Abstract:
Many colleges and universities are grappling with the ever-increasing challenge of understanding the impact that college has on students. Research has validated the fact that student engagement has a significant impact on academic performance, satisfaction with college experience, and graduation rates. Student engagement is the connection that students have, both academically and extracurricular, to their institution. Beyond that, student engagement is the manner in which higher education institutions help to foster student success. Undeniably, African American students at Historically Black Colleges and Universities (HBCUs) seem to face greater, perhaps unique, challenges in their quest to obtain a degree. The number of barriers that many of these students face is surreal, and the impact that HBCUs have on the educational outcomes of their students warrants further exploration. Exploratory in nature, this study examined the relationships between student engagement patterns and academic performance of undergraduate students at a public historically Black higher education institution in the Southeast. This study was ex post facto, since the data from students' responses to the survey items on the 2008 National Survey of Student Engagement (NSSE) had been collected prior to the design of this research study. In an effort to answer the research questions that were posed, the general linear model univariate analysis of variance statistical procedure was employed to model the relationships between the NSSE Benchmarks (i.e., academic challenge, active and collaborative learning, study-faculty interaction, supportive campus environment, and enriching educational experiences) and the outcome variable, student self-reported academic performance. The findings support the conclusion that student engagement has multi-dimensional effects. While a particular independent variable in one model might have had a statistically significant linear relationship with the dependent variable, student self-reported academic performance, that independent variable might not have had a statistically significant linear relationship when other independent variables were added to the model. For the most part, in some models, several of the independent variables did not have an additive effect and failed to contribute to the statistical significance, relationship, or prediction of the dependent variable.
