The retention impact of the Freshman Seminar on students with varying pre-college academic performance
Abstract
The Freshman Seminar has been a widely implemented retention intervention on college campuses for many years. The current study sought to determine what impact the seminar had on retention, as well as identifying predictors of second year retention and determining if an interaction existed between Freshman Seminar participation and academic preparation on retention. Utilizing guidance from Astin's and Tinto's models, the current study analyzed data from the 2002 freshman cohort at the University of Wisconsin-Whitewater, a regional public four-year institution. Of the 2,064 freshmen, 626 (30.3%) completed the one-credit elective Freshman Seminar. Using hierarchical logistic regression, the analysis showed high school class rank (p < .001) to be a retention predictor, but with a low odds ratio. Adding other variables, the most significant predictor was first-year GPA ( p < .041), demonstrating a higher effect size at Exp(B) = 3.709. Freshman Seminar impact was tested with the significant predictors from the first analysis. The seminar was not found to be a significant predictor of retention in this block. When tested, no interaction was noted between the Freshman Seminar and academic preparation suggesting that the seminar impact was similar regardless of academic preparation. Secondary analyses were then conducted. When the first-year GPA variable was removed, the Freshman Seminar showed statistical significance at p = .008, but had a low effect size. A chi-squared test was also run comparing the retention frequency for seminar participants and non-participants, demonstrating significance at χ2 (1, N = 2064) = 8.8436, p = .005. The current study, however, did not isolate the Freshman Seminar as the only variable impacting retention. A t-test was run to test for self-selection bias, demonstrating that better academically prepared students did not self-select to take the Freshman Seminar. Based on the current study's model, the conclusion must be drawn that the Freshman Seminar was not a significant predictor of retention. This is not an indictment on the Freshman Seminar. Rather, it is an acknowledgement that many factors impact student retention and completion of the Freshman Seminar is one of many. Evidence, particularly in the secondary analyses, suggested that the seminar impacted retention, but to a small extent.
This paper has been withdrawn.