Online Program Home
My Program

Abstract Details

Activity Number: 189 - Contributed Poster Presentations: Section on Statistical Education
Type: Contributed
Date/Time: Monday, July 29, 2019 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics and Data Science Education
Abstract #304784
Title: Predicting Undergraduate Student Success Using Geographically Weighted Logistic Regression
Author(s): James Roddy* and Samantha Robinson
Companies: University of Arkansas, Fayetteville and University of Arkansas
Keywords: Geographically Weighted Logistic Regression (GWLR); Local Spatial Modeling; Undergraduate Education; Student Success; Education Policy

Due to interest in undergraduate student success, including individual success in early coursework and overall timely college graduation, attempts to identify ‘at-risk’ students using both demographics and pre-college variables (e.g., SAT scores)are numerous. Interventions and programs on college campuses utilize these ‘at-risk’ identifications in an effort to increase student success. Despite the interest inaccurate identification of ‘at-risk’ students, few studies investigate the impact that pre-college location (i.e., hometown geographic location) has on student success. Geographically Weighted Logistic Regression (GWLR), a local spatial model, is used to explore the impact location has on student success of 2012-2014 freshman (i.e., first year) cohorts at one midsized, public university. Mappings of the results are presented, which can assist policy makers at institutions of higher education identify and intervene with ‘at-risk’ students. The predictive accuracy of these GWLR models, with and without covariates, is then compared to logistic regression models that do not take into consideration location, highlighting the benefits of spatial models in education research.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program