Analysis of Multiple Choice Questions Based On Classical Test Theory

The aim of this study was to use Classical Test Theory (CTT) software for qualitative and quantitative analysis of Multiple Choice Questions (MCQ) for Master students of health discipline. This study was carried out on two groups each including 50 undergraduate students of public health, and two groups of 50 and 26 undergraduate students of occupational health. The final exams were done to determine academic achievement of students using MCQ during the years 2004 and 2005 in the department of environmental health, school of health in Isfahan University of Medical Sciences. The questions were analyzed by CTT software in two phases before and after quantative analysis. Having qualitative analysis performed, the questions set for the first stage exam were modified. At the following year, another group of students were examined using those modified questions of the first stage. Reliability coefficient for two exams of industrial solid wastes and health of water, before the quantitative analysis of the questions, were obtained 0.576, and 0.359, respectively. After quantitative analysis of these questions, the coefficients values improved to 0.671 and 0.694, respectively. The method of question analysis in this study led to improving reliability coefficient in the second stage. This coefficient is still far away from desirable coefficient (reliability coefficient greater than 0.8). Therefore, these questions still need to be modified, and analyzed again using the CTT software.