Link to CSU home page

Information about this Journal

Call for Papers

Submission Guidelines

Events & Opportunities

List of Editorial Board members

List of Contributors

Link to ITL home page

Link to Exchanges home page

Demonstrating the
Use of Research in the Graduate Classroom:
"Cooperative Testing"

Pamela J. McCrory

Department of Educational Psychology and Counseling
California State University, Northridge


I collaborated with graduate students enrolled in a group counseling course in the Master of Science in Counseling program in a classroom project applying current educational research to their learning process. The research I selected reflected my teaching goals and several course objectives, as well as the conceptual framework and learning objectives of the Michael Eisner College of Education.

Project Background and Development

The first of my teaching goals was to engage students in an interactive partnership in the learning process. By focusing on a concern and challenge experienced by a significant number of students, namely, test anxiety, students were interested and invested in the process.

The second goal was to disseminate relevant research and to engage students in reflective and critical thinking about the application of research to real life problems.

The third goal was to incorporate a change in traditional methods of student assessment, as advocated by Stanford University researchers and teachers Philip Zimbardo, Lisa Butler, and Valerie Wolfe (2003): the use of cooperative testing and learning partnerships in which the sharing of knowledge and understanding between students is facilitated and valued. Such collaboration encourages positive interpersonal process and facilitates the creation of a learning community based on the belief that all individuals can learn and contribute. These objectives reflect the College philosophy, which can be viewed at

Zimbardo et al. (2003) report their own experiences in testing students in the traditional manner. In reviewing the literature on test taking, they conclude that examinations are experiences of great anxiety, interpersonal competition, and lowered self-esteem for many students, especially those whose individual performance is low. They discuss the particular disadvantage of traditional individual test taking for students who have lowered performance due to differences in linguistic and cultural background, learning style or test anxiety.

In their study, university students enrolled in introductory psychology courses worked in pairs to prepare for and take multiple-choice exams cooperatively. Significantly improved average test performance and reduced grade variability were found across four separate comparisons of students in the cooperative testing and traditional individual testing. In a follow-up survey completed immediately after the test, students reported the following positive benefits in their evaluation of their cooperative testing experience: "(a) reduction in test anxiety during study (81%), (b) reduced test anxiety during testing (88%), (c) encouraged sharing of knowledge (87%), (d) gave practice in negotiating differences, and (e) enhanced learning (52%)" (Zimbardo et al., 2003, p. 114).

In my group counseling course I used a multiple-choice exam based on the content of course text and lectures to introduce the cooperative testing approach. I shared with students a summary report of the research (Smith, 2003) and discussed the potential benefits and goals of the process, as well as their concerns about testing . Students were eager to participate in this experience. They were given the choice to prepare for and take the exam in pairs or individually. Of the 23 students in the course, 91% elected to take the exam with a partner. The administration of the exam in pairs was followed by completion of a Cooperative Testing Evaluation Form (Appendix A). The results of the test, the classroom distribution of test scores, and the results from 19 students who completed the evaluation form were distributed to the students. Class discussion of the research project included review of open-ended responses, frequency of responses to the evaluation form questions, and comparison with results reported by Zimbardo et al. (2003).

Comparison of Survey Results

Our classroom survey echoed the positive outcomes reported in the original research. All of the students in our class who took the test cooperatively reported preparing for the test in the same manner as if they were preparing to take the test independently and found sharing of knowledge during preparation helpful. Ninety-five percent of the students reported a decrease in their level of anxiety, both during the preparation and test taking experiences. All of the students indicated that they would utilize this approach again and would recommend it to other students. A sample of open-ended comments in the evaluation appears in Appendix B.

I conducted a class discussion regarding the findings of both the original research and the classroom project. We reviewed the limitations of the procedure, the settings to which the findings might be generalized, application to other educational settings, suggestions for variations in the procedure, and ideas generated about evaluation methods. Zimbardo et al. (2003) discuss the limitations of generalizing their results beyond their student population. A high degree of motivation to excel, academic diligence, and high level of academic achievement characterize students at Stanford and should be considered when generalizing to other student populations. In addition, the researchers suggest that cooperative testing in a residential college may have advantages over other settings in that students may have a more effective way of selecting appropriate partners for study and testing. In contrast, our students commute, tend to take classes together and know each other better than students in large undergraduate classes. The novelty of the cooperative testing method, which was a factor for both the original and our classroom project, could boost test performance but is unlikely to reduce positive outcomes in settings in which the procedure has become routine.

There was a striking difference between the original research and our project: 91% of the students in our class chose the cooperative testing method in contrast to 47% of the Stanford students (Zimbardo et al., 2003). To better understand this difference, we reflected on the original researchersí attempt to account for their studentís expectations regarding individual versus cooperative testing. The researchers administered a questionnaire with hypothetical testing options to a large group of Stanford introductory psychology students and found that about 35% would choose cooperative testing, and 39% were uncertain what option they would choose. In discussing these findings in both the actual and hypothetical testing situations, the authors surmise that factors such as the intense competition of undergraduate education at Stanford University, dislike of involuntary assignment to partners, the misattribution that partners were likely to do less test preparation, possible reluctance of shy, nonassertive students in requesting partners, as well as student age, may have influenced student choice. The researchers suggest that further investigations of such factors are needed.

Our project suggests that student variables such as age and experience may have an important impact on the perception and interest students have in utilizing team testing. Graduate students are older than undergraduates and often have considerable professional and life experience likely to impact their view of testing, their learning goals, and their degree of competitiveness with respect to grades. Support for the finding of a preference for cooperative testing by older students was found in the review of research cited in Zimbardo et al. (2003). The attitudes and academic values of our students may be different from those held by the Stanford undergraduates, and this may contribute to the considerable difference in our findings.

Degree of cultural diversity was not a factor included in the description of the Stanford sample. The students in our class and graduate program are culturally and linguistically diverse. Zimbardo and his colleagues (2003) discuss the potential detrimental effects of traditional competitive examination procedures for those with limitations in test performance due to such factors. The impact of cultural, linguistic and educational variables is important for future research.

A similar positive outcome along with the significantly greater number of students in our class who chose the cooperative testing option and also favored it for the future, compared to students in the Stanford study, suggests that cooperative testing may particularly fit graduate or undergraduate settings which are collaborative and socially supportive. Indeed, Zimbardo et al. (2003) call for the creation of just these kinds of "learning communities" (p. 123). Cooperative testing may be thought of both as a process to encourage such changes and as a method particularly valued by students in more collaborative academic environments.

Instructional Applications

To what other academic settings might cooperative testing be extended? Philip Zimbardo reports using cooperative testing in all of his courses, including more advanced level courses: "On the basis of positive student response to this testing format of academic gain and less reported testing pain across diverse settings, greater operational use of cooperative learning and testing seems warranted" (Zimbardo et al., 2003, p. 122). We encourage others to utilize the method at all levels of schooling. Introducing cooperative testing as a classroom learning tool to prepare students for the rigors of standardized testing may help to counter some of the negative effects of solo, competitive testing and foster the social and academic benefits of shared learning.

I found the entire process of discussing, demonstrating and reflecting on innovative research an effective way to implement my teaching philosophy and course objectives. Cooperative testing allowed me to appreciate this innovative process directly and to experience being part of a supportive classroom learning community. I witnessed a novel academic event: students excited to take an exam. Students used the social world to support their learning, which is, along with increased self-efficacy, a key benefit noted by the original researchers (Zimbardo et al., 2003). My personal experience was one of increased efficacy as an instructor. My students confirmed that increased self-efficacy was a shared experience.


Smith, D. (2003, February). Team testing boosts test scores, study finds. Monitor on psychology, 18.

Zimbardo, P. G., Butler, L. D., & Wolfe, V. A. (2003). Cooperative college examinations: More gain, less pain when students share information and grades. Journal of experimental education, 71 (2), 101-126.

Posted June 23, 2005.

All material appearing in this journal is subject to applicable copyright laws.
Publication in this journal in no way indicates the endorsement of the content by the California State University, the Institute for Teaching and Learning, or the Exchanges Editorial Board.
©2005 by Pamela J. McCrory.

·· exchanges ·· classroom ·· top of this page ·· | ITL home