Malia B Hardy, Sydney L Kostal, Ashna Nayak, Michelle L Rivers
Multiple-choice tests are efficient and objectively scored, but they provide limited insights into the reasoning underlying students' responses. Requiring students to justify their answers may encourage more generative processing while also providing diagnostic information about their understanding. To examine this possibility, the present research investigated whether answer justification influences multiple-choice performance, confidence, confidence-accuracy resolution, and answer revision. In two experiments, undergraduate students studied an introductory geology text before completing multiple-choice assessments. Specifically, in Experiment 1, students completed an initial set of items, later reconsidered the same items, and then responded to new, conceptually related items. During the reconsideration phase, students either explained their reasoning before rating their confidence or simply re-answered and rated their confidence. In Experiment 2, a larger sample completed all items within a single test while changes in selected responses were unobtrusively recorded. Across both experiments, we did not find sufficient evidence demonstrating that requiring justification improved performance, confidence, confidence-accuracy resolution, or revision likelihood under the conditions examined. Nevertheless, revisions produced a positive net benefit: wrong-to-right changes occurred more frequently than right-to-wrong changes, and greater confidence was associated with a lower likelihood of revising. Moreover, exploratory analyses indicated that justification content was associated with accuracy and confidence. Conceptual-understanding justifications showed the strongest association with accuracy, and several non-guessing justification types were associated with greater confidence than guessing. Although correlational, these patterns suggest that written justifications may provide useful diagnostic information about students' reasoning.