Impact on Students
In the redesign, did students learn more, less or the same compared to the traditional format?
The redesign produced little change in course grades. In the eight semesters immediately preceding the change to the redesigned course, the grade point average was 2.39 on a 4-point scale. The average for the first eight semesters of the redesign was 2.42—a difference that is not statistically significant. Relative grade outcomes by gender, ethnicity, and academic ability (self-reported or judged by high school grades and SAT scores) showed no clear effects from the switch to the redesigned format.
Overall performance on the common final exam appears to have improved. The average score was 63.5% in the two semesters immediately preceding the redesign and was 66.3% in the redesign years (68.1% if the first semester of the redesigned course is omitted). An analysis controlling for SAT scores and high school grades confirms this trend. Unfortunately, changes in the exams, including variations in the number of questions, cast some doubt on these results. The redesign had no discernible effect on students’ relative performance on final exams when classified by gender, ethnicity, and degree of academic challenge.
When common exam scores are broken down by topical area, there is one clear change: performance on questions concerning eigenvalues and eigenvectors declined in the redesigned course. This bears out course personnel’s general observation that the new system, as currently implemented, appears to work less well for conceptually difficult material. This was always a difficult topic in the traditional course, and it comes at the very end of the semester, but the results have certainly not improved with the redesign. This is an obvious target for improvement.
Other Impacts on Students
The percentage of students completing the course by achieving grades of D- or better improved, from an average of 80.5% in the two fall semesters immediately preceding the redesign to an average of 87.25% in the subsequent four fall semesters. Analysis by means of confidence intervals indicates that the improvement is statistically significant.
Program in Course Redesign Quick Links: