A comparison of psychometric properties of multiple choise and mixed tests prepared to measure the same cognitive feature

This research was conducted in two forms: multiple choise test and mixed test. This study was carried out to determine whether there is a significant difference between the item difficulty indices, item discrimination indices, test averages, test reliabilities and response times obtained by applying the "8th Grade Mathematics Achievement Test" and "8th Grade Science Achievement Test" online in a computer environment. These two forms are form multiple choice test, which was created using only the multiple choice item type, prepared to measure the same characteristics; The other one was prepared as form mixed test, using multiple choice, true/false, matching and short answer item types with more than one correct answer. The measurement tools were prepared and applied online on "testmoz.com", one of the Web 2.0 tools. The research is in the survey research model. The data was collected from the study group consisting of 173 eighth grade students studying in three private secondary schools selected from Yenimahalle and Keçiören districts within the borders of Ankara. As a result of the research, a significant difference was determined in favor of form multiple choice test in both tests between the item difficulty indexes of multiple-choice test items and test items requiring multiple choice, true/false, matching and short answers with more than one correct answer and form multiple choice test was found to be easier. No significant difference was found for the item discrimination power indices obtained from the multiple choice test and mixed test forms of the mathematics and science achievement tests. While there was no significant difference between the forms of the mathematics achievement test in terms of test reliability, a significant difference was found at the p <.05 level in favor of form multiple choice test in the science achievement test. While there was no significant difference in the response times between the forms of the mathematics achievement test, a significant difference was found at the level of p <.01 in favor of form mixed test in the science achievement test.