Predicting Community College Institutional Researchers' Perceptions of Machine Learning: A Mixed-Methods Approach

• In Event: Measurement and Assessment in Higher Education SIG (#64) Roundtable Session 1

Abstract

In the spring of 2021, community colleges (CCs) experienced a 9.5% decline in enrollment, continuing a multiyear trend that has made student retention and success a concern for CC leadership. There is interest in using tools to predict and improve enrollment and student outcomes. This study employs the results of a qualitative study on impressions of machine learning of CC Institutional Researchers (IRs) to develop a quantitative instrument to measure IR work activities, desire to use machine learning (ML), desire to learn ML, and barriers to the use of ML in a national sample of CC IRs. A structural equation model will be used to see which factors predict IRs’ desire to use or learn ML.

Authors

• Mark Andrew Perkins, University of Wyoming

• Michael Wade McDaniel, University of Wyoming

• Kristi Maria Grooms, University of Wyoming

• Craig Vaske, University of Wyoming

• W. Burt Reynolds, University of Wyoming

• Jonathan W Carrier, University of Wyoming

• Reed Scull, University of Wyoming