Comparative analysis of multicriteria decision making methods for postgraduate student selection

Abstract

Problem Statement: The educational recourses available for postgraduate education are limited compared with the resources for undergraduate education. An academic advisor is required for each graduate student. The academic staff allocates an important portion of his/her time for graduate studies of his/her students. Since the number of academic staff is limited, the number of graduate students should be limited, too. While a large number of students may attend an undergraduate course, this may not be possible for some graduate courses, which are designed especially for small student groups. Hence, the selection of postgraduate students is an important problem. The postgraduate student selection problem can be defined as selecting a subset of students from the applicants for a postgraduate program. Naturally, this problem is a multi-criteria decision-making problem since each applicant has several attributes and these attributes should be considered simultaneously during the selection process. Purpose of Study: The purpose of this study is to compare the performance of different multi criteria decision-making methods developed for ranking alternatives for the postgraduate student selection problem. The best method will also be used in the student selection process for the Defense Sciences Institute of Turkish Military Academy. Methods: This paper discusses three multi-criteria decision making methods developed for ranking alternatives, namely Analytic Hierarchy Process (AHP), Weighted Product (WP) and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). A performance measure is defined and the performance values of the methods are compared using real data gathered from the graduates of Defense Sciences Institute of Turkish Military Academy for illustration purposes. Findings and Results: According to the findings, AHP is the best of the three competitors. The ranking obtained by AHP is quite similar to the ranking of students with respect to their cumulative grade point average (CGPA) after their graduation from the program. Conclusions and Recommendations: Considering graduate student selection problem as a multicriteria decision-making problem is very important and enables universities to enroll graduate students with important attributes and a variety of strengths. This paper compares the performance of different multicriteria decision-making methods on graduate student selection problem using a single performance measure. A further study can be making such a comparison with multiple performance measures