Medical Education Multiple choice questions: A literature review on the optimal number of options RASHMI VYAS, AVINASH SUPE ABSTRACT Background. Single, best response, multiple choice questions (MCQs) with 4 options (3 distractors and 1 correct answer) or 5 options (4 distractors) have been widely used as an assessment tool in medical education in India and globally. Writing plausible distractors is time consuming and the most difficult part of preparing MCQs. If the number of options can be reduced to 3, it will make preparing MCQs less difficult and time consuming, thus reducing the likelihood of flaws in writing MCQs. We reviewed the literature to find out if the number of options in MCQ test items could be reduced to 3 without affecting the quality of the test. Methods. A systematic database search was done using the following question as a framework: How many options are optimal for multiple choice questions? Theoretical, analytical and empirical studies, which addressed this issue, were included in the review. Results. There was no significant change in the psychometric properties of the 3 options test when compared with 4 and 5 options. MCQs with 3 options improved the efficiency of the test as well as its administration compared with 4- or 5-option MCQs. MCQs with 3 options had a higher efficiency because fewer distractors needed to be prepared, took up less space and required less reading time, decreased the time required to develop the items and the time to administer, and more items could be administered in a given time thus increasing the content sampled. Conclusion. Our review of the literature suggests that MCQs with 3 options provide a similar quality of test as that with 4- or 5-option MCQs. We suggest that MCQs with 3 options should be preferred. Natl Med J India 2008;21:130–3 INTRODUCTION Multiple choice questions (MCQs) are used globally as test items for assessment in various fields of education. 1–4 In India, MCQs have been commonly used for undergraduate and postgraduate medical entrance and university examinations. 5–8 These are also used as an assessment tool for paramedical courses and formative assessment in the medical curriculum. 9–11 The main challenge in preparing MCQs is to construct good test items. 12–15 This requires a good knowledge of the content and understanding of the objectives of assessment as well as good skills in writing the items. 15–17 Though there are many guidelines for writing good test items, 18–21 these are usually not adhered to; leading to the preparation and administration of faulty MCQs. 8,22,23 The single, best-response type MCQ (the format that asks the student to choose the best answer from a list of possible answers) with 4 or 5 options is the most commonly used format. 14,22,24–28 For a distractor (wrong option) to be useful, it should represent a common misconception among students about the correct answer. 26 Writing plausible distractors is time consuming and the most difficult part of preparing MCQs. 17,26,29,30 Some of the main flaws in writing distrators include implausible distractors, more than one or no correct answer, the longest option being correct, and the use of ‘all of the above’ and ‘none of the above’ options. 22,23 Lack of training of the faculty in preparing MCQs and lack of time have been identified as important reasons that contribute to flaws in writing distrators/items. 1,8 Training faculty in the skills required for writing items and reducing the number of distractors in an MCQ could be some possible remedial measures. We hypothesized that if the number of options in an MCQ can be reduced to 3 from the commonly used 4 or 5, it would make MCQ writing less difficult and less time consuming, thus reducing the likelihood of flaws in writing MCQs. We, therefore, reviewed the existing literature to determine if the number of options in an MCQ can be reduced to 3 without affecting the quality of the test. METHODS A systematic database search was done using Science Direct, Pubmed, Ovid and ERIC search engines for the period 1960– 2007. Specific articles were also looked for in the Sage publications, JStor and Blackwell Synergy electronic resources. The articles were searched using the following question as a framework: How many options are optimal for multiple choice questions? The search terms included the keywords: MCQ, multiple choice questions, optimal distractors, optimal options, number of options, number of distractors, one best response, item analysis and item writing guidelines. All studies that assessed the 3-option format were included in the review. Studies not available in electronic format were excluded from the review. The initial results were screened for pertinence, which yielded 15 articles for reading. A review of the reference lists identified 8 additional articles of interest. RESULTS Studies on the optimal number of options were grouped into (i) theoretical and analytical studies, and (ii) empirical studies, © The National Medical Journal of India 2008 THE NATIONAL MEDICAL JOURNAL OF INDIA VOL. 21, NO. 3, 2008 130 Christian Medical College, Vellore 632002, Tamil Nadu, India RASHMI VYAS Department of Physiology Seth G.S. Medical College and K.E.M. Hospital, Parel, Mumbai 400012, Maharashtra, India AVINASH SUPE Department of Surgical Gastroenterology Correspondence to RASHMI VYAS; rashmifvyas@yahoo.co.in; rashmifvyas@gmail.com