Quality analysis of multiple choice questions(MCQs) examinations of noncontinuous undergraduate medical records

Introduction: There are Different methods for evaluating student stuff. One of the most commons is multiple choice questions (MCQs). If properly designed, it is a good way to measure student knowledge. Due to expansion using MCQs, This study was designed in order to review the quality of multiple choice question exam types of medical records students of Hormozgan University of Medical Sciences. Methods: In a descriptive and analytical study, all end of semester MCQ type examinations related to undergraduate degree of medical records course of Hormozgan University of Medical Sciences, including computer applications, English language, common classification systems, medical terminology 3, research methodology and data analysis of health indicators were selected. Then difficulty and discrimination index of all questions were calculated. For statistical analysis of the data, by taking advantage of Excel and SPSS software's, descriptive statistics, analysis of variance and regression methods were used. Results: The results showed that, from 128 questions, 105 questions (57.7%) had a suitable difficulty index, which the lowest rate was related to the research methods course (0.59) and the highest rate was related to the medical terminology3 (0.78). 133 questions (73.1%) were in weak and very weak discrimination index category, so that the lowest rate was related to English language (-0.05) and common systems (-0.03) and the most rate was related to the research methodology (0.23) and use of computer application (0.18). Results of analysis of variance showed a significant difference, between difficulty and discrimination index of six exams (P<0.05). There was no significant correlation between difficulty and discrimination indexes. Conclusion: Although, the difficulty index for all examinations were acceptable and appropriate but the results of discrimination index showed that, almost for all questions, the quality level is poor to very poor. So, it is better to remove these questions from the bank of questions or basically revise all the questions.