Practical Guidance for Writing Multiple-Choice Test Questions in Introductory Analytics Courses

Abstract

Test writing is a fundamental component of teaching. With increasing pressure to teach larger groups of students, conduct formal assessment of learning outcomes, and offer online and hybrid classes, there is a need for alternatives to constructed response problem-solving test questions. We believe that appropriate use of multiple-choice (MC) questions can alleviate some of these pressures. However, the results will only be acceptable to dedicated faculty if these questions are well written and effective in measuring student learning. In this article, we propose a structured framework consisting of six design principles that serve as a guide for writing MC items that promote higher-order thinking through the use of scenario-based and multistep problems. Then we demonstrate how to apply these principles to develop high-quality test questions in MC format for introductory analytics courses in statistics or operations research/management science.