Factors Affecting the Dropout Students Using Discriminant Analysis and Association Rule

Currently, the educational institutions face a lot of issues, such as high dropout rate, predicting the quality of student interaction, student's performance evaluation and the placement of the students. To overcome these issues of educational institutions, data mining techniques are used. One of the biggest challenges that higher education faces is scaling down the rate of dropout students. The aim of this paper was to select the subset of relevant features from a large set of features using the statistical tool Statistical Package for the Social Sciences (SPSS), and to find frequent item sets and to extract the hidden information from the large data regarding the factors that are responsible for student dropout using the Waikato Environment for Knowledge Analysis (WEKA) tool. This paper proposes a novel concept of discriminant analysis used for analysing the affects of 33 variables on the student dropout in higher education. It also indicates which variables are important in explaining a dropout student and the Apriori algorithm in mining association rule from a dataset containing dropout student data concerning women. For this study, data of the first year undergraduate students were collected randomly from a survey based on personal interview at the university. The generated knowledge will be quite useful for understanding the problem in a better way and to have a proper planning or decision to scale down the dropout rate.