Supporting comprehension: The advantages of multiple-choice over true-false practice tests.

Abstract

While work on improving comprehension has primarily focused on open-ended generative activities, closed-ended practice tests using inference-type questions may also benefit understanding from text. Four experiments were designed to investigate how practice tests, specifically in multiple-choice and true-false formats, may support comprehension. Experiments 1 and 2 compared the two practice test formats to rereading. Both formats improved performance on a final essay test in Experiment 1, but in Experiment 2, only multiple-choice practice enhanced performance on a short-answer (SA) test. Experiment 3 introduced feedback on practice tests, but found no added benefit on the final SA test, which remained consistently better for those who completed the multiple-choice as compared with the true-false version of the practice test. Finally, manipulating text availability during practice tests in Experiment 4 improved performance on the final SA test. However, multiple-choice practice consistently led to better SA performance than true-false, regardless of text availability. The present work illustrates that the benefits from a closed-ended practice test with multiple-choice questions can persist over a delay and transfer to a set of new comprehension questions. At the same time, the results also highlight important constraints in that subtle nuances in question design can impact the observed benefits of practice testing on learning outcomes.