Abstract

Evaluation of students' academic achievement is an unavoidable task for all faculty members of universities. The majority of academic institutions use multiple choice exams to evaluate their students. The question items that make up these exams need to possess certain psychometric properties to measure the student achievements. The purpose of this research was to examine the item properties of multiple choice exams used in the college of health of Kashan University of Medical Sciences. In this cross sectional descriptive study, a number of multiple choice exams used by the instructors were randomly selected. The difficulty index, discrimination index, and Cronbach alpha was calculated for every exam by using LERTAP 5.0 software prepared by the Assessment Systems Corporation of the United States to perform an item analysis. A total of 1343 multiple choice question items presented to the students at the college of health for evaluating 37 different subjects was analyzed. The results of analysis showed that the mean values of difficulty index and discriminating index were0.68 and 0.20, respectively. The average value of difficulty)0.68(and discrimination index)0.21(were relatively within the recommended values. However, some of the tests had value less than the recommended range and need careful reexamination. It was concluded that a feedback delivered to the instructors may improve these indexes. Further research is needed to examine this subject.