The purpose of the study was to develop a criterion-referenced test to measure student's achievement in educational evaluation using item response theory. To achieve this goal, the author constructed a 3-option multiple-choice achievement test of 48 items that was later administered to 348 students enrolled at the University of Bahrain. The findings of study revealed that the students' responses to 31 items fit the Rasch model assumptions while 17 items did not fit the model. All items of the final version of the test, however, were located within the range of the model's infit and outfit indicators. Also, the reliability estimates for persons and items were .87 and .93, respectively, indicating a high reliability of the test, and the maximum information extracted from the three-option test is obtained at the average ability levels. Based on these results, the author recommends using the developed test as a reliable measure of the level of university student's achievement in the subject of educational evaluation.

The purpose of the study was to develop a criterion-referenced test to measure student's achievement in educational evaluation using item response theory. To achieve this goal, the author constructed a 3-option multiple-choice achievement test of 48 items that was later administered to 348 students ...