Social cognitive model of adjustment to engineering majors: Longitudinal test across gender and race/ethnicity

• Conducted a longitudinal test of social cognitive model of adjustment to engineering majors
• The model was designed to explain students' satisfaction with and intentions to persist in their majors.
• A bidirectional version of the model offered good fit to the data across gender and racial/ethnic groups.
• Self-efficacy was the most reliable direct predictor of academic satisfaction and intended persistence.

We conducted a longitudinal test of a social cognitive model of academic adjustment in a sample of 732 engineering students. The model, designed to explain students' satisfaction with and intentions to persist in their majors, integrated features of social cognitive career theory's (SCCT) segmental models of satisfaction, interest, choice, and performance (Lent & Brown, 2006; Lent, Brown, & Hackett, 1994). Students completed measures of academic support, self-efficacy, outcome expectations, interests, satisfaction, positive affect, and intended persistence at three time points (at the end of their second, third, and fourth semesters in engineering). A bidirectional version of the model offered good fit to the data, both in the larger sample and across gender and racial/ethnic groups. Self-efficacy was the most reliable direct predictor of academic satisfaction and intended persistence across the third and fourth semesters, though other social cognitive variables also contributed, either directly or indirectly, to predictions at one time point or the other. We consider implications of the findings for further research and practice on academic adjustment and persistence in STEM fields.