Health Organizations: Theory, Behavior, and Development

Access to the Virginia Social Science Journal is granted under creative commons license CC BY-NC-ND 4.0. You may share this non-modified material for non-commer-cial purposes, but you must give appropriate attribution. Contact the author of this work for other uses. Suggested citation may be found on the last page of each article.

Ingroup Bias, Trust in Racial Groups, and Generalized Trust among U.S. Blacks and Whites

Jeffrey G. ToussainT & K. Jill KiecolT

Virginia Social Science Journal | Volume 54 | 2020 | Pages 40 —55

ABSTRACT We build on a conflict theory of trust to investigate how ingroup bias and trust in ethnoracial groups are related to generalized trust. The theory may help explain Blacks’ lower generalized trust, which is sometimes attributed to low trust of Whites. Combined data from the 2000 and 2004 American National Election Studies (N = 2,417) were analyzed. As predict- ed, ingroup bias undermined generalized trust, and ingroup trust boosted it. Unexpectedly, however, Blacks trusted Blacks and Whites about equally. Overall, trust in ethnoracial outgroups mattered more for generalized trust for Whites. The find- ings challenge the assumption that Blacks’ lower generalized trust stems from low trust of Whites. They also indicate that a conflict theory of trust better explains trust for Whites.

keywords: ingroup, outgroup, bias, trust, ethnoracial groups

AUTHORS Jeffrey G. Toussaint Virginia Wesleyan University & K. Jill Kiecolt Virginia Tech

INTRODUCTION

Generalized trust—the perception of how much “most people” can be trusted—is a central foundation of society (Ross, Mirowsky, and Pribesh 2001; Uslaner 2016). Trust implies the belief that other people are supportive and can be depended on (Hardin 2002). It enables people to establish mutually beneficial, cooperative relationships with others (Ross et al. 2001). Greater generalized trust engenders a greater willingness to cooperate with others and to engage in civic activities, even with people one does not know personally (Stolle 2002). By promoting faith in others, trust helps people and nations address societal problems (Uslaner 2016).

A prominent theory of trust, the conflict theory of trust, proposes that ethnoracial diversity undermines gener- alized trust (Putnam 2007). Evidence for that part of the theory is fairly weak and inconsistent (Abascal and Bal- dassarri 2015; Putnam 2007; Uslaner 2010; van der Meer and Tolsma 2014). However, other key parts of the theory, which involve how people think about their own and other racial groups, have not been tested. The theory pro- poses that ethnocentrism, that is, ingroup bias, increases with ethnoracial diversity. Ingroup bias fosters trust in one’s own ethnoracial group, but diminishes trust in oth- er ethnoracial groups, as well as generalized trust (Brewer 1981; Putnam 2007; Stolle 2002). We take ingroup bias as our starting point to test these parts of conflict theory.

We expand the theory to propose that trust in one’s own racial group and trust in other racial groups also will be related to greater generalized trust (Delhey, New- ton, and Welzel 2011; Stolle 2002). However, we do not know whether that is the case. The few studies that have examined trust in one’s own racial group and trust in other racial groups modeled them as separate outcomes (Abascal and Baldassarri 2015; Putnam 2007; Schmid, Al Ramiah, and Hewstone 2014) and did not investigate how they were related to generalized trust.

Examining trust of racial ingroups and outgroups ad- dresses a longstanding question of whom respondents have in mind when asked about trust in “most people.” Some assume that both Blacks and Whites are thinking mostly of Whites (Simpson, McGrimmon, and Irwin 2007) and that Blacks’ lower trust in people reflects low trust of Whites (Smith 1997). If so, Blacks’ trust in Whites and per- haps other racial outgroups should help explain the racial gap in trust. This study investigates how ingroup bias, trust in one’s own racial group, and trust in other races are related to each other and to generalized trust among Blacks and Whites. Data are from two national samples of U.S. adults.

BACKGROUND

We expand on the conflict theory of trust to investigate potential sources of differences in trust between Blacks and Whites. The theory involves two forms of trust,

Vol. 54 | 2020 | Virginia Social Science Journal Toussaint & Kiecolt | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | 41

generalized and identity-based trust. Generalized trust implies the belief that other people are supportive and will usually act in accord with one’s interests (Hardin 2002). Generalized trust in people helps sustain the social fabric (Putnam 2007; Ross et al. 2001), because it enables people to establish mutually beneficial, cooperative relationships with others (Ross et al. 2001). Trust helps in resolving issues and promotes cohesion in society. Moreover, it enhances economic performance (Knack and Keefer 1997) and promotes civic and political participa- tion (Delhey et al. 2011; Putnam 2007; Stolle 2002).

Identity-based trust, in contrast, is a more narrow form of trust. It refers to trust in people who belong to a given so- cial category (Freitag and Bauer 2013; Stolle 2002). In our expanded conflict theory of trust, identity-based trust re- fers specifically to trust in racial ingroups and outgroups. We discuss identity-based trust in more detail below.

Figure 1 shows the elements of the expanded conflict theory of trust that we test in our analysis: race/ethnicity, ingroup bias, trust in one’s racial ingroup, trust in racial outgroups, and generalized trust.1 In the sections below, we describe the connections among the elements.

Figure 1 shows that race/ethnicity (Black versus White) is predicted to be negatively associated with generalized trust. As noted above, numerous studies have established that Blacks have lower generalized trust than Whites, even after controlling for indicators of socioeconomic status (Abascal and Baldassarri 2015; Coverdill, Lopez, and Petrie 2011; Smith 2010; Uslaner 2010; Wilkes 2011). Numerous factors contribute to Blacks’ lower trust. Blacks experience discrimination across multiple institu- tions—the labor market, housing, and the criminal justice system. Structural inequality and residential segregation contribute to poorer neighborhood quality for Blacks, which fosters mistrust. Parents socialize their children to prepare to deal with the discrimination they will face from outgroups (Ross et al. 2001; Smith 2010). Blacks are expected to score lower than Whites on all the other vari- ables in the model as well (discussed further below).

Ingroup Bias and Trust Our theoretical model of generalized trust centers on orientations toward ethnoracial groups, which are likely to influence trust and to explain the racial gap. It has long been recognized that groups are “psychologically real” entities that people respond positively or negatively to, and that affect their behavior (Campbell et al. 1960:296; emphasis in original). According to social identity the- ory, people want to maintain positive group identities

that distinguish their group from other groups (Tajfel and Turner 1986). They do so by evaluating their ingroup favorably, but also more favorably than an outgroup. This sense of ingroup superiority to outgroups is termed ingroup bias (Brewer 1981).

Ingroup bias involves ingroup–outgroup comparisons, not merely favorable evaluations of one’s ingroup or unfa- vorable evaluations of outgroups (Bizumic and Duckitt 2012; Brewer 1981; Putnam 2007). Instead, ingroup bias is relative and variable (Brewer 1981). It ranges from evalu- ating one’s ingroup much more favorably than outgroups to (potentially) evaluating outgroups much more favor- ably than one’s ingroup.

However, because of persisting racial inequality, ingroup bias has different meanings for Blacks and Whites. For Whites, ingroup bias reflects cultural racism, ideas and images that convey White superiority (Williams and Mo- hammed 2013). For Blacks, ingroup bias implies resis- tance to cultural racism. We expect ingroup bias to be higher among Whites than Blacks. Blacks, like Whites, view their group very favor- ably; they endorse positive stereotypes of their group (e.g., being hardworking and intelligent) and reject nega- tive ones (Kiecolt and Hughes 2017). Nevertheless, mem- bers of higher-status groups tend to show more ingroup bias than members of lower-status groups (Hewstone, Rubin, and Willis 2002). Blacks rate their ingroup less pos- itively than Whites do (Bobo and Massagli 2001; Kiecolt and Hughes 2017), and they rate ethnoracial outgroups more favorably than Whites do (Bobo and Massagli 2001; Wodtke 2012). In a study that measured ingroup bias (dif- ferential racial ingroup and outgroup evaluations), Blacks showed little bias, whereas Whites showed ingroup bias vis-a-vis ethnoracial outgroups (Kiecolt and Hughes 2017).

As Figure 1 shows, ingroup bias should be related to greater trust in one’s own racial group (Brewer 1981; Put- nam 2007). When people shift to a collective identity as a member of a social category, they focus on their simi- larities to other group members (Tajfel and Turner 1986). As a result, they see less risk to trusting other members (Brewer 1981; Brewer and Yuki 2007). Consequently, peo- ple with higher ingroup bias should trust their ingroup more (Brewer 1999).

Ingroup bias is presumed to be associated with more negative attitudes toward outgroups. If so, it would be related to lower trust in racial outgroups and lower gener- alized trust. Research on ingroup bias and trust is sparse,

42 | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | Toussaint & Kiecolt Virginia Social Science Journal | Vol. 54 | 2020

but in one study, perceived superiority to racial outgroups was related to lower generalized trust among Blacks and Whites, especially if they felt closer to their group (Kiecolt and Hughes 2017). Based on the theory and research reviewed above, we predict that greater ingroup bias will be related to higher racial ingroup trust, lower racial outgroup trust, and lower generalized trust.

Trust in Racial Groups and Generalized Trust

In the U.S., race remains a primary source of social divi- sion (Bonilla-Silva 2018). For that reason, trust in one’s racial ingroup and racial outgroups are apt to be especial- ly salient forms of identity-based trust. As noted above, identity-based trust refers to trust in members of social categories or groups, only some of whom one knows (Brewer 1981; Stolle 2002). Identity-based trust can refer to trust in ingroup members, with whom one shares a so- cial identity, such as nationality, race, ethnicity, religion, and so forth (Stolle 2002). It also encompasses trust in members of outgroups (Freitag and Bauer 2013).

Like generalized trust, identity-based trust develops through people’s everyday experiences (Glanville and Paxton 2007; Ross et al. 2001). People interact with each other as members of social categories, such as ethnora- cial groups, even as they perform their social roles (Deaux and Martin 2003). People are likely to generalize from their positive (or negative) interactions with ingroup and outgroup members they know to other members of the category. Positive bonding interactions with ingroup members—based on strong ties (e.g., ingroup friend- ships) or weak ties (e.g., homogeneous associations) should promote ingroup trust (Stolle 2002). Similarly, positive bridging interactions with outgroup members— based on strong ties (e.g., interracial friendships) or weak ties (e.g., diverse associations) should promote outgroup trust (Stolle 2002).

Because people cannot interact with all ingroup or out- group members, making inferences about the trustwor- thiness of a group as a whole is challenging (Kramer et al. 2001). Nevertheless, people expect more positive behav- ior from ingroup than outgroup members, so they tend to trust them more as well (Kramer et al. 2001; Simpson et al. 2007; Smith 2010). Consequently, we expect that people will trust their racial ingroup more than racial outgroups. Previous cross-national research has shown that trust in outgroups (people of other nationalities and religions), as well as trust in people one knows personally, are posi- tively related to generalized trust (Delhey et al. 2011).2

In the U.S., generalized trust depends more on outgroup trust than on people one knows personally. However, we do not know how trust in one’s own racial group and trust in ethnoracial outgroups contribute to generalized trust. Based on Delhey et al.’s (2011) findings, however, we predict that both trust in one’s own racial group and trust in ethnoracial outgroups will be related to greater generalized trust.3 Interactions are the mechanism by which these occur. Both positive bonding and bridging interactions contrib- ute to generalized trust (Glanville and Paxton 2007; Stolle 2002). However, positive bridging interactions with mem- bers of outgroups should especially encourage general- ized trust (Marschall and Stolle 2004; Stolle 2002). In this regard, involvement in multiple voluntary associations fosters greater generalized trust by creating more diverse social networks (Glanville 2016). Interracial friendships also can encourage more positive views of outgroups (Pettigrew 1997).

Including both ingroup and outgroup trust as predictors of generalized trust enables us to address the longstand- ing question of whom respondents have in mind when asked about trust in “most people.” Some have specu- lated that both Blacks and Whites are thinking mostly of Whites (Marschall and Stolle 2004; Simpson et al. 2007; Smith 2010) and attributed Blacks’ lower trust in people to mistrust of Whites (Smith 1997). The more that gen- eralized trust depends on outgroup trust, the wider is a group’s radius of trust. If generalized trust for both Blacks and Whites depends more on trust of Whites than Blacks, then Blacks have a wider radius of trust (Delhey et al. 2011; Marschall and Stolle 2004; Smith 2010). Similarly, outgroup trust may be more strongly associated with generalized trust for Blacks than Whites.

We predict that Blacks will trust racial outgroups less than Whites do, based on a previous finding that used a composite measure of ethnoracial outgroup trust (Abas- cal and Baldassarri 2015). If so, outgroup trust will help explain the racial difference in generalized trust. Rath- er than using a composite measure of outgroup trust, though, we examine how trust in specific ethnoracial outgroups contributes to generalized trust. In our multi- racial society, Blacks’ and Whites’ perceptions of ethnora- cial outgroups may differ (Bobo and Hutchings 1996), and these perceptions may affect trust differently.

We expect to find lower ingroup trust among Blacks than Whites as well. Previous research has found that Blacks trust members of their racial ingroup less than Whites do, even after controlling for socioeconomic status (Abascal and Baldassarri 2015). Trust in one’s racial ingroup is pre-

Vol. 54 | 2020 | Virginia Social Science Journal Toussaint & Kiecolt | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | 43

dicted to contribute to greater generalized trust, so lower ingroup trust among Blacks also would help explain their lower generalized trust.

We also explore how trust in one’s own and other racial groups are related to each other. Putnam (2007) suggests that ingroup trust is not necessarily negatively correlated with outgroup trust. He reasons that people with more bonding capital (ties to similar others) are likely to have more bridging capital (ties to dissimilar others) (Putnam 2007). If so, strong ingroup ties and ingroup trust should be related to greater trust in racial outgroups. Never- theless, he leaves open the possibility that ingroup and outgroup trust may be positively related, negatively re- lated, or unrelated to each other (Putnam 2007). Because evidence is lacking, we make no prediction.

METHODS

Data

We combined data from two cross-sectional surveys, the 2000 and 2004 American National Election Studies Time Series (ANES; Burns et al. 2016; University of Michigan 2016), administered by the Survey Research Center at the University of Michigan. These were the only two years that the ANES asked about trust in racial groups; no more recent data are available. In both years, the weighted samples were representative of noninstitutionalized, En- glish-speaking adults 18 years of age or older in the Unit- ed States. In 2000, respondents were interviewed either face-to-face or by telephone; in 2004 respondents were interviewed face-to-face. In each year, interviews were conducted pre- and post-election. In 2000, response rates for the pre-election survey were 61.2% for face-to-face interviews and 57.0% for telephone interviews. For the post-election survey, response rates were 86.0 for face- to-face interviews and 85.8% for telephone interviews. In 2004, response rates were 66.1 for the pre-election survey and 88.0 for the post-election survey. Only respondents who were interviewed both pre- and post-election in either 2000 or 2004, were included in the analysis. The combined analytic sample consisted of 2,080 White and 337 African American respondents (N = 2,417). Weights provided by the Survey Research Center for each year ad- justed for sampling, nonresponse, and poststratification factors. We controlled for survey year in the multivariate analyses. Measures

Generalized trust. A scale of generalized trust used in previous research (Coverdill et al. 2011; Hughes and

Thomas 1998; Wilkes 2011) summed scores on three items. Respondents were asked whether most people can be trusted or one can’t be too careful in dealing with peo- ple, whether most people take advantage of others or try to be fair, and whether people try to be helpful or mostly look out for themselves. We coded trusting responses as 1, mistrustful responses as 0. The scale ranged from 0−3. Cronbach’s alpha was 0.78.

Trust in racial ingroup and racial outgroups. The mea- sures of trust in racial groups were based on ratings of trustworthiness as a general group characteristic of African Americans, Asian-Americans, Hispanic-Americans, and Whites, from 1–7. Seven meant that almost all the people in the group tended to be trustworthy, one meant that most people in the group were untrustworthy, and four meant that the group was no closer to one end or the other. Ingroup trust was measured as Blacks’ rating of African Americans or Whites’ rating of Whites. Outgroup trust in Blacks or Whites was measured as Blacks’ rating of Whites or Whites’ rating of African Americans. Outgroup trust in Hispanics was measured as respondents’ rating of Hispanics. Outgroup trust in Asians was measured as respondents’ rating of Asians.4

Ingroup bias. Ingroup bias was measured by differential racial ingroup–outgroup ratings on two sets of items that measure racial stereotypes (Bobo and Johnson 2000; Kiecolt and Hughes 2017). Respondents rated Blacks, Hispanics, Asians, and Whites on (1) how unintelligent/ intelligent they thought most group members were, from 1 = unintelligent to 7 = intelligent, and (2) how lazy/hard- working they thought most group members were, from 1 = lazy to 7 = hardworking. Racial ingroup ratings averaged Blacks’ and Whites’ ratings of their own group on the two items. Outgroup ratings averaged ratings of each of the other three groups (Blacks or Whites, Hispanics, and Asians) on the two items. We also created a composite measure that averaged ratings of the two racial stereo- types (unintelligent/intelligent and lazy/hardworking) for all three outgroups.

Each respondent’s mean ingroup rating minus their mean outgroup rating for a specific outgroup yielded a difference score that ranged from –6 to +6. For example, a Black respondent who rated their ingroup as 5 and Hispanics as 3 would score +2 on ingroup bias regarding Hispanics. A White respondent who rated their ingroup as 1 and Asians as 3 would score –2, indicating outgroup bias toward Asians. Sociodemographic characteristics. The analyses con- trolled for age, gender, education, marital status, region,

44 | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | Toussaint & Kiecolt Virginia Social Science Journal | Vol. 54 | 2020

and survey year. Age and education are positively related to trust (Wilkes 2011). Women and married people tend to have higher trust than men and unmarried people, respectively (Smith 1997; Wilkes 2011). Trust in people is lower in the South than in other regions of the U.S. (Put- nam 2007; Smith 1997). Generalized trust has declined in recent decades (Wilkes 2011). Dummy variables were race (0 = White, 1 = Black), gen- der (0 = male, 1 = female), marital status (not married or cohabiting = 0, married or cohabiting = 1), region of residence (0 = non-South, 1 = South) and survey year (0 = 2000, 1 = 2004). Age was measured in years. Education was measured as 1 = grade 8 or less, 2 = grades 9–11, 3 = high school diploma, 4 = more than 12 years of school- ing, 5 = Associate of Arts degree, 6 = Bachelor’s degree, 7 = postgraduate or professional degree. All sociodemo- graphic characteristics except race were mean-centered in the regression analyses. Data Analysis

We computed descriptive statistics on the study vari- ables for the total sample, Blacks, and Whites. We also performed paired t-tests to examine differences between ingroup and outgroup trust. We then computed correla- tions between trust in one’s racial ingroup and trust in the three racial outgroups for the total sample and for Blacks and Whites separately. Next, we regressed ingroup trust and trust in the three outgroups on race and the control variables, then added ingroup bias to the models. Finally, we performed ordinal regressions of generalized trust in people, separately for each outgroup. In each analysis, Model 1 regressed trust in people on race, ingroup bias, and the control variables. Model 2 added ingroup trust and outgroup trust. We also tested for interactions of race and outgroup trust, to test whether outgroup trust mattered more for Blacks than Whites.5 We show significant interactions graphically rather than reporting coefficients, as recommended (Mustillo, Lizardo, and McVeigh 2018). Standard errors of test statistics were adjusted for survey design effects.

RESULTS

Table 1 provides summary statistics and compares Blacks and Whites on the study variables. As predicted, Blacks had much lower trust in people than Whites. Only 12% of Blacks, compared to 42% of Whites, were at the highest level of trust. Blacks also trusted other Blacks, their racial ingroup, less than Whites trusted other Whites. Surpris- ingly, Blacks trusted Whites more than Whites trusted

Blacks. The two groups did not differ on trust of Hispanics or Asians.

Paired t-tests revealed differences between ingroup and outgroup trust (not shown). On average, Whites trusted other Whites more than they trusted Blacks (t = 24.51, p < .001), but Blacks trusted Blacks and Whites about equally (t = 0.82, NS). As predicted, on average both Blacks and Whites trusted their ingroup more than Hispanics (t = 3.81, p < .001 for Blacks; t = 21.92, p < .001 for Whites). Ingroup trust also was higher than trust of Asians for both Blacks (t = 1.95, p < .05) and Whites (t = 11.41, p < .001). We also investigated how ingroup trust and outgroup trust were related to each other. In data not shown, for Blacks, ingroup trust was positively related to trust in Whites (r = 0.55, p < .001), Hispanics (r = 0.58, p < .001), and Asians (r = 0.53, p < .001). For Whites, ingroup trust was positively related to trust in Blacks (r = 0.34, p < .001), Hispanics (r = 0.38, p < .001), and Asians (r = 0.52, p < .001).

As predicted, ingroup bias was generally greater among Whites than Blacks. On average, Whites’ ingroup bias vis-à-vis Blacks was 0.91, whereas Blacks’ ingroup bias vis-à-vis Whites was –0.14, where a score of 0 indicated no bias. Similarly, Whites showed more ingroup bias vis-à-vis Hispanics than Blacks did. In contrast, Blacks and Whites did not differ on ingroup bias toward Asians. Both groups showed less ingroup bias toward Asians than other outgroups. In fact, average scores were below 0, indicat- ing slightly more favorable ratings of Asians than their respective ingroups. Table 2 shows the OLS regressions of racial ingroup trust and racial outgroup trust. Looking first at racial ingroup trust, as expected, greater ingroup bias was related to higher racial ingroup trust. (For ingroup trust, ingroup bias is the average of ingroup bias toward all three out- groups.) As against Table 1, the racial difference in trust in one’s own racial group was no longer significant when ingroup bias was included in the model. That is, Blacks’ lower ingroup bias helps explain their lower ingroup trust.

The other three columns of Table 2 show regressions of trust in racial outgroups. On trust in a Black or White outgroup, here, too, the racial difference disappeared when ingroup bias was controlled. As predicted, great- er ingroup bias was associated with lower trust in the outgroup. Blacks and Whites also did not differ on trust in Hispanics or trust in Asians. As expected, greater ingroup bias was related to lower trust in those groups as well.6 Of the sociodemographic characteristics, only age was

Vol. 54 | 2020 | Virginia Social Science Journal Toussaint & Kiecolt | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | 45

consistently and positively related to ingroup and out- group trust. Education was unrelated to trust in one’s ingroup or trust in a black or white racial outgroup. It was positively related to trust in Hispanics and Asians, howev- er. Note also that the level of ingroup and outgroup trust did not change from 2000 to 2004.

Table 3 presents the ordinal logistic regression analyses of generalized trust in people. The coefficients are odds ratios (OR). Odds ratios of less than 1 indicate a negative relationship, and odds ratios greater than 1 indicate a positive relationship. For each racial outgroup, Model 1 shows the effect of race, ingroup bias, and the sociode- mographic characteristics. Model 2 added trust in one’s own racial group and in a specific racial outgroup.

The first set of models involve a Black or a White out- group. In Model 1, as expected, Blacks had much lower generalized trust than Whites (OR = 0.35, p < .001). Greater ingroup bias was related to lower generalized trust. Model 2 added trust in one’s own racial group and in a racial outgroup, in this case, either Blacks or Whites. The effect of ingroup bias was no longer significant. As predicted, trust in one’s racial group was related to higher generalized trust, as was trust in a Black or White outgroup. Notably, the effect of outgroup trust was the same for Blacks and Whites. Moreover, as race remained significant, low trust in Whites does not explain the racial difference in generalized trust.

The next set of analyses involved Hispanics as an out- group. Model 1 shows that ingroup bias vis-à-vis Hispan- ics was related to lower generalized trust. Model 2 added ingroup trust and outgroup trust in Hispanics. Both were positively related to generalized trust, and ingroup bias vis-à-vis Hispanics was no longer significant.

When Asians were the racial outgroup, the pattern differed. Ingroup bias vis-à-vis Asians was not related to generalized trust. As before, ingroup trust was related to greater generalized trust. For the total sample, trust in Asians also was related to generalized trust. However, as Figure 2 shows, trust in Asians interacted with race. It was only significant for Whites.

In Figure 2, the solid line shows that for Blacks, trust in Asians was unrelated to being at the highest level of gen- eralized trust. The dotted line shows that for Whites, the likelihood of being at the highest level of generalized trust was greater at higher levels of trust in Asians. Whites who trusted Asians the least (1 or 2 on a scale from 1 to 7) were no more likely than Blacks to have a high level of trust in people.

The effects of the sociodemographic characteristics were consistent with those of previous studies. Age, education, and being married or cohabiting were positively related to generalized trust. Residents of the South had lower generalized trust than others. Finally, generalized trust was lower in 2004 than it was in 2000. DISCUSSION

This study has examined previously untested elements of the conflict theory of trust (Putnam 2007), which links in- group bias and trust in one’s own and other racial groups to generalized trust. In addition, we expanded the theory to examine how ingroup and outgroup trust are related to generalized trust. Doing so enabled us to test an expla- nation for why Blacks have lower generalized trust than Whites do (Coverdill et al. 2011; Smith 2010; Wilkes 2011). Using nationally representative samples of U.S. Blacks and Whites, we tested the hypothesis that lower trust in Whites and other ethnoracial outgroups helps explain Blacks’ lower generalized trust. We also examined how trust in one’s own racial group and ingroup bias contrib- ute to generalized trust. Our study adds to the literature on generalized trust, trust in ingroups and outgroups, and race and trust. Three findings are especially noteworthy. First, our findings challenge the assumption that Blacks’ low trust in Whites or other racial outgroups largely accounts for their lower generalized trust. In our data Blacks trusted Whites more than Whites trusted Blacks, and Blacks and Whites did not differ on trust in Hispanics or Asians. This result differs from a previous finding that Blacks have lower overall outgroup trust than Whites (Abascal and Baldassarri 2015). However, it is consistent with the finding that Blacks and Whites showed equally trusting behavior (Simpson et al. 2007). Consequently, low trust in Whites does not explain Blacks’ lower gener- alized trust. We also found that Blacks’ trust in Whites and Whites’ trust in Blacks had the same positive association with generalized trust, no more and no less. Similarly, both Blacks’ and Whites’ trust in Hispanics was related to greater generalized trust, to the same extent. For Blacks, though, trust in Asians was unrelated to generalized trust. For Whites, trust in all racial outgroups was related to greater generalized trust. On balance, outgroup trust matters somewhat more for Whites than Blacks. These findings indicate that the radius of trust is narrower for Blacks than for Whites, contrary to what we might have expected (Marschall and Stolle 2004). They also point to the need to inquire more finely into which outgroups

46 | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | Toussaint & Kiecolt Virginia Social Science Journal | Vol. 54 | 2020

influence trust in people.

Ethnoracial intergroup dynamics also may influence generalized trust differently for Blacks and Whites. For Whites, low trust in ethnoracial outgroups may stem from viewing outgroups as different and alien, and as consti- tuting a threat to Whites’ dominant status (Blumer 1958; Bobo and Hutchings 1996; Jardina 2019). Abascal and Baldassari (2015) found that for Whites, but not racial mi- nority groups, living among more outgroup members was related to lower generalized trust. For Blacks, low trust in ethnoracial outgroups likely results more from racial alienation, a collective sense of group disenfranchise- ment rooted in long-standing, collective experiences of racial inequality and discrimination (Bobo and Hutchings 1996). On balance, Blacks had lower trust in their own group than Whites did. As ingroup trust contributed to high- er generalized trust, Blacks’ lower ingroup trust helps explain their lower generalized trust. For Blacks, both ingroup and generalized trust may depend more on trust in people they interact with in daily life, such as neighbors, relatives, and friends. Everyday interactions themselves are influenced by structural factors related to historical and continuing racial inequality. These include community characteristics such as average education, neighborhood characteristics such as residential stability, and economic well-being (Abascal and Baldassarri 2015; Ross et al. 2001).

In addition, other situational and subjective factors influence both Blacks’ reports of ingroup trust and their trusting behavior toward others. For example, despite declaring themselves as mistrustful, poor urban Blacks selectively form supportive exchange relationships with some neighbors and other community members (Raudenbush 2016). We echo Raudenbush’s call to discov- er the multiple frames people use in assessing how much to trust others.

On a related point, our measure of ingroup trust referred to a group’s trustworthiness. Understanding more about how people assess a group’s trustworthiness may inform judgments of trustworthiness more generally. Assess- ments of trustworthiness are based on an entity’s per- ceived competence or ability, benevolence, and integrity (Mayer, Davis, and Schoorman 1995). How do people rate their racial ingroup and racial outgroups on these charac- teristics? What factors influence those ratings, e.g., inter- actions with group members? How do ratings of a group’s competence, benevolence, and integrity contribute to a group’s perceived trustworthiness?

Second, based on conflict theory, we predicted that in- group bias would be related to all three types of trust—to greater ingroup trust, lower outgroup trust, and lower generalized trust. Ingroup bias had most of the predicted effects. It was related to greater ingroup trust and to low- er trust in all ethnoracial outgroups. Ingroup bias toward Black or White or Hispanic outgroup was related to lower generalized trust, but its effect became nonsignificant when ingroup and outgroup trust were added to the models. As Blumer (1958) observed, ingroup bias is not just an individual phenomenon; it is an outcome of histor- ical, economic, and political processes that shape notions of racial groups. Its negative association with general- ized trust reflects historical processes of competition between groups, especially for Whites and Blacks (Bobo and Hutchings 1996). In this regard, ingroup bias vis-à-vis Asians was lower than for the other groups, and it was not related to generalized trust. The stereotype of Asians as a “model minority,” despite their diverse ethnic origins and histories (Frey 2015), may explain these findings. Third, racial ingroup trust is positively related to trust of racial outgroups. The more people trust members of their racial ingroup, the more they trust racial out- groups. These findings suggest that bonding capital (ties to similar others) and bridging capital (ties to people in outgroups) are positively associated with each other, as Putnam (2007) proposed. This finding again suggests that the mechanism is positive interactions. Glanville (2016) points out that positive interactions with outgroup members humanizes them and promotes tolerance that extends to other outgroups (Pettigrew 1997).

As noted above, higher levels of generalized trust are related to greater economic growth (Knack and Keefer 1997). The dynamics of trust among racial groups also are likely to have economic consequences at the aggregate level. In particular, Whites’ trust in racial outgroups also may be related to greater economic growth, because it lessens racial discrimination in hiring. One study linked racial discrimination in hiring to firm survival. Specifical- ly, firms that discriminated against Black applicants in an audit study conducted six years before the 2008 recession were less likely to have survived the downturn (Pager 2016).

Our study has four limitations. First, the data were cross-sectional, so we cannot conclusively establish causal direction among the variables in our models. We assumed that trust in one’s racial ingroup and trust in racial outgroups contribute to generalized trust, just as trust in people one knows personally does (Glanville and Paxton 2007), because identity-based trust also is more

Vol. 54 | 2020 | Virginia Social Science Journal Toussaint & Kiecolt | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | 47

specific. Possibly, however, the reverse occurs, or iden- tity-based trust and generalized trust are reciprocally related. By analogy, research on involvement in voluntary associations and generalized trust finds that trust fosters membership in voluntary associations, but that mem- bership influences trust more strongly (Glanville 2016; Paxton, Hipp, and Marquart-Pyatt 2011). Longitudinal studies of how ingroup and outgroup trust are related to generalized trust are needed to address this question. Second, although we combined data from two ANES surveys, the number of Black respondents was small, and there were too few members of other ethnoracial groups to include in the analysis. Third, the measures of racial ingroup and outgroup trust were not entirely parallel to the measure of generalized trust. The three-item gen- eralized trust scale included an overall assessment of whether others can be trusted, as well as others’ per- ceived helpfulness versus self-interest and fairness versus exploitation. The measures of racial ingroup and out- group trust simply asked about a group’s trustworthiness versus untrustworthiness. Even so, both sets of measures referred to “most people”—people in general, or people in a specific ethnoracial group. Fourth, the data were from the 2000s. Unfortunately, no more recent data on trust in ethnoracial groups were available. Nevertheless, the data enabled us to address long-standing questions about how trust in racial outgroups is related to gener- alized trust. from 2000 to 2004 was part of a downward trend (Wilkes 2011). Political events around that time may have affected generalized trust. The 2000 election saw the end of Bill Clinton’s presidency, which was marred by his impeachment. By 2004, the political polarization fur- thered by Clinton’s impeachment grew with differences of opinion on the Iraq war (Lewis-Beck et al. 2008). Factors such as political party identification and political ideolo- gy, as well as growing economic inequality (Uslaner 2016) also may have influenced generalized trust.

Our findings need further examination in future research. In addition to ingroup bias, ingroup trust, and outgroup trust, comprehensive models would include community characteristics, such as racial composition, community social inequality, and racial residential segregation, as well as measures of intergroup contact and perceived threat from racial outgroups (Abascal and Baldassarri 2015; Pettigrew and Tropp 2011; Uslaner 2010; van der Meer and Tolsma 2014). Such models would help eluci- date the intergroup dynamics that influence generalized trust.

Second, future research needs to investigate whether the dynamics of ingroup bias, racial ingroup and outgroup trust, and generalized trust have changed since the early

2000s. Whites’ racial identity has become more politi- cized, and Blacks’ racial identity remains a psychologi- cal resource for fighting oppression (Jardina 2019). The Black Lives Matter movement emerged, and White racial backlash intensified (Anderson 2014). Donald Trump was elected, hate crimes have spiked, and White nationalist rhetoric has become more prominent. In the post-Obama era, ingroup bias and trust in racial ingroups and out- groups may have a greater impact on generalized trust.

Third, the relationships between ingroup bias and types of trust may differ for ethnoracial groups aside from Blacks and Whites. For example, the dynamics of ingroup bias, trust in racial groups, and generalized trust among Hispanics, Asians, and others may differ. Finally, aside from racial identity-based trust, we need to know more about how identity-based trust on other dimensions, such as religious affiliation and party identification, relate to each other and to generalized trust. END NOTES

1 In addition to conflict theory, two other theories—con- tact theory and constrict theory—have guided research on generalized trust. Contact theory proposes that eth- noracial diversity affords more opportunities for positive contacts with outgroups, thereby reducing ingroup bias and encouraging greater outgroup trust and generalized trust (Putnam 2007). Most studies that measure inter- group contact find that it is associated with more posi- tive outgroup attitudes and greater trust of ethnoracial outgroups (Hewstone 2015; Pettigrew and Tropp 2011; Schmid et al. 2014). Thus, contact theory and conflict theory make the same predictions about the effects of ingroup bias. Constrict theory proposes that ethnoracial diversity discourages contact with both ingroup and outgroup members (van der Meer and Tolsma 2014), but makes no predictions about the effects of ingroup bias.

2 The authors label trust in people one knows personally, such as friends, neighbors, and coworkers, as ingroup trust. However, trust in people one knows personally and trust in members of a social category are distinct dimen- sions of trust (Freitag and Bauer 2013). Trust in people one knows personally, termed particularized trust, is more limited. Particularized trust contributes to greater generalized trust (Glanville and Paxton 2007), but it is outside the scope of this study. 3 Scholars also have examined two other types of trust— moral (or moralistic) trust and strategic trust. Moral trust rests on the assumption that other people share one’s values, that they are part of one’s moral community. It

48 | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | Toussaint & Kiecolt Virginia Social Science Journal | Vol. 54 | 2020

is thought to be a stable orientation, learned early in life and not updated with experience (Uslaner 2016). Stra- tegic trust refers to expectations that others will act in one’s interest, based on knowledge of their motivations and intentions from previous exchanges (Hardin 2002; Smith 2010). Trust develops more readily in networks where people can monitor others’ behavior (Coleman 1990). Identity-based trust likely is related to both moral and strategic trust. People assume that ingroup members share more of their values than outgroup members do (Blumer 1958; Bobo and Hutchings 1996). People interact more with ingroup members, so they can more easily assess their trustworthiness. For cogent analyses of how moral and strategic trust differ from generalized trust, see Smith (2010) and Stolle (2002). Moral trust and strategic trust are outside the scope of this study.

4 We verified that generalized trust and ingroup/outgroup trust were empirically distinct constructs (Freitag and Bauer 2013). We performed a confirmatory factor analy- sis in which we modeled generalized trust and ingroup/ outgroup trust as either one or two dimensions. The two-dimensional model fit the data much better. It had a lower chi-squared, a higher score on the comparative fit index (CFI), and a smaller Akaike’s information criterion (AIC). Fit statistics for the two-dimensional model were chi-squared = 246.38, df = 13, p < .001, CFI = 0.965, AIC = 35074.192. Fit statistics for the one-dimensional model were chi-squared = 2718.19, df = 14, p < .001, CFI = 0.592, AIC = 37544.003. Further results are available from the authors.

5 We also tested whether the ordinal logistic regression models violated the proportional odds assumption, that is, whether the slopes of any independent variables differed across the thresholds of the dependent vari- able. Only the effects of age and education violated the assumption. They had larger positive effects on trust at higher levels of trust than at the threshold between the lowest and second lowest level. Because age and educa- tion are not our major explanatory variables, we report standard ordinal logistic regression results for ease of pre- sentation. These results are available from the authors.

6 Because respondents rated trust in four groups, the significance of the predictors might be inflated. Of great- est interest is the effect of ingroup bias. We can apply a Bonferroni correction by multiplying the p value of each regression coefficient by 4, the number of groups that respondents rated. In each case, p = .0000. Even round- ed down from p = .00004, the adjusted p values are p = .00016. Hence, the effects of ingroup bias remain signifi- cant.

Vol. 54 | 2020 | Virginia Social Science Journal Toussaint & Kiecolt | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | 49

Table 1: Descriptive Statistics for Study Variables Total sample Blacks Whites

Range Mean (S.E.)/ Proportion

N N N

Trust in people 0–3 2,252 309 1,943

0 0.21 0.41 *** 0.17***

1 0.19 0.28 0.18

2 0.23 0.19 0.23

3 0.37 0.12 0.42

Trust in racial 1–7 4.73 2,375 4.42 *** 303 4.78 *** 1,952 Ingroup (0.03) (0.08) (0.03)

Trust in Black or 1–7 4.09 2,227 4.40 * 302 4.04* 1,949 White outgroup (0.03) (0.14) (0.03)

Trust in Hispanics 1–7 4.14 2,240 4.20 300 4.13 1,940 (0.03) (0.07) (0.03)

Trust in Asians 1–7 4.44 2,233 4.34 295 4.46 1,938 (0.03) (0.08) (0.03)

Trust in outgroups 1–7 4.23 2,227 4.32 295 4.22 1,932 (0.03) (0.08) (0.03)

Ingroup bias toward

–6–+6 0.76 2,220 –0.14 *** 297 0.91*** 1,923

Blacks or Whites (0.04) (0.11) (0.04)

Ingroup bias toward

–6–+6 0.56 2,171 0.06 *** 286 0.65*** 1,885

Hispanics (0.04) (0.07) (0.04)

Ingroup bias toward

–6–+6 –0.10 2,160 –0.21 287 –0.09 1,873

Asians (0.03) (0.08) (0.03)

Control Variables

Age 18– 47.03 2,413 43.02 * 337 47.74* 2,076 96 (0.50) (1.04) (0.58)

Female 0, 1 0.54 2,417 0.54 337 0.54 2,080 Education 1–7 4.03 2,415 3.53 *** 337 4.12*** 2,078 (highest degree) (0.05) (0.09) (0.05)

Married/cohab- iting

0, 1 0.58 2,417 0.35 *** 337 0.63*** 2,080

South 0, 1 0.36 2,417 0.63 *** 337 0.31*** 2,080 Year (2004 = 1) 0, 1 0.41 2,417 0.49 * 337 0.40* 2,080 Note: Means (and standard errors) or proportions for weighted Ns; unweighted Ns are shown. *p < .05. **p < .01. ***p < .001.

50 | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | Toussaint & Kiecolt Virginia Social Science Journal | Vol. 54 | 2020

Table 2. Estimated Regression Coefficients from OLS Regressions of Trust in One’s Racial Ingroup and Racial Outgroups

Trust in One’s Own Racial Group

Trust in Black or White

Outgroup

Trust in Hispanics

Trust in Asians

N = 2,249

N = 2,330

N = 2,279 N = 2,266

Race (Black = 1) –0.04 –0.08 –0.02 –0.05 Ingroup biasa 0.90*** –0.34*** –0.29*** –0.15*** Age 0.01** 0.003* 0.004* 0.004* Woman 0.00 0.02 –0.01 –0.003 Education 0.02 0.04 0.06** 0.05* Married or cohabiting 0.11 0.16* 0.15 0.12 South 0.06 0.02 0.05 –0.001 Year (2004 = 1) –0.01 –0.07 –0.02 –0.07 Constant 4.55*** 4.34*** 4.31*** 4.45*** R-squared 0.19 0.15 0.10 0.04 ***p < .001; **p < .01; *p < .05 (two-tailed tests).

aFor ingroup bias, outgroups were measured as (1) a composite of all three groups for ingroup trust, (2) Blacks or Whites for trust in Black or White outgroup, (3) Hispanics for Hispanic outgroup, and (4) Asians for Asian outgroup. Note: All control variables were centered at their means.

Vol. 54 | 2020 | Virginia Social Science Journal Toussaint & Kiecolt | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | 51

Table 3. Estimated Odds Ratios from Ordinal Regressions of Generalized Trust Black or White Outgroup Hispanic Outgroup Asian Outgroup (N = 2,289) (N = 2,244) (N = 2,230) Model 1 Model 2 Model 1 Model 2 Model 1 Model 2 Race 0.35*** 0.32*** 0.37*** 0.36*** 0.38 *** 0.39*** (Black = 1) Ingroup biasa 0.89* 0.95 0.92* 0.96 0.98 0.96 Racial 1.17** 1.15** 1.17** ingroup trust Trust in Black 1.39*** or White racial outgroup Trust in 1.37*** Hispanics Trust in Asians 1.23*** Control variables Age 1.02*** 1.02*** 1.02*** 1.02*** 1.02*** 1.02*** Female 1.00 1.02 0.99 1.01 0.97 0.98 Education 1.38*** 1.38*** 1.38*** 1.37*** 1.38*** 1.37*** Married or cohabiting 1.47*** 1.39** 1.44*** 1.36** 1.46*** 1.41*** South 0.74** 0.72** 0.76* 0.74* 0.76* 0.75* Year (2004 versus 2000) 0.71** 0.72** 0.72** 0.72** 0.72** 0.73** Cut 1 –1.80 0.25 –1.77 0.17 –1.73 –0.09 Cut 2 –0.73 1.37 –0.70 1.29 –0.65 1.01 Cut 3 0.38 2.53 0.42 2.45 0.46 2.15 F 45.76 49.46 44.73 53.41 45.94 50.17 ***p < .001; **p < .01; *p < .05 (two-tailed tests). aFor ingroup bias, outgroups were measured as (1) Blacks or Whites for trust in Black or White outgroup, (2) Hispanics for Hispanic outgroup, and (3) Asians for Asian outgroup. Note: All control variables were centered at their means.

52 | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | Toussaint & Kiecolt Virginia Social Science Journal | Vol. 54 | 2020

Figure 1. Conceptual Model of How Race/Ethnicity, Ingroup Bias, and Trust in Racial Ingroup and Racial Outgroups Affect Generalized Trust in People

Ingroup Bias

Trust in Own Racial Group

Generalized Trust in People

+

–

Trust in Other Racial Groups

+

–

+

Race/Ethnicity

Vol. 54 | 2020 | Virginia Social Science Journal Toussaint & Kiecolt | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | 53

54 | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | Toussaint & Kiecolt Virginia Social Science Journal | Vol. 54 | 2020

REFERENCES

Abascal, Maria and Delia Baldassarri. 2015. “Love Thy Neighbor? Ethnoracial Diversity and Trust Re- examined.” American Journal of Sociology 121 (3):722-82.

Anderson, Carol. 2014. “Ferguson Isn’t About Black Rage against Cops. It’s White Rage against Progress.” Washington Post.

Bizumic, Boris and John Duckitt. 2012. “What Is and Is Not Ethnocentrism? A Conceptual Analysis and Political Implications.” Political Psychology 33 (6):887-909.

Blumer, Herbert. 1958. “Race Prejudice as a Sense of Group Position.” Pacific Sociological Review 1 (1):3-7.

Bobo, Lawrence D. and Devon Johnson. 2000. “Racial Attitudes in a Prismatic Metropolis: Mapping Identity, Stereotypes, Competition, and Views on Affirmative Action.” Pp. 81-161 in Prismatic Metropolis: Inequality in Los Angeles, edited by L. D. Bobo, M. L. Oliver, J. M. Johnson, Jr. and A. Valenzuela, Jr. New York: Russell Sage

Bobo, Lawrence D. and Michael P. Massagli. 2001. “Ste- reotyping and Urban Inequality.” Pp. 89-162 in Urban Inequality: Evidence from Four Cities, edit- ed by A. O’Connor, C. Tilly and L. Bobo. New York: Russell Sage Foundation.

Bobo, Lawrence and Vincent L. Hutchings. 1996. “Per- ceptions of Racial Group Competition: Extending Blumer’s Theory of Group Position to a Multiracial Social Context.” American Sociological Review 61 (6):951-972.

Brewer, Marilynn. 1981. “Ethnocentrism and Its Role in In- terpersonal Trust.” Pp. 214-231 in Scientific Inqui- ry and the Social Sciences, edited by M. B. Brewer and B. E. Collins. San Francisco, CA: Jossey-Bass.

Brewer, Marilynn B. 1999. “The Psychology of Prejudice: Ingroup Love or Outgroup Hate?” Journal of So- cial Issues 55 (3):429-429.

Brewer, Marilynn B. and Masaki Yuki. 2007. “Culture and Social Identity.” Pp. 307-322 in Handbook of Cultural Psychology, edited by S. Kitayama and D. Cohen. New York: Guilford Press.

Burns, Nancy, Donald R. Kinder, Steven J. Rosenstone, Virginia Sapiro, and University of Michigan. Insti- tute for Social Research. 2016. American National Election Study, 2000: Pre- and Post-Election Sur- vey. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor].

Campbell, Angus, Philip E. Converse, Warren Miller, and Donald Stokes. 1960. The American Voter. New York: John Wiley & Sons.

Coleman, James S. 1990. Foundations of Social Theory. Cambridge, MA: Belknap/Harvard.

Coverdill, James E., Carlos A. Lopez, and Michelle A. Petrie. 2011. “Race, Ethnicity and the Quality of Life in America, 1972-2008.” Social Forces 89 (3):783-805.

Deaux, Kay and Daniela Martin. 2003. “Interpersonal Net- works and Social Categories: Specifying Levels of Context in Identity Processes.” Social Psychology Quarterly 66 (2):101-117.

Delhey, Jan, Kenneth Newton, and Christian Welzel. 2011. “How General Is Trust in “Most People”? Solving the Radius of Trust Problem.” American Sociological Review 76 (5):786-807.

Freitag, Markus and Paul C. Bauer. 2013. “Testing for Measurement Equivalence in Surveys: Dimensions of Social Trust across Cultural Contexts.” Public Opinion Quarterly 77:24-44.

Frey, William H. . 2015. Diversity Explosion: How New Ra- cial Demographics Are Remaking America. Wash- ington, DC: Brookings Institution Press

Glanville, Jennifer L. 2016. “Why Does Involvement in Voluntary Associations Promote Trust? Examining the Role of Network Diversity.” Sociological Inqui- ry 86 (1):29-50.

Glanville, Jennifer L. and Pamela Paxton. 2007. “How Do We Learn to Trust? A Confirmatory Tetrad Analysis of the Sources of Generalized Trust.” Social Psy- chology Quarterly 70 (3):230-242.

Hardin, Russell. 2002. Trust and Trustworthiness, The Russell Sage Foundation Series on Trust. New York: Russell Sage Foundation.

Hewstone, Miles. 2015. “Consequences of Diversity for Social Cohesion and Prejudice: The Missing Di- mension of Intergroup Contact.” Journal of Social Issues 71 (2):417-438.

Hewstone, Miles, Mark Rubin, and Hazel Willis. 2002. “Intergroup Bias.” Annual Review of Psychology 53:575-604.

Hughes, Michael and Melvin E. Thomas. 1998. “The Con- tinuing Significance of Race Revisited: A Study of Race, Class, and Quality of Life in America, 1972 to 1996.” American Sociological Review 63 (6):785- 795.

Jardina, Ashley. 2019. White Identity Politics. Cambridge, United Kingdom: New York, NY, USA : Cambridge University Press.

Kiecolt, K. Jill and Michael Hughes. 2017. “Racial Identity and the Quality of Life among Blacks and Whites in the U.S.” Social Science Research 67:59-71.

Vol. 54 | 2020 | Virginia Social Science Journal Toussaint & Kiecolt | Ingroup Bias, Trust in Racial Groups, and Generalized Trust | 55

Knack, Stephen and Philip Keefer. 1997. “Does Social Capital Have an Economic Payoff? A Cross-Coun- try Investigation.” Quarterly Journal of Economics 112 (4):1251-1288.

Kramer, Roderick M., Benjamin A. Hanna, Steven Su, and Jane Wei. 2001. “Collective Identity, Collective Trust, and Social Capital: Linking Group Identi- fication and Group Cooperation.” Pp. 173-196 in Groups at Work: Theory and Research, edited by Marlene E. Turner. Mahwah, NJ: L. Erlbaum.

Lewis-Beck, Michael S., William G. Jacoby, Helmut Nor- poth, and Herbert F. Weisberg. 2008. The Ameri- can Voter Revisited. Ann Arbor, MI: University of Michigan Press.

Marschall, Melissa J. and Dietlind Stolle. 2004. “Race and the City: Neighborhood Context and the Develop- ment of Generalized Trust.” Political Behavior 26 (2):125-153.

Mayer, Roger C., James H. Davis, and F. David Schoor- man. 1995. “An Integrative Model of Organiza- tional Trust.” Academy of Management Review 20 (3):709-734.

Mustillo, Sarah A., Omar A. Lizardo, and Rory M. McVeigh. 2018. “Editors’ Comment: A Few Guidelines for Quantitative Submissions.” American Sociological Review 83 (6):1281-1283.

Pager, Devah. 2016. “Are Firms That Discriminate More Likely to Go out of Business? .” Sociological Sci- ence 3:849-859.

Paxton, Pamela , John R. Hipp, and Sandra Marquart-Py- att. 2011. Nonrecursive Models: Endogeneity, Reciprocal Relationships, and Feedback Loops, Quantitative Applications in the Social Sciences (Qass) Series #07-168. Thousand Oaks, CA: Sage.

Pettigrew, Thomas F. 1997. “Generalized Intergroup Con- tact Effects on Prejudice.” Personality and Social Psychology Bulletin 23 (2):173-185.

Pettigrew, Thomas F. and Linda R. Tropp. 2011. When Groups Meet: The Dynamics of Intergroup Con- tact. New York: Psychology Press.

Putnam, Robert D. 2007. “E Pluribus Unum: Diversity and Community in the 21st Century: The 2006 Johan Skytte Prize Lecture.” Scandinavian Political Stud- ies 30 (2):137-174.

Raudenbush, Danielle. 2016. “”I Stay by Myself”: Social Support, Distrust, and Selective Solidarity among the Urban Poor.” Sociological Forum 31 (4):1018- 1039.

Ross, Catherine E., John Mirowsky, and Shana Pribesh. 2001. “Powerlessness and the Amplification of Threat: Neighborhood Disadvantage, Disorder, and Mistrust.” American Sociological Review 66 (4):568-591.

Schmid, Katharina, Ananthi Al Ramiah, and Miles Hew- stone. 2014. “Neighborhood Ethnic Diversity and Trust the Role of Intergroup Contact and Per- ceived Threat.” Psychological Science 25 (3):665- 674.

Simpson, Brent, Tucker McGrimmon, and Kyle Irwin. 2007. “Are Blacks Really Less Trusting Than Whites? Revisiting the Race and Trust Question.” Social Forces 86 (2):525-552.

Smith, Sandra Susan. 2010. “Race and Trust.” Annual Review of Sociology 36:453-475.

Smith, Tom W. 1997. “Factors Relating to Misanthropy in Contemporary American Society.” Social Science Research 26 (2):170-196.

Stolle, Dietlind. 2002. “Trusting Strangers—the Concept of Generalized Trust in Perspective.” Österre- ichische Zeitschrift für Politikwissenschaft 31:397- 412.

Tajfel, Henri and John C. Turner. 1986. “The Social Iden- tity Theory of Intergroup Behavior.” Pp. 2-24 in Psychology of Intergroup Relations, edited by S. Worchel and L. W. Austin. Chicago, IL: Nelson-Hall.

University of Michigan, Institute for Social Research. Center for Political Studies. 2016. American Na- tional Election Study, 2004: Pre- and Post-Election Survey. Ann Arbor, MI: Inter-university Consortium for Political and Social Research.

Uslaner, Eric. 2010. “Segregation, Mistrust and Minori- ties.” Ethnicities 10 (4):415-434.

Uslaner, Eric M. 2016. “Who Do You Trust?” Pp. 71-83 in Interdisciplinary Perspectives on Trust: Towards Theoretical and Methodological Integration, ed- ited by E. Shockley, T. M. S. Neal, L. M. PytlikZillig and B. H. Bornstein. Cham, Switzerland: Springer International Publishing.

van der Meer, Tom and Jochem Tolsma. 2014. “Ethnic Di- versity and Its Effects on Social Cohesion.” Annual Review of Sociology 40:459-478.

Wilkes, Rima. 2011. “Re-Thinking the Decline in Trust: A Comparison of Black and White Americans.” Social Science Research 40 (6):1596-1610.

Williams, David R. and Selina A. Mohammed. 2013. “Racism and Health I: Pathways and Scientif- ic Evidence.” American Behavioral Scientist 57 (8):1152-1173.

Wodtke, Gregory T. 2012. “The Impact of Education on In- tergroup Attitudes: A Multiracial Analysis.” Social Psychology Quarterly 75 (1):80-106.

SUGGESTED CITATION Toussaint, J. G. and Kiecolt, K. J. (2020). Ingroup Bias, Trust in Racial Groups, and Generalized Trust among U.S. Blacks and Whites. Virginia Social Science Journal, Vol. 54 pp. 40-55

Copyright of Virginia Social Science Journal is the property of James Madison University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.