both have over 50,000 immigrant arrivals who hail from Mexico and India, the top two sending countries to the United States (Chakravorty et al. Reference Chakravorty, Kapur and Singh2017; Portes and Rumbaut Reference Portes and Rumbaut2014; Zong and Batalova Reference Zong and Batalova2016, Reference Zong and Batalova2017).^{Footnote 2}

In addition, we sampled U.S.-born Whites and Blacks because they represent key populations in these metropolitan areas and vary on racial and socioeconomic dimensions that typically grant them higher and lower status, respectively, in U.S. society. Relatedly, we asked U.S.-born Whites and Blacks about their contact with and attitudes toward Mexican and Indian immigrants because these groups vary on status markers that exemplify the bifurcation in skill levels among post-1965 immigrants to the United States, which lead them to be perceived, respectively, as quintessential “low economic status” and “high economic status” immigrant groups (Portes and Rumbaut Reference Portes and Rumbaut2014).^{Footnote 3} On average, foreign-born Mexicans have low levels of education and higher levels of employment in lower-skilled sectors of the economy; they also register low levels of English language proficiency and, following decades of intensifying border and interior immigration enforcement, high levels of undocumented status (Bean et al. Reference Bean, Brown and Bachmeier2015; Garip Reference Garip2017; NASEM Reference Waters and Gerstein Pineau2015; Telles and Ortiz Reference Telles and Ortiz2008). In contrast, foreign-born Indians are among the most highly educated and residentially dispersed immigrant groups; they are often employed in higher-skilled sectors of the economy, have considerable fluency in English, and register low rates of undocumented status, despite exhibiting strong internal variation (Chakravorty et al. Reference Chakravorty, Kapur and Singh2017; Leonard Reference Leonard, Waters and Ueda2007; Mishra Reference Mishra2016; Portes and Rumbaut Reference Portes and Rumbaut2014; Sandhu Reference Sandhu2012).^{Footnote 4} Such pointed differences in the two groups’ socioeconomic positioning also correspond to different host reactions. Americans today are more likely to perceive Mexicans to be poorly educated, to worry they may never assimilate, and to racialize them as illegal or even inhuman (Chavez Reference Chavez2008; Huntington Reference Huntington2004; Jiménez Reference Jiménez2010; Lee and Fiske Reference Lee and Fiske2006), whereas Americans often view Indians as well educated, smart, and talented (Lee and Zhou Reference Lee and Zhou2015), even as Americans may feel some competition with Indian immigrants (Jiménez and Horowitz Reference Jiménez and Horowitz2013; Lee and Fiske Reference Lee and Fiske2006; Samson Reference Samson2013) or see them as non-White. Indeed, Asian immigrants are viewed more positively than Latin American immigrants in public opinion (Pew Research Center 2015), which may have implications for their contact experiences with the U.S.-born in everyday life.

Overall, the SINAP dataset is uniquely suited to address the present research goals as it includes assessments of intergroup contact experiences and intergroup attitudes among members of both U.S.-born groups who were representatively sampled from a general population in two different metropolitan areas, and who completed surveys in relation to two immigrant groups who are differentially positioned in terms of social and economic status. (For more detail about the full samples and procedures included in the SINAP dataset, which also queried foreign-born Mexican and Indian immigrants, see Appendix A.)

DISCUSSION AND CONCLUSION

How do we understand patterns of contact between the U.S.-born and immigrants within twenty-first century America, where these groups do not interact with one another in an isolated, binary fashion, but rather interact and respond to each other in multi-ethnic settings, which are shaped by the weight of a long and unequal historical past? Little research has endeavored to model the effects of such contact. We move beyond simply examining Blacks’ and Whites’ attitudes toward immigrants—a key element of the “context of reception” that influences immigrant incorporation (Portes and Rumbaut Reference Portes and Rumbaut2014)—to examine how these attitudes are influenced by Whites’ and Blacks’ direct contact experiences with each other, and with immigrants themselves. Using original representative survey data from U.S.-born Whites and Blacks living in metropolitan Atlanta and Philadelphia, we focus on how Blacks’ and Whites’ interracial contact with one another, even controlling for their direct contact with immigrants, contributes to their receptivity toward immigrants. This approach usefully adds to existing research on STEs by (1) focusing on these effects within the highly racially stratified U.S. context, (2) drawing on representative samples of adults across two metropolitan areas, (3) attending carefully to group status in relations between U.S.-born and immigrant groups, and (4) highlighting ways in which contact relations can foster inclusionary and positive attitudes toward immigrants, not merely those that are exclusionary and hostile.^{Footnote 11} Consistently for Whites, and in some cases for Blacks, we find that more frequent contact with the other U.S.-born group meaningfully contributes to greater receptivity toward immigrants. Such associations remain significant when controlling for the effects of both groups’ direct contact experiences with immigrants themselves, and when taking into account other possible explanatory or confounding factors.

The analyses presented here confirm the importance of secondary transfer effects in shaping attitudes toward immigrants, beyond the effects stemming from direct contact between U.S.-born and foreign-born individuals [H1].^{Footnote 12} The findings also confirm that secondary transfer effects are less pronounced among U.S.-born Blacks than among U.S.-born Whites [H2]. However, our results did not support our third hypothesis—that Whites’ contact with Blacks might be more closely linked to receptivity toward lower-status immigrant groups, whereas Blacks’ contact with Whites might be more closely linked to receptivity toward higher-status immigrant groups. Instead, Whites’ contact with Blacks corresponded with greater receptivity toward both immigrant groups, while Blacks’ contact with Whites only corresponded with greater welcoming toward a lower-status immigrant group (Mexicans) than toward a higher-status immigrant group (Indians).

Cross-sectional data alone cannot fully explain the processes or mechanisms driving these patterns of results. However, they may help to rule out a few alternatives. First, it bears re-emphasizing that across all our analyses the STEs for contact frequency are uniformly positive; when and where STEs emerge in our data, more frequent contact with the other U.S.-born group is always associated with greater (not less) receptivity toward immigrants. As discussed above, this shows that the STEs uncovered here are not driven by a negative process of Whites engaging in social comparison between Blacks and the two immigrant groups.^{Footnote 13}

Second, the present analysis also shows that the STEs uncovered here are not driven by large disparities in the reported quality of Whites’ or Blacks’ contact with each other or in their attitudes toward the two immigrant groups. We showed earlier that, overall, U.S.-born Whites and Blacks do not differ significantly in terms of how much contact they report having with each other or with Mexicans or Indians. Nor do they differ significantly in terms of how much they welcome or desire to know Mexicans or Indians. Thus, we cannot argue that Whites’ attitudes somehow have greater room to shift by virtue of Whites having more contact with immigrants than Blacks do, or because Blacks are generally more receptive to immigrants than Whites are.

Finally, the present analysis suggests that the STEs indicated by the models are not driven by the differential economic positioning of the two immigrant outgroups. Instead, we argue that the shared racial positioning of the two immigrant outgroups likely provides a better account for why we find roughly equal evidence of STEs toward Mexicans and Indians among U.S.-born Whites, and perhaps also greater evidence of STEs among Whites than among Blacks. Perhaps U.S.-born Whites, given their dominant status in the American racial hierarchy, perceive both Mexican and Indian immigrants collectively as “non-Whites,” which could help to explain why Whites’ more frequent contact with Blacks contributes to greater White receptivity toward both groups of immigrants, regardless of the latter’s very different economic positioning. Some U.S.-born Blacks, given their subordinate status in the U.S. racial hierarchy, may instead perceive greater affinity with various racialized immigrant-origin groups, especially those who are strongly vilified, as Mexicans are in the present moment (e.g., Williams and Hannon Reference Williams and Hannon2016). Correspondingly, Blacks’ attitudes toward immigrants may be driven more by the salience of long-standing racial divides (see Bobo and Hutchings Reference Bobo and Hutchings1996; Wilkinson Reference Wilkinson2015) than by perceived similarities in economic status between Whites and higher-status immigrants, or between Blacks and lower-status immigrants. Of course, it is also possible that Blacks’ more frequent contact with Whites could intensify feelings of competition with and resentment of immigrants, and reduce Blacks’ receptivity toward all immigrant groups. However, this possibility does not appear to be supported by our analyses; the mean scores for welcoming and interest in knowing suggest neutral to positive attitudes toward immigrants, and the coefficients we uncover in modeling for STEs suggest only positive associations between U.S.-born Blacks’ contact experiences and receptivity toward immigrants, even after controlling for their perceptions of economic threat.

Thus, in our view the present findings uphold the central role of contact frequency in secondary transfer effects, even when controlling for contact quality and economic threat. ^{Footnote 14} But they also highlight a need for more research to better understand the range of plausible processes and mechanisms that might underlie STEs, especially as they may operate in relation to different host and immigrant outgroups. Using the SINAP dataset, we are not able to test directly the role that perceived similarity between outgroups may play in facilitating STEs (see Pettigrew Reference Pettigrew2009; Tausch et al. Reference Tausch, Hewstone, Kenworthy, Psaltis, Schmid, Popan, Cairns and Hughes2010), so one fruitful step would be to include more direct measures of Whites’ and Blacks’ perceptions of similarity (economic, racial, and cultural) between various immigrant outgroups. This would allow for more direct examination of the dimensions of intergroup similarity that are deemed most predictive of STEs in contexts where recent immigrant streams introduce new dimensions of intergroup differences, while aligning this work more closely with existing STE studies on intergroup contact. A second fruitful approach, since experimental contact studies already provide convincing evidence that more frequent contact leads to improved attitudinal outcomes rather than vice versa (e.g., Binder et al. Reference Binder, Zagefka, Brown, Funke, Kessler, Mummendey, Maquil, Demoulin and Leyens2009; Tausch et al. Reference Tausch, Hewstone, Kenworthy, Psaltis, Schmid, Popan, Cairns and Hughes2010; see also Pettigrew and Tropp Reference Pettigrew and Tropp2011), would be to collect longitudinal data on STEs. This would allow for the testing of multiple mediating mechanisms at once, over several points in time. Finally, a third fruitful approach would be to test for secondary transfer effects in other locations (either within the United States or abroad), using other host and/or immigrant groups, or using other status characteristics that shape group positioning, to see if the conclusions presented here are generalizable to other settings and groups. For instance, one could imagine applying our model to other contexts with multiple, highly-stratified host groups, such as Cataluña, Belgium, Australia, South Africa, Canada, or beyond. Given the rise of prejudice toward Muslim migrant communities in Europe and increasingly in the United States (Bail Reference Bail2015; Bozorgmehr and Bakalian Reference Bozorgmehr and Bakalian2009; Cainkar Reference Cainkar2009; Selod Reference Selod2018; Zolberg and Long 1999), it could be especially fruitful to incorporate an analysis of other status characteristics, such as religion or language, alongside race and economic status to disentangle which ones in particular are most important for observed STEs.^{Footnote 15}

The findings of the research here, however, provide an essential starting point. They represent a novel extension of STE research including new groups in new contexts, as well as a novel linking of that literature with immigration and race/ethnicity scholarship. Engagement, for example, with immigrant incorporation, group position, or even racial triangulation theories (see Kim Reference Kim1999; Xu and Lee Reference Xu and Lee2013) is not typical in STE research, yet the present study’s findings provide theoretical support for a common argument that runs across all three of those literatures, which is that the U.S. racial hierarchy shapes new immigrants’ reception and incorporation. Further, our results suggest the utility of integrating relational measures of contact and attending to economic as well as racial dimensions of stratification into such an interdisciplinary endeavor. Our findings suggest that Whites may be forming attitudes toward immigrants based largely on perceptions of the latter’s non-White racial—and less so economic—status, yet Whites and Blacks are not interpreting their encounters with one another in negative ways. Rather, we observe positive connections between interracial contact and immigration attitudes among the U.S.-born, and we do so in two metropolitan areas that were purposefully selected for their long histories of White-Black inequality and their high levels of ongoing White-Black segregation (Logan and Stults Reference Logan and Stults2011), which should create a more conservative test of how interracial contact, if and when it occurs, might translate into other positive outcomes.