ABSTRACT
Social class bicultural identity integration research demonstrates that integrated social class identities are linked with better health, well-being, and academic performance among first-generation students. Here, we demonstrate that exposure to college graduates in students’ home neighborhoods before college is positively related to higher social class bicultural identity integration (Study 1), that the effect of identity integration on academic performance is mediated by academic self-efficacy (Study 2), and that the effects of identity integration on acculturative stress, life satisfaction, and overall health outcomes observed at a large, public university replicated at selective, private universities (Study 3). This suggests that the identity integration framework is a useful theoretical lens to conceptualize and predict health and performance outcomes for first-generation students.
Disclosure of potential conflicts of interest
No potential conflict of interest was reported by the author(s).
Notes
1. See Appendix I for all materials.
2. Year in school was not collected. Typical samples from introductory psychology classes at this university are majority first-year. Please see studies 2 and 3 for breakdowns.
3. 2014 is an accurate reflection of when participants would have been living in these neighborhoods.
4. Please see Appendix IV for a correlation table with all neighborhood-level predictors.
5. Please see Appendix IV for a correlation table with all school-level predictors.
6. Please see Appendix II for all materials.
7. Logistic regression examined differences in attrition by race, age, gender, income, social class, subjective SES, and social class bicultural identity integration. A test of the full model against a constant-only model was not significant, X2(7) = 10.32, p = .17. No variable made significant contributions to prediction (all p’s > .23). Thus, the missing data was completely at random.
8. First-generation students have lower belonging and higher perceived cultural mismatch across college (Phillips et al., Citation2020; Tibbetts et al., 2016).
9. An original item, “I haven’t been mixing too well with the opposite sex lately,” was revised to, “I am not satisfied with my dating relationships lately,” to make it more inclusive to LGBTQ participants.
10. Given the interactions of identity integration and ethnicity on social integration and belonging, we conducted separate mediation analyses for ethnic majority and minority participants, but found no significant mediation. Additionally, we conducted moderated mediation analyses (Model 7; Hayes, Citation2018) with ethnicity as a moderator, but the model was not significant.
11. Please see Appendix III for all materials.
12. All findings were the same between the two private universities.
13. To compare GPA across Studies 2 and 3, as well as Studies 2, 4, and the replication in the supplement from Herrmann and Varnum (Citation2018a), we calculated I2 and Cochran’s Q (Cumming, 2012). The total sample size was 1395. GPA had substantial heterogeneity, I2 = 92.46%, Q = 79.58, p < .01 (Higgins, 2003). Additionally, we analyzed effect size across studies using fixed effects, weighted by sample size. All correlations were Fisher’s z transformed and converted back to Pearson correlations. Overall, the effect was significant, M r = .11, Z = 5.65, 95% CI [.08, .15], p < .001. Those high in social class bicultural identity integration have higher GPAs.