The Comparison of Psychometric Properties of Standardised and Make Up Maths and Turkish Subtest Questions in TEOG

Abstract

The aim of this study is to compare the psychometric properties of Turkish and math subtests in common and make-up examination of TEOG applied in 2013-2014 school year. Turkish and math tests of the common and make-up examinations were investigated comparatively within the frame of psychometric properties such as item difficulty, item discrimination index, test reliability and validity. Research was carried out via item scores matrix of 9773 students for Turkish common exam, 9485 students for math common exam, 2747 students for Turkish make-up exam and 2323 students for math make-up exam. It was consulted to expert opinion for the content validity study. According to this, in order to determine and compare the acquisitions that is measured in common and make-up examinations of TEOG, question-acquisition matching form and question assessment form was prepared by researcher. As a result of research, it has been observed that Turkish and math subtests test and item statistics are different in significantly. Content validity which based on expert opinions is in similar level for Turkish and math subtests. The results of the study revealed that math tests of the common and make-up exams items didn't measure the same things in terms of construct validity. Unlike math, Turkish tests of the common and make-up exams items measure the same things in terms of construct validity.