It is difficult to get students to think critically about their own beliefs about things that matter to them. This is a three-part exercise that addresses this problem.

Part I^[1]

The students are asked to pick some belief they care about and to give the best argument for it they can think of. We usually ask them to hand this in at the end of Week 2. Because this is very early in the course, we do not expect a very high standard of argument and we do not allocate many marks to this part of the assignment. Usually it is worth 5% of the final grade, and we grade it generously: anyone who has given an argument at all gets 3/5.

Part II

The students are asked to critically evaluate the Part I argument of one of their fellow-students and to provide constructive suggestions about how to improve it. Part II is due around the midpoint of the course or slightly later, and is worth 10% of the final mark for the course.

Part III

The students receive constructive feedback on their Part I argument from both the instructor and a fellow student (who has evaluated their argument in Part II). At the end of the course, hopefully having a better understanding than before of the difference between good and bad reasons for belief, the student improves their original argument in response to the feedback, presenting their improved argument both in standard form and as an argument tree and explaining why they have made the changes that they have (and, if they have rejected some of the suggestions given by their fellow-student or by the instructor, explaining why they have). Sometimes the feedback convinces them that their original conclusion was too strong, or even that it was completely misguided: in such cases, they are allowed to argue for whatever they now believe about the topic at issue. Part III is worth 15% of the final mark for the course.

Results

The final versions of the arguments are almost always greatly improved, and in those cases where students really have argued for something they feel strongly about, we hope (and we are explicit about this in class) that the process has provided a model for future examination of their own reasons for belief.

Practicalities and problems

• We anonymise the Part I arguments before distributing them to fellow-students. This seems particularly important since we have asked the students to argue for conclusions that they care about, and they may be sensitive about others in the class knowing what they have argued for. Genuine anonymity is not possible in a very small class: we have used this assignment only in classes of 50 or more students.
• An earlier version of this assignment did not include instructor feedback on the Part I arguments: the only feedback was from a fellow-student. This disadvantaged those students whose arguments were incompetently evaluated: the current version of the assignment ensures that all students get some useful feedback.
• There remains a problem for students whose Part I arguments are incompetently evaluated: if we give them the incompetent or unconstructive feedback from their fellow-student, they have the extra task of evaluating the feedback and seeing that it isn’t useful. Since we, the instructors, have graded Part II and provided comments on it, one solution is to give the student receiving the peer feedback our comments on the peer feedback (with the grade removed). This is what we standardly do. There are however some cases in which the peer feedback is completely useless, and there are also cases in which the peer feedback is hostile. In those cases, we do not provide the student with the peer feedback, but we make sure that there is plenty of instructor feedback.
• This assignment is labor-intensive from a grading point of view. It is important that the grading is done by an experienced instructor: for it to be gradable in a reasonable amount of time, the grader has to be able to quickly see what can be done to fix an argument, and this is a skill which can only be developed with long practice. The grading is time-consuming even for an experienced instructor, but we think the pay-off justifies the time spent—the three-part assignment helps students to see the usefulness of critical thinking and shows them how it can be applied to things that really matter to them.