Identifying the Most At-Risk First Time in College Students
presented by Bernadette Jungblut, Ronald Atwell, University of Central Florida
Date of webinar: February 10, 2010
|
*Materials for this webinar are not currently available.
REGISTRATION INFORMATION
To register for this webinar download this document, fill it out, and either mail or fax it to the CSRDE.
|
Abstract
Data-mining and logistic regression techniques are used to predict the First Time in College students most likely to be at risk for retention failure. Using incoming students' academic preparation and demographic characteristics, we predict and test their likely achievement gaps in terms of retention, probation status, and term GPA. Factors employed to make these predictions (based on a risk score generated for each incoming student) include the following: financial aid, scholarships, and grants; high school overall GPA, class rank, class size, and percentile; SAT total and component scores; ACT total and component scores; high school English, math, science, and social science units and GPA earned; ethnicity, gender, and intended university academic program/major. The results have been used to identify which students most need the Knight Success Program; to design that program; and to improve the retention and persistence of the program participants. Over the past four (4) years, we have reassessed this program's performance and undertaken additional data analyses to inform decisions taken to improve the program's design, implementation, and effectiveness.
|
|