Does the Response Options Placement Provide Clues to the Correct Answers in Multiple-choice Tests? A Systematic Review

• Published in: Applied measurement in education Vol. 35; no. 2; pp. 133 - 152
• Main Authors: Lions, Séverin, Monsalve, Carlos, Dartnell, Pablo, Blanco, María Paz, Ortega, Gabriel, Lemarié, Julie
• Format: Journal Article
• Language: English
• Published: Philadelphia Routledge 03.04.2022; Taylor & Francis Ltd
• Subjects: Educational tests & measurements; Guidelines; High Stakes Tests; Item Response Theory; Multiple Choice Tests; Standardized Tests; Systematic review; Teacher Made Tests; Test Bias; Test Construction; Test Items
• ISSN: 0895-7347; 1532-4818
• DOI: 10.1080/08957347.2022.2067539

Multiple-choice tests are widely used in education, often for high-stakes assessment purposes. Consequently, these tests should be constructed following the highest standards. Many efforts have been undertaken to advance item-writing guidelines intended to improve tests. One important issue is the unwanted effects of the options' position on test outcomes. Any possible effects should be controlled through an adequate response options placement strategy. However, literature is not straightforward about how test developers arrange options. Therefore, this research synthesis systematically reviewed studies examining adherence to options placement guidelines. Relevant item features, such as the item source (standardized or teacher-made tests) and the number of options were considered. Results show that answer keys' distribution across tests is often biased, which might provide examinees with clues to select correct options. Findings also show that options are not always arranged in a "logical" fashion (numerically, alphabetically...) despite being suited to be so arranged. The reasons underlying non-adherence to options placement guidelines are discussed, as is the appropriateness of observed response options placement strategies. Suggestions are provided to help developers better arrange items options.