Testing models of reciprocal relations between social influence and integration in STEM across the college years

The present study tests predictions from the Tripartite Integration Model of Social Influences (TIMSI) concerning processes linking social interactions to social integration into science, technology, engineering, and mathematics (STEM) communities and careers. Students from historically overrepresented groups in STEM were followed from their senior year of high school through their senior year in college.  Based on TIMSI, we hypothesized that interactions with social influence agents (operationalized as mentor network diversity, faculty mentor support, and research experiences) would promote both short- and long-term integration into STEM via social influence processes (operationalized as science self-efficacy, identity, and internalized community values).  Moreover, we examined the previously untested hypothesis of reciprocal influences from early levels of social integration in STEM to future engagement with social influence agents.  Results of a series of longitudinal structural equa...