Student/faculty interaction, risk status, academic identity and persistence of undergraduate students
The purpose of this dissertation was to ascertain whether student/faculty interaction, risk status and academic identity impact persistence of undergraduate students. While research shows that student/faculty interactions influence student persistence and success, the nature of student/faculty interactions that are most conducive to academic identity and persistence in conjunction with risk status has not been explored. This is a topic of importance that goes beyond simply an academic tool utilized by the institution to measure graduation rates. Persistence affects the reputation of the institution and translates into funding the University depends on to thrive. Further, while much research has been done on student interaction with faculty, little research on student/faculty interaction in conjunction with risk status and academic identity has been explored. There remains a gap in the literature specifically on the impact of student-faculty interactions, risk status (as it is defined as first-generation, African American and Hispanic undergraduate students) and academic identity on the persistence of that this study addresses. OLS Regression and Binary Logistic Regression analysis was used to investigate the relationships between student/faculty interaction, risk status, academic identity and persistence. Findings indicate that the relationship between student/faculty interaction and persistence is most effective when the aspect of academic identity is also included as a factor, though students will not persist necessarily based on academic identity alone. This study suggests that an increase in the frequency and perceived quality of student/faculty interaction will increase levels of behavior consistent with a strong academic identity, which in turn will increase the rate of persistence. Research suggests the possibility that this concept will also influence persistence rates of high risk students, though this finding is not conclusive based on this study. Limitations and implications for further research are discussed.