A Data Mining Approach for Identifying Predictors of Student Retention from Sophomore to Junior Year

Abstract

Student retention is an important issue for all university policy makers due to the potential negative impact on the image of the university and the career path of the dropouts. Although this issue has been thoroughly studied by many institutional researchers using parametric techniques, such as regression analysis and logit modeling, this article attempts to bring in a new perspective by exploring the issue with the use of three data mining techniques, namely, classification trees, multivariate adaptive regression splines (MARS), and neural networks. Data mining procedures identify transferred hours, residency, and ethnicity as crucial factors to retention. Carrying transferred hours into the university implies that the students have taken college level classes somewhere else, suggesting that they are more academically prepared for university study than those who have no transferred hours. Although residency was found to be a crucial predictor to retention, one should not go too far as to interpret this finding that retention is affected by proximity to the university location. Instead, this is a typical example of Simpson's Paradox. The geographical information system analysis indicates that non-residents from the east coast tend to be more persistent in enrollment than their west coast schoolmates.

How Does Learning Analytics Contribute to Prevent Students’ Dropout in Higher Education: A Systematic Literature Review

“…Research on this topic began in the 1970s and 1980s [37][38][39][40]; these approaches are still the basis for developing new solutions [41][42][43][44][45]. However, it is necessary to note that dropout can happen throughout the academic course (from the first to the second year [46], and from the second to the third year [47]). Therefore, the risk factors may be differentiated.…”

Educational data mining to support identification and prevention of academic retention and dropout: a case study in introductory programming

“…The application of data mining in the educational context is widely explored in the EDM literature (Romero & Ventura, 2020), especially to predict students performance (Salloum, Alshurideh, Elnagar, & Shaalan, 2020) and to understand the factors that most impact in students success/failure. Most of techniques employ data regarding students economic and demographic information (Yu, DiGangi, Jannasch-Pennell, & Kaprolet, 2010), as well as examination results and course enrolment (Li, Ding, & Liu, 2020;Adekitan & Salau, 2019).…”

Development of a Predictive Model of Student Attrition Rate

“…Meanwhile, several international studies have been conducted. In [8], explores student retention probability based on freshmen' background data, including demographic, academic, and course participation data. Methodologically, they find that, transferred hours, residency, and background as critical retention determinants utilizing multivariate adaptive regression splines, neural networks, and classification trees.…”

“…Attrition among final-year students has significant effects on both the people and the affected institutions in today's educational system. Attrition results in costs for all parties, whether they are in terms of resources, time, or money [1], [2]. As a result, institutions of higher education face a significant challenge in preventing educational attrition [3].…”