Test Format, Learning Confidence, and Perceptions of Teaching Effectiveness

Abstract

Both multiple-choice and short-answer tests can be beneficial to learning in the classroom. However, fact-based multiple-choice questions, because they include the correct answer as a response option, could lead to inflated estimates of learning and higher evaluations of teaching effectiveness. The objective of this study was to examine the effects of test format on perceptions of learning and teaching. Undergraduate students ( N = 123) completed either a multiple-choice or short-answer test based on a brief psychology lesson. Then, they predicted their performance on later tests of knowledge and evaluated the quality of the lesson. Taking a multiple-choice test led to predictions about future performance that were 10% higher than those taking a short-answer test. No consistent differences emerged in participants’ evaluations of teaching effectiveness between the two test formats. These results suggest that test format may significantly influence students’ confidence in their ability to produce answers on future tests. Teachers who want to take advantage of testing as a metacognitive learning tool should adopt question formats that utilize recall of information from memory without retrieval cues.