Children’s and Adults’ Predictions of Black, White, and Multiracial Friendship Patterns

Cross-race friendships can promote the development of positive racial attitudes, yet they are relatively uncommon and decline with age. In an effort to further our understanding of the extent to which children expect cross-race friendships to occur, we examined 4to 6-year-olds’ (and adults’) use of race when predicting other children’s friendship patterns. In contrast to previous research, we included White (Studies 1 and 2), Black (Study 3), and Multiracial (Study 4) participants and examined how they predicted the friendship patterns of White, Black, and Multiracial targets. Distinct response patterns were found as a function of target race, participant age group, and participant race. Participants in all groups predicted that White children would have mostly White friends and Black children would have mostly Black friends. Moreover, most participant groups predicted that Multiracial children would have Black and White friends. However, White adults predicted that Multiracial children would have mostly Black friends, whereas Multiracial children predicted that Multiracial children would have mostly White friends. These data are important for understanding beliefs about cross-race friendships, social group variation in race-based reasoning, and the experiences of Multiracial individuals more broadly. Cross-race friendships (i.e., friendships between members of two different racial backgrounds) can play an important role in the development of positive racial attitudes (for a review, see McGlothlin, Edmonds, & Killen, 2007). These friendships teach us about within-group variation (two individuals of the same racial group need not share the same interests), the inaccuracy of stereotypes (because of within-group variation, judgments on the basis of racial group membership can be incorrect), and the wrongness of prejudice and race-based exclusion (not interacting with someone because of their racial group membership is unfair and unjust; Crystal, Killen, & Ruck, 2008). Indeed, research has shown that U.S. children from racially heterogeneous schools, compared with those from racially homogenous schools, are less likely to interpret ambiguous situations in terms of race (McGlothlin & Killen, 2010), and British children from racially heterogeneous kindergartens, compared with those from racially homogenous kindergartens, are less likely to make trait attributions on the basis of race (Rutland, Cameron, Bennett, & Ferrell, 2005). Thus, cross-race friendships decrease race-based biases and distance between racial groups; increase close, collaborative, and supportive intergroup CONTACT Steven O. Roberts sothello@umich.edu Department of Psychology, University of Michigan, 530 Church Street, Ann Arbor, MI 48109, USA. JOURNAL OF COGNITION AND DEVELOPMENT 2017, VOL. 18, NO. 2, 189–208 http://dx.doi.org/10.1080/15248372.2016.1262374 © 2017 Taylor & Francis relationships; and foster beliefs about social fairness and equity (see also Graham, Munniksma, & Juvonen, 2014). Despite the importance of cross-race friendships, they are relatively uncommon and their prevalence declines as children grow older. For example, during first grade, children have more same-race friends than cross-race friends (Aboud, Mendelson, & Purdy, 2003; Finkelstein & Haskins, 1983), and by sixth grade, the already low number of cross-race friendships declines even further (Graham & Cohen, 1997). Furthermore, the cross-race friendships that do develop are often less intimate and long-lasting compared with samerace friendships (Aboud et al., 2003; Fletcher, Rollins, & Nickerson, 2004). Given that cross-race friendships are rare and decline with age, it is important to understand what expectations children hold about them, which could provide critical information regarding the extent to which children see these kinds of relationships as likely and also the extent to which they themselves are likely to engage in them. In one study, Kinzler, Shutts, Dejesus, and Spelke (2009; see Experiment 3) examined children’s use of race when making friendship predictions. They showed a sample of White U.S. children (aged 5 years) photos of Black children and White children and asked them to indicate which child they themselves wanted to be friends with. Children were more likely to select White friends than Black friends. Similarly, Shutts, Pemberton Roben, and Spelke (2013) presented White U.S. preschoolers with a series of photos of Black children and White children. Each target child was introduced by name and was described as engaging in a certain activity (e.g., “This is Jessica. She is playing hide-and-seek today with all her friends”). Next, two response images were presented—one image of a Black child and one of a White child—and the preschoolers were asked to indicate which response image was a friend of the child. Four-year-olds predicted that children would be friends with children of their same race, demonstrating that they inferred more same-race friendships than cross-race friendships. Critically, though, previous studies have focused on the dichotomous categories of “Black” and “White” to assess children’s friendship predictions. Although this method was useful, given that it provided careful experimental control and a maximally distinctive task that young children could use, it was nevertheless limiting because it oversimplified race, which may therefore have oversimplified our understanding about racebased concepts (see Dunham & Olson, 2016; Kang & Bodenhausen, 2014). By giving children clear-cut racial exemplars, previous researchers may have made race maximally salient and therefore overestimated the extent to which children used race to predict friendships. Thus, as a critical next step toward furthering our understanding of children’s friendship predictions, research that goes beyond dichotomous and clear-cut notions of race to include greater perceptual variation is needed, as it more closely reflects real-world variation. Friendship predictions for multiracial children Children with both Black and White parentage (henceforth referred to simply as Multiracial children) are one of the youngest and fastest-growing demographic groups in the United States (Pew Research Center, 2015; U.S. Census Bureau, 2011). They challenge the dichotomous notion that one is either Black orWhite and represent a weakening of race-based social boundaries and an increase in cross-race relationships (Lee & Bean, 2007; Liebler, 2016; Morning, 2009; Rockquemore, Brunsma, & Delgado, 2009; Young, Sanchez, & Wilton, 2013). Research has shown that social relationships have important implications for 190 S. O. ROBERTS ET AL.