Identifying Institutional Peers Through Cluster Analysis

Abstract

In an era of evidence-based decision-making, institutional researchers utilize benchmarking as a means of evaluating and improving university performance. University leadership use benchmarking metrics like six-year graduation rates to compare their performance to their peers, identify leading national institutions, and to discover best practices towards achieving institutional goals. In this session, I discuss how to identify peer institutions for the University of Houston using IPEDS data. I employ cluster analysis to calculate distance measures and then group institutions based on their institutional characteristics. At the end of this session, attendees will understand how to identify appropriate metrics, how to execute clustering methods, how to evaluate their analyses, and how to visualize cluster groups to communicate their findings.

Date
Mar 2, 2020 3:30 PM — 4:15 PM
Location
Marriott Plaza San Antonio
555 S. Alamo St, San Antonio, TX 78205

Dataset codebooks: Institution Data, Faculty Data

Avatar
Jorge Martinez
Director, Data Science

Director of Data Science at Houston Independent School District.