Regency C
Student Retention Modeling by Use of Pre- and Post-Enrollment Data (303973)
*Sima Sharghi, Bowling Green State University
Kevin Edward Stoll, Bowling Green State University
Keywords: Retention, Modeling, ACS, American Community Survey, Tree based modeling, Statistical Learning,Random Forest, Decision Tree, Predictive Analytics, GBM
Retention affects student success and university reputation and impels universities to invest in creative ways to increase retention. Bowling Green State University (BGSU) uses extensive research and modeling to help understand retention and promote student success. Here we expand upon the use of statistical learning to model retention at BGSU. The modeling applies various tree based statistical learning methods to predict semester retention rates per student across their academic journey. In addition to the pre and post-enrollment predictors, we gather important demographic predictors from the 2017 American Community Survey to better predict and understand retention. Particularly, estimates of the median housing value, median income, and population density of the block group corresponding to students' home address are gathered. We elaborate on variables playing an important role in the students' success and retention throughout and discuss how BGSU can use this information to take appropriate actions.