Lesson 2 – David Thompsons’ two-stage collaborative testing to enhance engagement

The Technique

I started using two-stage collaborative testing about four years ago after I was introduced to it at a Scholarship of Teaching and Learning (SoTL) conference. The process was outlined, and then we were asked to take a short version of the Canadian citizenship test in two different ways using the technique. It was the test experience as an individual and then again as part of a group that got me hooked! The experience was a fun, engaging, and enjoyable way to be tested, and I thought my students would appreciate it also.

I use two-stage testing in my research course for two multiple-choice tests I give my learners. Each learner takes the test on their own first, hands in their answers, and then takes the same test again with a small group of 3-4 students. On their own, they are in control of the resulting score but, in a group, they must rely on multiple “experts” and discussion/debate to identify an answer their group agrees upon for each question.

Research is a collaborative endeavor, and I want my learners’ experience to reflect that.

---

Background reading that inspired me to try the technique
Rieger, G. W., & Heiner, C. E. (2014). Examinations that support collaborative learning: The students’ perspective. Journal of College Science Teaching, 43(4), 41-47.

---

How I use It

Key principles that make this technique work include the following:

• Learners’ individual test scores have more test weight (85% of the test score) than the group score (15% of the test score) – weights are then combined to 100% of the overall test weight.
• Nothing is graded or returned until both steps are completed.
• The group test is allotted about half the time that the individual test is allotted.
• I also state that being a “lone wolf” is okay – if learners want to participate in Stage 2 (group testing), they must opt in before we start. If they opt out, the individual score is the only mark.
• I have a no-risk “trick”: the group mark can only raise or maintain an individual score. If a group score is lower, I keep the individual score.*

The Stages

• I state that we will have two multiple-choice tests using a two-stage process that will engage them and help them learn during testing, and I go over the rules (i.e., the principles stated above).
• I point students to the Google doc where they can create their own small groups (of 3-4 students) for Stage 2.
• I make note of students who choose to OPT OUT of the group testing phase.

Stage 1

• Individuals take the test and submit their test Scantron and test papers.**

Stage 2

• Each student group identifies a recorder who will mark answers and submit the test.***
• The recorder takes a Scantron sheet and a test to read out questions to their group.
• Groups debate the pros/cons of possible answers and decide upon their group answer for each question.
• The group recorder submits the test.

After Testing

• I mark OPT OUT students first.
• I mark remaining individual tests and then move on to group tests.
• I match up individual tests to groupings done by students.
• I release both the individual and group scores for each student.

* Student marks are increased by participating in the group testing but not by much because of the weighting (usually 1-2%).

** If a student has extra test time (e.g., as accommodation), that student begins Stage 1 testing earlier than others: for example, if the class starts at 2:00 pm, they can start the test at 1:45 (or whatever their accommodation requires).

*** When using  Zoom for Stage 2 testing, one student should login and share a screen in the group breakout room.

Feedback from Learners

I’ve found that the classroom experience with group testing adds lively, engaging discussion to determine “group think” answers. My students have shown a very high adoption of the technique ( 90+%), and an explosive debate occurred when sharing answers from the Stage 1 testing.

What My Students Had to Say:

“I liked the 2 stage test that was implemented in this course, not only did it help give a little boost but most importantly it helped me to understand what I got right vs wrong and why after completing the test by myself.”

“The 2-step test was a very neat concept and I enjoyed being able to collaborate with my peers.”

“I really enjoyed the test concept for this course. I thought giving us a chance to redo the test as a group was a great idea. I found myself understanding the concepts and the questions better due to having to explain my rationale for my answer choices to my peers.”

“Continue to offer the two part first test. I really liked the opportunity to talk/go over the material a second time. It was a new and enjoyable way to write a test.”

Metrics from Implementing Over a Four-year Period

Test Means: Individual Compared to Group Across Years and Tests

2019 Test 1 Individual: 72 Group: 85 Test 2 Individual: 73 Group: 86

2020 Test 1 Individual: 74 Group: 88 Test 2 Individual: 75 Group: 84

2021 Test 1 Individual: 72 Group: 84 Test 2 Individual: 75 Group: 86

2022 Test 1 Individual 74 Group 81 Test 2 Individual 72 Group: 78

A Short Task to Challenge You!

Let’s try a two-stage test cycle! Even the most “expert” Canadian will benefit from a bit of discussion to answer these questions.

Stage 1

Come up with an answer to all questions below, in under 3 minutes, without googling;-)

1. What is the capital of Ontario; Ottawa or Toronto?
2. Where is St. John’s located; New Brunswick or Newfoundland?
3. Which is farther north’ Vancouver or Montreal?
4. In which month is Canadian Thanksgiving; October of November?
5. Is Boxing Day a federal statutory holiday; yes or no?

Stage 2

Now that you have answered, find the 2-3 people closest to where you are right now, and ask them to help you answer the questions one more time. See if, as a group, you can answer them in just 2 minutes.

One Final Task

Is this something you can use in your classroom? How might you utilize it? If you share your results somewhere on social media, share the link to this lesson for context.