Skip to main content
Utah's Foremost Platform for Undergraduate Research Presentation
2021 Abstracts

A High Impact Engaged Demand Survey: Application of Data Mining in Student Retention Strategy Using Decision Tree Module

Presenters: Jiyeon Maeng, Business, organizational leadership
Authors: Jiyeon Maeng, Angie, Yang Huo
Faculty Advisor: Yang Huo, Business, organizational leadership
Institution: Utah Valley University

One of the biggest challenges that higher education faces is to improve student retention
 (Alhohani, 2016: Kerby, 2015: Mah, 2016: Lin, 2012: Tight, 2020). Student retention has become a critical indicator for an academic performance and enrolment management. The purpose of this study is to identify why drop out from the course/program, why they transfer or leave college or university and to follow up with intervention options and appropriate strategies to enhance student retention. We utilize the Qualtrics, web-based survey software, to create surveys, collect and store data, and produce the descriptive statistical analysis. We collect 201 sample from business management students at Utah Valley University register for spring and summer semester 2020. The collected data include demographics such as student study history, gender, academic level, GPA, marital status, ethnicity, total earned credit, job situation (full or part time) and institutional factors regarding the dropout. We apply the Decision Tree/Regression Tree using CART (Classification and Regression Trees) method as an analytic tool to identify a group, discover relationship between groups and predict future events and for segmentation, stratification, data reduction and variable screening, interaction identification, and category merging the paths from root to leaf represent classification rules. The results of CART regarding the demographics indicate that the most effective attributes are job situation, earned credit hour, ethnicity, and academic level. For the dropout reasons regarding the institutional factors show that the most effective attributes are financial situation (scholarship), quality of teaching, job related course, student accountability (personal issues: motivation, emotion). The decision tree module can help spot students ‘at risk’ in the context of major selection, course/class management by evaluating the course or module suitability, and tailor the interventions to increase student success and retention and improve student’s performance.