Adolescents’ Friendship Selection Based on Caste, SES and Academic Achievement: The Role of School Diversity

• In Event: 1-217 - Intergroup Friendships in Cultural Context: Harnessing the Power of Social Network Approaches

Integrative Statement

Adolescents growing up in Nepal experience a society that is characterized by social hierarchies and strong social inequalities (Gurung, Suwal, Pradhan, & Tamang, 2012). To what extent do these adolescents select their friends based on social criteria related to social status? For example, is it more important to have friends from the same caste or friends with high academic achievement? Does attending schools with a high number of students from different castes and different SES enhance the formation of friendships with students from different social backgrounds?
To answer these questions, we used social network analysis in order to better understand the relative importance of classmates’ gender, SES, caste, and academic achievement for friendship selection since all effects can be analyzed in one single model (see Figure 1). In particular, we assumed that the likelihood of choosing friends on the basis of social status criteria would be less pronounced in schools that were more diverse with regards to SES and caste than schools that were more homogeneous. In addition to studying friendship selection, we investigated potential positive effects of friendship formation. Based on intergroup contact theory (Tropp & Prenovost, 2008), we assumed that the formation of friendships with students from low SES families predicted increases in adolescents’ sympathy towards the poor.
Nepalese adolescents from a rural area attending grades 8 to 10 (M = 14.58, SD = 1.22) in five different schools (i.e., 16 school classes) nominated their best friends from their school class at two measurement times (5-7 months apart; N = 449T1, N = 438T2). At T1, they answered questions regarding family housing and property (to estimate their SES) and their last name (indicating their caste). Intergroup sympathy was assessed by asking students about their hypothetical reactions to three situations describing the social exclusion of a low SES student. Schools provided students’ academic record. School diversity was based on the number of students from different castes and the ratio of students with low and high SES.
Results from a multi-group analysis in RSIENA (see Table 1) revealed that adolescents’ friendship selection was based on gender, SES, caste, and academic achievement. In particular, adolescents were more likely to nominate friends from higher than lower SES. In addition, adolescents tended to form friendships with students from the same caste (i.e, caste homophily). Surprisingly, rather than SES or caste, similarity in academic achievement seemed to be the strongest predictor for friendship selection. Most importantly, adolescents’ friendship selection based on caste and academic achievement was less pronounced in schools with higher diversity than schools with lower diversity. Lastly, supporting the intergroup contact hypothesis, adolescents who had more friends with a low SES increased in their intergroup sympathy over time.
These findings add to the important question of how adolescents weigh different social criteria to select their friends. Moreover, the results suggest that adolescents attending heterogeneous school contexts may have a greater openness to choose friends from different castes. Thus, the findings highlight the potential of diverse school environments to increase social inclusion.

Authors

• Jeanine Grütter, University of Zurich / University of Maryland
  Presenting Author

• Sandesh Dhakal, Tribhuvan University
  Non-Presenting Author

• Melanie Killen, University of Maryland
  Non-Presenting Author