Measuring students’ learning gains with pre/post assessment

John R. Buck; Kathleen E. Wage

doi:10.1121/1.5067885

Pre/post protocols assess students with the same instrument at the start and end of a course to measure changes in their understanding. In contrast, traditional ”final exam” assessment measures only the students’ understanding but not their progress. Since pre/post testing measures learning gains, it is an essential tool for evaluating the effectiveness of different learning formats, e.g., standard lecture versus active learning. Concept inventories (CI's) are frequently administered as a pre/post assessment. A CI is a multiple-choice exam that emphasizes conceptual understanding over computational skills and has wrong answers designed to match common misconceptions. Inspired by Hake’s [1998] pioneering study of pre/post test data from the Force Concept Inventory [Hestenes et al., 1992], we developed the Signals and Systems Concept Inventory (SSCI). In almost two decades since its development, scores of instructors have administered the SSCI to thousands of students in dozens of countries. Analyzing pre/post data from sixty-two signals classes at multiple institutions, we found students in active learning classes learned significantly more than students in lecture classes. Our findings are consistent with both Hake’s pioneering study and with Freeman et al.’s 2014 meta-analysis of 225 studies that found that active learning classes reduced the failure rate by one-third.