ÇIRAK, Gülçin Ankara Üniversitesi - EBF - Ölçme ve Değerlendirme ABD, Turkey , ÇOKLUK, Ömay Ankara Üniversitesi - EBF - Ölçme ve Değerlendirme ABD, Turkey

The Usage of Artifical Neural Network and Logistic Regresssion Methods in the Classification of Student Achievement in Higher Education

22435

In this study, according to the results of the survey conducted by the researcher entitled Variables Which Affect the Success of University Students” artificial neural networks and logistic regression methods’ performances were compared to indicate the estimated future success of the students who are registered to different faculties and programs. The overall academic grades of the students are taken as the dependent variable. The general academic achievement grade average is a permanent variable, it turns out to be a discontinuous variable. The group which was studied with the research which is a relationally browsing model consists of 419 students who were at their 3rd year in the 2011-2012 education and teaching year. This group were students of Ankara University in the departments of educational sciences and Language and History-Geography. According to this study, neural network analysis has a higher right classification probability when compared to logistic regression analysis.

71

Student Achievement , Classification , Prediction , Artifical Neural Network , Logistic Regression

Mediterranean Journal Of Humanities

79