Collaborative testing (aka team-based testing) is a collaborative learning technique that has received a fair amount of attention from researchers (LoGiudice, Pachai, and Kim 2015). It is a practice that typically involves students taking a test individually and then taking the same test – or a similar test – collaboratively in small groups. I suspect that part of the reason that it has gained attention from researchers is that its effectiveness as a pedagogical practice can be quantified in a rather straightforward way: you can measure students’ learning with a third individual assessment and compare the scores to a control group in which students took the previous tests individually.

Now, retaking an exam (or taking a parallel form of an exam) with a group of people who have also studied the material means that the group test scores are almost always higher than the individual test scores. This can raise red flags related to overall grade inflation and the potential for freeloading by some students. However, this can be mitigated by weighting the collaborative tests less than the individual tests.

Collaborative testing challenges you to change the way that you think about testing. Instead of viewing quizzes and exams as summative assessments, they become more formative.* It’s true that the grade for a student who didn’t prepare for the test will benefit from collaborating with other students, but that student will also likely learn something about the material through that collaboration, which wouldn’t happen with only an individual test. Additionally, the prepared students are unlikely to suffer any negative implications from the collaborative testing. To the contrary, explaining an answer to another student could help reinforce their own understanding of the material.  

The motivation for this practice of repeated testing likely stems from research on the testing effect that shows that retrieval practice enhances long-term retention of information (Roediger and Karpicke 2006). This effect holds true regardless of whether feedback is provided after the test (though feedback certainly enhances the effect; Roediger and Butler 2011). Further, retrieval practice appears to be a more effective way of consolidating information into memory than using the same amount of time restudying the material (Roediger and Karpicke 2006).

There is also evidence that the testing effect goes beyond just retention of answers to specific test questions, which is great because if all we were doing was getting students to remember that the answer to question number 12 on a quiz is option C, that would be pretty disappointing. The effects of testing have been shown to transfer to different question formats and wording, as well as to questions that test analogous forms of reasoning (Pan and Rickard 2018). Research even suggests that testing can improve future learning of subsequent information (this phenomenon is known as the forward testing effect; Yang, Potts, and Shanks 2018). In other words, giving students tests is great.

Now, the above research really just shows that the testing effect exists, and that repeated retrieval practice has a number of positive effects. When it comes to team-based testing, the question is whether allowing students to do at least some of that testing in groups further benefits learning above and beyond what is gained from individual testing. Unfortunately, the best I can say here is that it doesn’t appear to hurt. Research that tests whether there are differences in learning (based on scores on a subsequent individual assessment) for collaborative testing and individual testing is mixed, with some studies showing an additional benefit for collaboration and others showing no difference (LoGiudice, Pachai, and Kim 2015). That’s not great, but it’s not horrible either.

The good news is that we can at least say that if you are planning on using practice tests in your course, allowing students to collaborate on them won’t negatively impact their learning. Though that may not be a ringing endorsement, there are other reasons that are more clearly supported by research that might encourage you to consider using class time for collaborative testing:

1. You can reduce students’ test anxiety and increase their perceptions of their own learning. Though it’s not clear whether students actually learn more when they use collaborative testing, research consistently shows they report greater perceived learning and less test anxiety when they are allowed to collaborate on tests (LoGiudice, Pachai, and Kim, 2015). The discrepancy between the findings for actual learning and perceived learning serves as a good reminder to pay attention to how learning outcomes are measured in research, but student perceptions are important too. If students feel like an activity is helping them learn, they are going to be more likely to engage with it.
2. You can take advantage of short answer tests without dramatically upping the amount of time you spend grading. One study on the testing effect in a class setting found that the testing effect was particularly robust when short-answer questions were used for the practice test rather than multiple choice (McDaniel et al. 2007). This was true, even though the final assessment was multiple choice. This is good to know, but grading short answer tests can be time-consuming (and tedious), particularly if you have a large class. To gain the benefit while managing your own workload, administer an individual multiple-choice test first, followed by a collaborative short-answer test. 
3. You may be having them do other types of collaborative activities in class that aren’t necessarily as effective. In an experimental study on test-enhanced learning, researchers (Stenlund, Jönsson, and Jonsson 2017) found that students who were randomly assigned to use class time for a collaborative test showed greater long-term retention of the material than students who were assigned to use the same amount of class time to engage in group discussion (with access to their notes and texts). Even though the discussion groups were given the same set of questions as the testing group, they didn’t reap the same benefits. A cognitive explanation for this is that the discussion groups didn’t have to practice retrieving information from memory because they had access to their notes. However, there is also a social explanation for this. The discussion groups typically reported spending 50% or less of their time actually working on the questions. It’s possible that the collaborative testing format provides structure and clearer expectations, which help students use their time effectively.

* When we think about assessment, it’s useful to break it into two types: formative and summative. A prototypic summative assessment would be a final exam. The primary goal there is to assess students’ learning.  Formative assessment, however, has the goal of using assessment as a tool to teach. Grading a draft of a paper and providing feedback would be a good example of this. Because the paper is a draft, you aren’t assuming that the student is demonstrating the level of mastery that you expect on a final draft. Instead, this is an opportunity for them to learn by doing and to receive targeted guidance.

References

LoGiudice, Andrew B., Amy A. Pachai, and Joseph A. Kim. 2015. “Testing Together: When Do Students Learn More through Collaborative Tests?” Scholarship of Teaching and Learning in Psychology 1, no. 4: 377–89. https://doi.org/10.1037/stl0000041.

McDaniel, Mark A., Janis L. Anderson , Mary H. Derbish, and Nova Morrisette. 2007. “Testing the Testing Effect in the Classroom.” European Journal of Cognitive Psychology 19, no. 4-5: 494–513. https://doi.org/10.1080/09541440701326154.

Pan, Steven C., and Timothy C. Rickard. 2018. “Transfer of Test-Enhanced Learning: Meta-analytic Review and Synthesis.” Psychological Bulletin 144, no. 7: 710–56. https://doi.org/10.1037/bul0000151.

Roediger, Henry L., and Andrew C. Butler. 2011. “The Critical Role of Retrieval Practice in Long-term retention.” Trends in Cognitive Sciences 15, no. 1: 20–27. https://doi.org/10.1016/j.tics.2010.09.003.

Roediger, Henry L., and Jeffrey D. Karpicke. 2006. “The Power of Testing Memory: Basic Research and Implications for Educational Practice.” Perspectives on Psychological Science 1: 181–210.

Stenlund, Tova, Fredrik U. Jönsson, and Bert Jonsson. 2017. “Group Discussions and Test-Enhanced Learning: Individual Learning Outcomes and Personality Characteristics.” Educational Psychology 37, no. 2: 145–56. https://doi.org/10.1080/01443410.2016.1143087.

Yang, Chungliang, Rosalind Potts, and David R. Shanks. 2018. “Enhancing Learning and Retrieval of New Information: A Review of the Forward Testing Effect.” npj Science of Learning 3, no.8. https://doi.org/10.1038/s41539-018-0024-y.

David Buck, associate professor of psychology, is the 2020-2022 Center for Engaged Learning Scholar. Dr. Buck’s CEL Scholar project focuses on collaborative projects and assignments as a high-impact practice.

How to Cite this Post:

Buck, David. (2021, January 29). Why Letting Your Students Collaborate on Exams isn’t a Bad Idea [Blog Post]. Retrieved from https://www.centerforengagedlearning.org/why-letting-your-students-collaborate-on-exams-isnt-a-bad-idea