Building a Predictive Model for Short-Term Student Success Using Learning Analytics

• In Event: Roundtable Session 20
  In Roundtable Session: Making It Through: Predicting College Student Success, Persistence, and Attainment

Abstract

Learning Analytics has been introduced to higher education in the recent years and has been embraced by some colleges to be a powerful tool in effectively predicting student learning outcomes. Our study intended to answer the following questions: 1) what information collected from Accuplacer, Noel-Levitz College Student Inventory, and student background can be mined to identify characteristics of at-risk students, and 2) can predictive models using learning analytics provide administrators with timely and adequate information to develop intervention strategies for at risk-students? Psychological predictor variables along with several student characteristics were found to be associated with at-risk students. We intend our study to be an example for how learning analytics research results lead to action plans for improving student retention.

Authors

• Jere Turner, Manchester Community College

• Hui-Ling Chen, Saint Anselm College