Teaching-Learning Center: Learning Matters

A site made possible by a 2001 - 2002 grant from the Carl D. Perkins Foundation

Collaborative Testing: One Instructor’s Experience

Peter Wooldridge

Collaborative learning is a term that is almost inevitable in any discussion of pedagogy in the community college classroom. To many, the term refers to some type of group activity that is designed to engage students and force them to use each other as a resource, and perhaps, in the process, allow them to serve as both a “teacher” and a “learner.” While few instructors would disagree that group activities can provide valuable learning opportunities, it is my experience that many of these same instructors are hesitant to use them. There are two primary reasons why this is so. The first is the very real perception that there is not enough time to teach all that has to be taught. Those who fall in this camp feel, rightfully so, that setting up and running collaborative or group activities takes too much time. This is especially true in the community college classroom where instructors are frequently teaching student skills along with content.

The second reason springs from what I believe to be a lack of confidence on the part of the instructor. I am convinced that many simply feel that there is too much risk in group activities, i.e., outcomes are harder to predict and control. Thus, the instructor has to work harder to help influence the outcomes in the direction of the learning objective. This can be an exciting challenge but, for people teaching 15 to 20 hours a week, it can also seem overwhelming. Add to that the fact that the “personality” of the class can make or break a group activity and it becomes clear why so many instructors steer clear of this learning approach.

One area in which collaborative or group work is not typically considered to be appropriate is testing. The traditional testing model is, of course, an individualized event in which students work alone without the use of notes, textbooks or other people. I believe that a collaborative approach can also be used effectively within the testing environment. To accept and implement such an idea, however, requires a significant change in one’s “frame of reference” about the goals of assessment. For several years, I have included a group testing component in my General Psychology classes. Students are placed into groups of three to four on the day of the test and complete the multiple-choice component of the test as a group. They do not have to agree on the answer as a group and each member may choose whatever answer he or she wishes. Students who want to work alone may do so and students who come in late must work alone. The class members, prior to the first test, decide what is considered to be “late.” Over the past five semesters, it has ranged from five to eight minutes. Once a student has completed the first part of the test, she or he goes on to complete the short answer and essay components alone.

Admittedly, this activity reflects my philosophy that testing can and should provide a learning opportunity. Instructors who do not agree with this particular belief will perhaps find what I do questionable. However, I have discovered that there are important benefits for both my students and to me.

First, and most immediately, there appears to be a reduction in students’ anxiety about the test. Knowing that there is someone else to rely on seems to provide some students with a dose of self-confidence. This may help to reduce the significant test anxiety that I see in many of my students.

Second, students actually seem to enjoy the group test, especially after they recover from the feeling that they are cheating. I have been amazed at the level of discussion that ensues when there is disagreement about an answer. Individual students will defend their answer with great passion and actually appear to be having a good time while they do so! This discussion provides an excellent opportunity for students to teach each other and to reason out their own ideas as they face questioning and opposition.

The students’ discussions have also provided a rather surprising opportunity for me. As I wander through the classroom and listen to what students tell each other, I hear what they “heard” (or believe they heard) in class lectures and discussions. It has been an eye-opening (and ear-opening) experience, and it has forced me to attend more directly to the fact that significant “distortion” is going on as information flows from me to the student. This has prompted me to provide new and more varied examples, to revise or drop examples I have used in the past, to spend more time on specific topics that I have been able to identify as particularly confusing and to assess my students more frequently to correct errors of interpretation early on. It has, in fact, made me a better teacher.

As with every teaching tool, there are problems. Not surprisingly, some students arrive unprepared for the test but reap great benefits if they happen to be grouped with well-prepared students. Because of this, I am careful to make the multiple choice portion of the test no more than 60% of the grade and, in my experience, it is rare that an unprepared student is able to do well enough on the rest of the test to finish with a grade of “C” or higher. Despite this, the creation of groups has to be given some thought and the instructor has to guard against certain “alliances” that might develop.

I have also seen individual students who I have identified as having low confidence in their academic skills change a correct answer to an incorrect one because of a group member or members who dominate the discussion. This is especially likely to happen if there is a lot of disagreement in the group about the answer. It is difficult to watch, and my urge to intervene has to be kept in check. I have recently instituted a new policy that is meant to provide students with a second chance on these difficult questions. After the group has completed the multiple choice, I allow the students to pick any two or three questions (depending on the length of the test) they wish, and I will tell them if their answer is correct or incorrect; although if the answer if incorrect, I do not tell them what the correct answer is. They will then have an opportunity to change the answer. I have found that those “low confidence” students who were initially correct about an answer that the group changed show genuine delight at having been correct, and I am now trying to gather data on whether this experience may affect their motivation in the classroom.

Another potential problem lies in the fact that there are students who wish to work alone. I am careful to let students know that this is an option. Few students pick this option, but it is not clear to me whether some students want to work alone but don’t because they feel that they might be singled out in some way by working individually. The pressure to conform to the group “norm” can be strong, and working alone for no obvious reason might make some students feel too “salient.” I have tried to remedy this by introducing the idea of group tests on the first day of class and then discussing it several times prior to the first test. Each time we discuss it, I am careful to let them know that working alone is acceptable and that students in previous classes have chosen this option. However, I am not certain that this effectively addresses the problem.

A final problem is that different groups of students finish at different times. When a group finishes, the members move on to the individual portion of the test while students in other groups are continuing to discuss multiple-choice questions. This creates a noisy environment. Although I have never had a student complain, I worry that this may interrupt the students’ concentration. I have not found a very effective method for dealing with this except to provide a tentative time for the multiple choice component to be completed and to remind students to be as quiet as possible when other students move to the short answer/essay portion of the test.

None of the problems I have discussed outweigh the benefits I see in this collaborative testing approach. While I have not gathered formal data, I believe that this method most benefits those students who are academically strong and those students who are in the middle of the academic continuum. The academically strong students benefit from the opportunity to verbalize the logic behind the questions while the “middle of the continuum” students benefit from both the modeling of the stronger students and the opportunity to defend their own reasoning. Based on an informal examination of final grades, it appears that this has translated into somewhat higher grades for the “middle students.” The grades of the stronger students are generally near the top so there is a ceiling effect for this group.

Those students with weak academic skills tend to remain passive in the groups and to follow the lead of the strongest voice. Their grades have not generally improved nor have they gotten worse. At present, I am using different groupings of students to try and prompt a more active role by these students, which I hope will translate into higher course grades.

In this article, I have illustrated one way in which a collaborative approach might be used in a nontraditional way. I hope to prompt others to think about how collaborative approaches might be used in various ways in their own classrooms. Pushing ourselves to be more creative in the classroom can be a difficult and sometimes uncomfortable experience, but it is clear that to do so can provide unexpected opportunities for everyone involved.