Investigation of Psychometric Properties of Multiple-Choice Items Developed By Turkish Teachers

Abstract

The main purpose of this research is to determine the psychometric properties of the items (questions) developed by Turkish teachers. The research was conducted with 320 teachers who teach Turkish to 8th grade students in public schools in Ankara in during the 2020-2021 academic year. Turkish subtest items of LGS (High School Enterance System) central exam verbal section were evaluated by field experts and a blueprint was created. Each achievement in the blueprint was randomly assigned to a teacher and he/she was asked to develop a multiple-choice item which has moderate difficulty and high discrimination to measure this achievement. The multiple-choice items written by the teachers were applied to the 8th grade students and the psychometric properties of the items were estimated from the students’ responses. According to the results of the research, it was found that the content validity ratios, reliability and validity coefficients, item factor loadings, and differential item function values of the items developed by the teachers were at an insufficient level. Item difficulties were easy and the discrimination indexes were at a level that could be taken to the test by only correction.