A Predictive Model For Benchmarking Academic Programs (pbap) Using U.S. News Ranking Data For Engineering Colleges Offering Graduate Programs

<p>Improving national ranking is an increasingly important issue for university administrators. While research has been conducted on performance measures in higher education, research designs have lacked a predictive quality. Studies on the U.S. News college rankings have provided insight into the methodology; however, none of them have provided a model to predict what change in variable values would likely cause an institution to improve its standing in the rankings. The purpose of this study was to develop a predictive model for benchmarking academic programs (pBAP) for engineering colleges. The 2005 U.S. News ranking data for graduate engineering programs were used to create a four-tier predictive model (pBAP). The pBAP model correctly classified 81.9% of the cases in their respective tier. To test the predictive accuracy of the pBAP model, the 2005 U.S .News data were entered into the pBAP variate developed using the 2004 U.S. News data. The model predicted that 88.9% of the institutions would remain in the same ranking tier in the 2005 U.S. News rankings (compared with 87.7% in the actual data), and 11.1% of the institutions would demonstrate tier movement (compared with an actual 12.3% movement in the actual data). The likelihood of improving an institution's standing in the rankings was greater when increasing the values of 3 of the 11 variables in the U.S. News model: peer assessment score, recruiter assessment score, and research expenditures.</p>

Grantor dc:publisher

University of Central Florida

Year

2005

Author dc:creator

• Chuck, Lisa Gay Marie

Contributors dc:contributor

• Tubbs, LeVester

• U.S. News rankings
• college rankings
• benchmarking
• predictive model
• classification discriminant analysis
• engineering college rankings
• Education
• Educational Leadership

Language dc:language

English

Identifier

CFE0000431

OAI identifier oai:identifier

oai:stars.library.ucf.edu:etd-1297