Students don’t like them—that almost goes without saying. They prefer unit exams that include only material covered since the previous exam. And they’d like it even better if the final wasn’t a comprehensive exam but one last unit test. But students don’t always prefer what research shows promotes learning and long-term retention, and that is the case with this study of the effects of cumulative exams in an introductory psychology course.
Exam performance of students in two sections of the course was analyzed. In the noncumulative section, students took three 50-question multiple-choice exams and a comprehensive final in which 55 percent of the questions covered material from across the course. In the cumulative section, students took three 50-question multiple-choice exams, but the second and third exams both included 10 questions covering material from the previous exams. Students in this section took the same final as those in the noncumulative section. Additionally, students in both sections took a follow-up exam two months after the course ended. That exam included 50 multiple-choice questions covering course content, and they were not questions that appeared on the exams taken during the course.
Students were also surveyed about their exam preparation, studying, and preferences. As predicted, those taking the noncumulative section were happier with the exam format than those in the cumulative section. Other than that there were no significant differences in the students’ perceptions of things such as exam difficulty, their study methods, or the number of hours they reported they spent studying for the exams.
But there were differences in their performance on the final, on chapter quizzes, and in their overall course grade. All these differences favored students in the section with cumulative exams.
The researcher also divided students in each section into low- and high-scoring groups based on the scores earned in the first exam. (College GPA could not be used to group the students as most of them were first-semester college students.) These groupings revealed some of the most interesting findings from this study. For students in the high-scoring group, quiz grades were unaffected by section, but students in the low-scoring group of the cumulative section did better on the quizzes than did the low-scoring students in the noncumulative section. This difference was statistically and academically significant. Low-scoring students in the cumulative section averaged a B quiz grade and those in the noncumulative section a C quiz grade. The same effect was seen in final course grades, with those in the cumulative section ending up with a B average and those in the noncumulative section with a C+ average. It emerged again in scores on the exam taken two months after the course ended. High-scoring students’ long-term retention was not affected by the experimental manipulation, but low-scoring students remembered more if they had the four cumulative exams.
The researcher explains the results this way: “Most likely, having multiple cumulative exams motivates low-scoring students to engage in behaviors that promote better performance and long-term retention. High-scoring students probably already have the motivation to engage in these types of behaviors.” (p. 18)
Is it worth risking student disfavor by giving cumulative exams? If those exams promote long-term retention (and this isn’t the only research work supporting that finding), the risk is worth taking. Students can be told about this research, content covered previously can be regularly mentioned in light of current content, and teachers (or high-scoring classmates) can work with students on those study strategies that effectively prepare them for cumulative questions.
Reference: Lawrence, N. K. (2013). Cumulative exams in the introductory psychology course. Teaching Psychology, 40 (1), 15–19.
