Co-enrollment Density Predicts Engineering Students’ Persistence and Graduation: College Networks and Logistic Regression Analysis
Studies in educational evaluation(2021)
摘要
College retention is a concern for educational institutions and researchers. This concern is particularly acute in engineering for reasons including workforce shortages, economic competitiveness, social justice, and socioeconomic equity. This study presents the evaluation of co-enrollment density (CeD) for engineering students at eight medium and large American public universities over 24 years. CeD is a novel metric estimated using enrollment records that may predict retention in 4-year bachelor of science programs in engineering. Graduation and persistence were fitted to CeD with logistic regression. Students in denser co-enrollment clusters-high CeD-tend to graduate more than their classmates in less dense neighborhoods-low CeD. The regression models predict graduation with odds ratio intervals 95 % CIs [3.24, 4.81] and area under the receiver operating curve [0.76, 0.80]. CeD is more sensitive to students who do not persist, particularly after the first year, so CeD's cut-off points may be indicators for dropouts' risk.
更多查看译文
关键词
Students persistence,Logistic regression,Engineering education,Receiver operating characteristic curves,Network analysis
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要