In the personality course I teach, I administer several multiple choice tests during the semester. I used to feel rather uncomfortable when I passed back the exams. On those exam review days, it seemed to me, the classroom atmosphere was – well, not pleasant. Either students wanted to argue the points, or, alternatively, they sat, sullen-appearing, in response to greeting their grades.

I brought this up one day in a teaching seminar that is part of the Teaching Excellence Program at the University of New Hampshire, and came up with a new plan of action, courtesy of my colleague, Ed O’Brien, a cognitive psychologist who sometimes works in the Teaching Excellence program.

Here was my new plan: Rather than ignore or try to silence the arguments over test items that greeted my return of the exams, or allow my discussions with students to escalate into a mass griping session, I would try to channel the students' energy and direct it toward their own further learning (and my own, as it turned out). I worked out this two-class solution to the problem: In Class 1, I hand back corrected but ungraded exams for discussion; in Class 2, I hand back the final, graded, exams. Here, in greater detail, is what I have been doing recently.

Class 1: Handing Back the Corrected Exam

During Class 1 -- the next class immediately follwing the exam -- I begin by saying something like this:

There are no grades on these exams. I am handing back your exams today and the grading so far involves only designating which of your answers were wrong.

As you also know, no multiple choice test is perfect, and it may be that there are some alternatives that I ought to have graded as correct.

Today, I want you to go over the test with me and identify any item alternatives that, maybe in your opinion, I should have given credit for. That is, today your job is to identify other possible test alternatives that might be right.

This is independent of any curve for the exam or any final grades I assign to the exam. All we are interested in today is whether there are some additional alternatives to questions that I ought to have graded as correct.

You might identify an answer alternative as correct if you can: (a) find evidence in the book for it, (b) find evidence in favor of it from my lectures or from another documented, reputable source, or (c) if the question was worded in such a way that the correct alternative was not obviously the best response.

This is not majority rule – just because (almost) everyone picked the same, wrong alternative doesn’t make it right!

I then appoint 9 or so group leaders (from a class of about 60 people) and give each of them a “leader’s sheet”. They then form 9 or so groups.

Once the group is formed, each class member puts their name down on the leader's sheet (the students get in-class participation credit for joining the exercise). Each group is then assigned to identify one alternative on the test they believed I ought to have credited (they may also "pass" -- that is, say that they cannot find such an alternative). Once their group had done this, they place their nominated test alternative (e.g., Question 45—alternative C) on the board in front of the class. This part of the process is usually over after 30 min. or so.

At about 30 minutes before the end of class (and more if the students are done earlier) I go through the various item nominations, and explain why I scored a given alternative as incorrect, and listen to the given group (and others in the class) explain why they believe it ought to be credited. Sometimes I agree with them, and credit the item, sometimes not, and sometimes I defer my judgment until later.

The result is that we have a very active dialog about the test in a structured way. The students generally appreciate this (see below: Things That Go Right and Wrong).

Class 2: Handing Back the Graded Exam (allow 5-10 minutes for this)

In the next class meeting, I hand back the corrected, graded exam. I first credit any alternatives I have agreed to credit, and then curve the exam if needed, and finally, assign a letter grade. At this point, I simply pass back the exam and tell the students they have been credited, and mention if there has been a curve or not. The students generally have a good sense as to what their likely grades are going to be by this time, and they generally are able to work on new course material without being overly distracted by the return of the exam at the beginning of class.

Things That Go Right and Wrong

This approach has a number of advantages – and potential drawbacks -- but generally the advantages win out.

Things that go right: The class is almost always very energetic about going over the exam. They have an extra chance to review material they missed in a more spirited way than usual. I learn what they learned clearly and what they did not, as well as what they thought was fair or unfair on the exam.

Things that go wrong: On bad review days, I credit too many alternatives (am too easy) and the bright students complain. More commonly, I don't appear to credit a sufficient number of alternatives, or otherwise seem stubborn, or there is an item the class and I disagree about and the class is adamant (and so am I). They complain (occasionally on the final evaluation forms, although this is rare). My guess, however, is that they still evaluate the course better than they would otherwise.

To avoid bad review days, I often like to have a graduate student come into class with me, who I can discuss alternatives with. Due to scheduling issues, however, that isn’t always possible.

Variations on a Theme

I have tried more or less elaborate versions of this. Originally, for example, I had each of the student groups pass their nominated response (that is, their choice of which alternative to credit) to a second group, which then evaluated what the first group said, and either agreed or disagreed with whether it should be credited.

Another possibility I am thinking about (but haven't tried yet) is to experiment with having the class nominate an expert panel – a sort of supreme court of students – to evaluate the alternatives with me. After hearing my side and the class's side, the students would vote in a closed ballot system as to whether they think a given item ought to be credited or not (or rank the alternatives as to which is most important to credit).

One professor liked this approach to reviewing exams but didn’t want to give up an entire class period for an exam review. He had students keep their exam booklets – and gave them a copy of the key immediately after the exam (he kept their exam answer sheets for grading). He then told them they had until midnight that day to e-mail suggestions of additional alternatives that ought to be graded as correct (along with their reasons).