Predicting Degree Completion through Data Mining
Abstract
Universities and colleges continuously strive to increase student retention and degree completion. The U.S. Department of Education has set the goal of preparing a society with individuals capable to “understand, explore and engage with the world” specific skills that can be achieved through STEM majors. Currently, considerable student data are collected and there is a latent opportunity to make the available information useful for determining the factors that influence retention and completion rates. Analyzing student data with those aims is vital for intentional student advising. To this end, this research presents the application of decision trees to predict degree completion within three years for STEM community college students. Decision trees also enable the identification of the factors that impact program completion using non-parametric models by classifying data using decision rules from the patterns learned. The model was developed using data on 283 students with 14 variables. The variables included age, gender, degree, and college GPA, among others. The results offer important insight into how to develop a more efficient and responsive system to support students.
Recommended Citation
T. Cardona et al., "Predicting Degree Completion through Data Mining," Proceedings of the ASEE Annual Conference & Exposition (2019, Tampa, FL), American Society for Engineering Education (ASEE), Jun 2019.
Meeting Name
ASEE Annual Conference & Exposition (2019: Jun. 16-19, Tampa, FL)
Department(s)
Engineering Management and Systems Engineering
Keywords and Phrases
Student Retention; Decision Trees; Degree Completion; Engineering Education
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2019 American Society for Engineering Education (ASEE), All rights reserved.
Publication Date
24 Jun 2019