مرکز منطقه ای اطلاع رساني علوم و فناوري - Investigation of Item Bias in 2014- November Transition Exam from Primary to Secondary Education In Terms Of Gender

Abstract :

Tests and items are expected not to show different characteristics in terms of the answers of individuals who are at the same ability level but from different groups. The objective of this study was to indicate whether Turkish, math, science and technology and Turkish Republic History and Atatürk s courses put forth any difference due to the gender of the individuals or not in TEOG examination. The study was based on a descriptive model, as the objective was to describe a situation as it exists. The research includes all students who took part in the TEOG examination in November in 2014. The sample of the research consists of 40000 students that were chosen via random sampling method: 52.14% percentage of which consists of male and 47.86 of female. It was determined whether the items contains Differential Item Function (DIF) or not using the Rasch model and via the Winstep package software. Prior to determining DIF identification, it was examined whether Turkish, math, Revolution History and Kemalism and science and technology courses provide unidimensionality and local independency or not in order that the analysis processes can fulfill the requirements of the Rasch model. Following the acceptance of unidimensionality 12 items in Turkish subtests, 20 in math subtests, 18 in science and technology subtests and 14 in Revolution History and Kemalism subtest were subject to DIF analysis making up a total of 64 items. In the analysis for the subtests, no contrast value, that is suitable for the DIF, could be found. In other words, there couldn’t be found any item that shows any different functions in terms of gender in subtests.. This result indicates that there is no item that works for the behalf of or to the detriment of students from different gender groups.