December 3, 2008
Teaching Strategies: Frequent Exams = Better Results for Students
It’s not a new finding — in general, more exams lead to better grades—but it’s always nice when research confirms some of our best practices in teaching.
In the educational assessment study referenced below, the students were enrolled in two sections of an introductory statistics course for sociology majors. Both sections had the same instructor, same text, and same material presented in class. Students enrolled in each were similar in terms of gender and year in school. In the control section, students took two midterm exams (one at the end of the sixth week, the other at the end of the 12th week) and a two-hour cumulative final. In the experimental section students took an exam every other week starting at the end of the second week, for a total of six exams, plus the same cumulative final. Students were given one-third the amount of time for each of the biweekly exams.
As for the better results, students given the biweekly exams scored, on average, about 10 percentage points, or one letter grade, higher on the exams taken during the semester. They scored about 15 percentage points higher on the final than those students who only took the two midterm exams.
Measuring the Impact of Testing Frequency on Student Performance
There were some other persuasive results related to the impact of testing frequency on student performance. More than 11 percent of the students in the control section withdrew from the course. Not one student in the experimental section did.
Moreover, students in the experimental section evaluated both the course and the instructor more highly. Seventy-one percent rated the instructor as “one of the best” compared with instructors of other courses they had taken. In the control section only 36 percent gave that rating to the instructor. In this case the same instructor taught both sections. Forty-nine percent in the section with the biweekly exam said that they would definitely recommend the course to friends, compared with 14 percent in the section with midterm exams.
The faculty researchers who completed this analysis suggest several reasons for these dramatic results. First, students had less material to learn for the biweekly exam, which made them less likely to cram for the exams. Second, they got feedback earlier and more often, which helped them adjust their study behaviors. Third, repeated experience taking the exams increased their feelings of competence and confidence, and that in turn increased their motivation to study and do well.
The results could be explained by any one of these reasons, or these explanations may well have had a cumulative effect. Whatever the cause, the fact that students do better when we test them more often has been confirmed yet again.
Reference: Myers, C.B., and Myers, S.M. (2007). Assessing assessments: The effects of two exam formats on course achievement and evaluation. Innovative Higher Education, 31, 227–236.
Excerpted from Frequent Exams: Better Results for Students, The Teaching Professor, June-July 2007.