Multiple–True–False Questions Reveal the Limits of the Multiple–Choice Format for Detecting Students with Incomplete Understandings

By having students select one answer among several plausible options, multiple–choice (MC) questions capture a student's preferred answer but provide little information regarding a student's thinking on the remaining options. We conducted a crossover design experiment in which similar groups of introductory biology students were assigned verbatim questions in the MC format or multiple–true–false (MTF) format, which requires students to separately evaluate each option as either true or false. Our data reveal that nearly half of the students who select the correct MC answer likely hold incorrect understandings of the other options and that the selection rates for individual MC options provide inaccurate estimations of how many students separately endorse each option. These results suggest that MC questions systematically overestimate question mastery and underestimate the prevalence of mixed and partial conceptions, whereas MTF questions enable students and instructors to gain a more nuanced portrait of student thinking with little additional effort.