A Study of Factors Influency Student Dropout Rate Using Data Mining

Abstract

ABSTRACT

The objective of this research is to study factors affecting student dropout rate using data mining. The information of students in the Department of Computer, Faculty of Science and Technology, Chiang Mai Rajabhat University was used as data in this study. The data were divided into two sets. The first set was obtained from former institution containing 1,433 records of data. This data set was analyzed with classification data-mining technique based on decision tree and C4.5 algorithm. The analysis results showed that the top three factors that have impact on dropout of the students are previous study result (GPA), previous study field, and educational background, with accuracy values of 72.02, 70.11, and 68.13, respectively. The second data set contains 2,568 records of current study results. This set was used to determine courses that affect dropout of the students based on association rules data mining technique and Apriori algorithm. The courses found to have impact on the students' dropout were Computer, English, Mathematics and Physics. Results from this research can be useful for developing a prediction model of student dropout. In addition, the research findings can provide some guidelines for universities to solve the dropout problem of their students.