Social cognitive predictors of Latinx and White engineering students' academic satisfaction and persistence intentions: Exploring interactions among social identities and institutional context

Abstract Engineering has long lacked gender and racial diversity among its ranks, with Latinx and White women substantially underrepresented in engineering. The present study explored the role of social cognitive variables in engineering academic satisfaction and persistence intentions with a diverse sample of 1022 engineering students using the integrative social cognitive model of academic adjustment (Lent et al., 2013). Results indicated that (a) the hypothesized bidirectional model fit the data well for the full sample, subsamples based on race, gender, institution, race × gender, and race × institution, (b) all but six parameters were significant and in the expected direction for the full sample, (c) model parameters were invariant across race, gender, institution, and race × gender, but differed across race × institution subsamples, and (d) the relations within the model explained a significant amount of variance in engineering academic satisfaction and persistence intentions for the full sample and sub-samples. Implications of the findings are aimed at educational and career interventions focusing on retaining Latinx and women in engineering.