Does it stick? - Investigating long-term retention of conceptual knowledge in mechanics instruction

Abstract

By administering the Concept Assessment Tool for Statics (CATS) as a retest to engineering students and graduates, the retention of basic concepts in mechanical engineering is explored. Results from the retest from a sample population are compared to results from a larger population of posttests, i. e., tests after all relevant instruction. The sample is a subset of the population for which each member's posttest result is known. The sample population of 268 individuals is analyzed and grouped into three sub-populations based on the time interval between the posttest and retest. Overall we find that normalized differences between the sample retest and posttest is positive, showing a gain of understanding since the posttest at all retention intervals. It is hypothesized that gains in the retest relative to the posttest are a structural artifact of posttest being administered at the end of the course, but well before the exam period, an interval of sometimes up to eight weeks that is usually accompanied by intensive preparation and review of the course material. There is some evidence of forgetting when retention intervals are compared with one another. Sample members who remain actively engaged in the subject matter (as revealed by survey questions administered with the posttest) actually show learning gains and not only retention as a function of the retention interval.