Abstract:

This study aimed to determine to what level of accuracy can the variables such as pre-service teachers’ General Mathematics, Pure Mathematics, Analysis I, Analysis II, Geometry, Linear Algebra-I, Analysis3, Special Teaching Methods 2, Elementary Number Theory, Algebra, Problem Solving variables which level of classify the students’ performance in a graduate education achievements. The purpose of this research is to be able to make an effective prediction regarding the students’ success in post-graduate education with Logistic Regression Analysis (LRA) which is used as an effective prediction method in various sectors, as an alternative to traditional methods in the field of education. Relational screening model was employed in this study. 139 primary mathematics pre-service teachers who were selected randomly among students who are studying in or graduated from Marmara University, Educational Sciences Institute and Marmara University, Atatürk Educational Faculty and constitute the sample of this study. Logistic regression analysis technique was used because dependent variable had categorical structure. The results of the investigation showed that eleven variables included, were statistically significant. Estimates of teacher candidates' academic achievement in the process of performing undergraduate students received passing grades for the courses during their training, have been considered and the data of the Marmara University Faculty of Education students of Primary School Mathematics teacher used as predictive variables, tried to be prescribed by the achievements of the students for graduate education It was observed that with the given logistic regression model the rate of correct classification was 92%. The findings of the study reveal that 89,7% of the students who were successful in post-graduate education achievements, and 93,8% of students who were not successful was estimated with the correct classification.

Keywords: