Predicting First Year Student Attrition in Higher Education: A Design Theoretic Approach
Abstract
It is known that retaining customers is more cost effective than attracting new customers. Retaining students in higher education mirrors this business scenario. Multiple academic journals have examined student attrition and... [ view full abstract ]
It is known that retaining customers is more cost effective than attracting new customers. Retaining students in higher education mirrors this business scenario. Multiple academic journals have examined student attrition and retention through various lenses ranging from demographic and academic factors to qualitative studies presenting grounded theory of understanding student retention. There remains an unexplored area in student retention: designing and modeling student retention through advanced analytics to predict student attrition a priori. The fields of machine learning and computational statistics provide the ability to sort and classify huge amounts of data in real time; enabling predictive reporting to key stakeholders of students that are at risk of dropping out. Predictive reporting allows education and advising specialists provide interventions for students are at risk. This paper proposes a design theoretic process for predicting student retention in real time and examines post facto a dataset with this process.
Authors
-
Caleb Bradberry
(Radford University)
-
Andrew Ray
(Radford University)
-
Jeff Pittges
(Radford University)
Topic Area
Topics: Information Technology, Decision Support Systems, and Cybersecurity - click here w
Session
IT3 » IT Issues - I (11:30 - Thursday, 5th October, West A)
Paper
Predictions_first_year_students__SEINFORMS_2017_revised.pdf
Presentation Files
The presenter has not uploaded any presentation files.