Using Text Mining to Analyze Student Reviews to Improve Retention: A Case Study in Historically Black Colleges and Universities (Poster 10)

• In Event: AERA Poster Session Five
  In Poster Session: Division D Section 2 Poster Session

Abstract

This study analyzes first-year students' online reviews of four-year historically black colleges and universities (HBCUs) to provide insights for improving the first-year student retention rates in HBCUs. Our study collected more than 13000 online student comments from 55 HBCUs. We used text mining analytical approaches, including Latent Dirichlet Allocation, n-grams, and sentiment analysis with deep learning models, to extract and identify institutional characteristics that were most frequently commented by students and students’ sentiments towards them. We ran the hierarchical multiple regression to determine how the topics and student sentiments are associated with student retention. Finally, based on the results, we formulated strategies and suggestions to improve the first-year student retention rates of four-year HBCUs.

Authors

• Tonghui Xu, University of Massachusetts - Lowell
  Presenting Author

• Xiaobai Li, University of Massachusetts - Lowell
  Non-Presenting Author

• Yan Wang, University of Massachusetts - Lowell
  Non-Presenting Author

• Hsien-Yuan Hsu, University of Massachusetts - Lowell
  Non-Presenting Author