In a 1992 article in College Teaching, authors Mealy and Host identify three types of students who report high levels of anxiety during exams; those who lack adequate study skills, those who can study but are easily distracted during an exam, and students who mistakenly believe they know how to study but still do poorly on the exam and blame the instructor or think the exam was “unfair”. Seventy-five percent of these students with high anxiety want instructors to conduct some kind of review before the test. They report being less anxious after attending these review sessions.
In surveying students in my math-heavy classes, they indicated a preference for two exams instead of one. They also wanted to average the scores on those two exams or drop the lowest score. I wondered if dropping the lowest scores would make the first exam more like a practice exam and as such it might more effectively motivate students to study. I decided to explore some second-chance options more fully.
In one section of my course, I gave two midterm exams and one final exam. For the first midterm exam, I provided the solutions to the exam problems online afterwards and we reviewed the exam during the next class session. The following week, I gave a second midterm exam—not the same problems but ones at a similar level of difficulty—and dropped the lowest score.
In the second section, I gave one midterm and one final. Scores on both tests were similar, with the average midterm score just three percent higher than the final. However, in the first section, where students essentially got two attempts at their midterms, the scores were 12 percent higher than the scores in section two. In addition, the students in the section with two midterms scored seven percent higher on their final exams than the section with just one midterm and final. I believed part or all of the improvement may have resulted from a decrease in exam anxiety. I decided to continue my exploration of these options. [perfectpullquote align=”right” bordertop=”false” size=”22″] Knowing what they missed on the first exam seems to give students confidence.[/perfectpullquote]
In a third section, I gave two midterm exams and two final exams with the lowest score dropped for each. Compared to the section with one midterm and one final, their average scores on the midterm and final exams increased by 11 percent and 18 percent, respectively.
Giving second chances can help or hurt students. If knowing that there’s a second exam causes students not to study because another exam will follow, then there’s no learning benefit. But there are ways around this potential liability. I announce that the first exam will be easier than the second one and that gives students a powerful reason to study. I don’t say how much easier the first exam will be and in fact it ends up being just about as difficult as the second exam. And even if they do poorly on the first exam, having a chance to fix that, also motivates study for the second exam.
Knowing what they missed on the first exam seems to give students confidence. They understand what they did wrong and what they need to study further to make sure they don’t make those mistakes again. If students are satisfied with their scores on the first exam, they did not have to take the second exam. About 50 percent of the students took the second exams to improve their scores.
The principle of least effort postulates that animals, people, and even machines will inherently choose the path of least resistance. This principle applies to many areas, including teaching. When we get overextended, there’s the temptation to cut corners. The idea of having to create and grade more exams and then not count some of those grades, can feel like a lot of unnecessary work, especially with large classes. However, if it does yield better grades, as it has in my courses, that usually means more learning. If second chances mean my students are learning more, I will put up with the extra work.
Reference: Mealey, D. L., and Host, T. R., (1992). Coping with test anxiety. College Teaching, 40 (4), 147-150.
Milan Toma is an assistant professor in the Department of Mechanical Engineering at New York Institute of Technology’s College of Engineering and Computing Sciences.
