Beliefs regarding the usefulness of true-false tests are mixed. Many of these opinions stem from research comparing true-false test performance to that on traditional paper-and-pencil tests. However, little is known about how true-false test scores relate to performance measures requiring knowledge application, or whether different scoring algorithms vary in their ability to predict such performance. To address these gaps, we examined the relationships between traditional and modified true-false scoring methods and outcomes on a business simulation designed to assess complex knowledge application. Our results showed that posttest true-false scores were associated with simulation performance, with the gap between high and low scorers widening over time. Scoring formats that incorporated confidence ratings demonstrated higher reliability and predictive power, but were not substantially more correlated with performance than traditional methods. These findings suggest that true-false tests can serve as effective measures of performance on complex tasks.
Copyrights © 2026