Two-Stage Exams

Two-stage exams are an innovative exam style perfect for classes that use a lot of student student interaction. In a two-stage exams students complete the first, individual, portion of the exam as normal, then they hand in their individual work, get into groups of 4, and complete the same exam again with a group of their peers. In the words of one student:

“Great idea! The group exams give you a chance to go over your answers to the exam while you still care about the questions.”

The student response to this style of exams is excellent both in terms of how they feel about the exams and also in terms of retention.

Below is a collection of resources for those of you interested in learning more about two stage exams. 

Our research group has published the results of two studies on two stage exams. One looks at student retention of content:
Two Stage Exams Gilley and Clarkston

The appendices for this paper are here:
Appendix1and2forMeasuringLearningInTwo-StageCollaborativeExams

And the second that looks at student experiences:
http://www.cwsei.ubc.ca/SEI_research/files/Rieger-Heiner_2-stage-Exams_JCST2014.pdf

If you are interested in trying a two stage exam in one of your classes, the video below explains how the exams are run here at UBC:
A good example of a two stage exam in action is one of the early videos on this page:
Two short two page newsletters that describe the exams:
Tips for two stage exams:
And Using group exams in your classes
A collection of two stage exam resources from Catherine Rawn:

8 thoughts on “Two-Stage Exams

  1. Dana Bjornson

    Thanks for this blog post. Because of its existence, I ran my first two-stage test in my Math 10 classes last week. I think things went fairly well, although I will make a couple of changes for next time. One negative that I came across was that in a couple of groups, the understanding was so limited, that the group mark was not as high as the rest of the class’. The groups were formed by the students themselves and for the most part, they were “friend groups”. I think for my second attempt, I will have students pair up with one person of their choice, then I will pair up the couples. I recognise that this may work in a lecture theatre but with 30 students, it is doable. Hopefully, this doesn’t backfire, though. I would hate for students’ anxiety to increase due to being with people that they were not bffs with!

    Reply
    1. fhmjones

      Hello Dana,
      Thank you for this comment. It sounds like you made an excellent first pass. Your solution to the challenge of “friend groups” sounds like a good idea. I will pass this to colleagues who may not see this post because it is a nice idea for compromise. Other ideas can be found by entering “two-stage” into the search box at the CWSEI website – http://www.cwsei.ubc.ca/resources/index.html.

      Reply
      1. Dana Bjornson

        This looks like an amazing resource! Thank you!!!

        To follow up on my second round of testing, I allowed students to pick one friend, then I paired the pairs. I have also approached my highly anxious students individually, to talk to them about the process and what would help them the most. It seemed to work really well! All of the groups seemed to benefit more equitably. #teacherwin #studentwin

        Reply
        1. bgilley Post author

          That’s great Dana,

          I”m glad to hear that the students responded well to the changes. They are really wonderful (and quite simple) exams, I know I’ll never go back to individual only.

          Thanks for following up!

          Reply
  2. Rob Poole, PhD

    Hi

    I teach a course in management and use a variation on the theme

    I give the students a 10 minute quiz every day that covers the main points I want to discuss. Then I put them in groups to take the same quiz.

    Then that material for several chapters in then covered in an individual exam.

    The concept is that students are more likely to ask another student a question on the quiz than the instructor. The students see the relationship of the material to the lecture and can ask better questions in class and we do not have to focus on the minutia.

    Further, they have several pages of good notes to study for the test.

    Additionally, they can see the difference in their own scores and group scores and learn why group decisions are generally better than individual decisions. This is a learning lesson about groups that is also put into practice when they turn in a group project.

    The course teaches future managers how to lead and control groups. Both theory and practice is explored. This is just one tool.

    Reply

Leave a Reply to Rob Poole, PhD Cancel reply

Your email address will not be published. Required fields are marked *