Does Discussion Make a Difference?

Here’s the scenario: Students are taking a chemical thermodynamics course. The instructor solicits clicker responses to a conceptually based multiple-choice question. Students answer individually, write a brief explanation in support of their answer, and indicate how confident they are that their answer is correct. They are then encouraged to discuss their answers with two or three (self-selected) other students. After that discussion, they have the opportunity to change their answer if they wish, write another explanation for the answer, and once again indicate their degree of confidence in their answer. Do you think that discussion would make a difference—particularly, would it make a difference in their understanding of the concept?

That’s the protocol students followed in the research referenced below. In one cohort, students saw how the rest of the class answered the question before their discussion, and in a second cohort they did not.

The results came down pretty substantially on the side of discussion. “A statistically significant number of students who originally had the correct multiple-choice answer had a higher value code assigned to their explanation after group discussion, and therefore demonstrated more explicit understanding of difficult concepts in chemical thermodynamics.” (p. 1,482) In other words, even though they had correctly answered the question after discussing it with peers, students had a richer understanding of the answer. The same was true for students who initially answered the question incorrectly. Regardless of whether they corrected their answer or answered incorrectly again, in both cases they improved the code value of their explanations. Only when students changed a correct answer to an incorrect one did the code value of their explanations decline. However, the number of students who changed correct answers was small compared to the number who changed from incorrect to correct answers.

Whether or not students saw answer results before discussion did not seem to make a difference in whether answers were changed or in the quality of the explanations offered for the answers. Confidence in the correctness of the answer was enhanced when students saw the class response and it agreed with their answer. Likewise, when they saw the answer chosen by the majority of the class and it was not the answer they selected, their confidence diminished.

Interestingly, in this study students spent on average seven minutes in discussion. Perhaps their interactions were richer because they not only answered the question but had written an explanation supporting the answer they chose. Also of note, extra credit was awarded to students who answered correctly, which probably served to motivate participation in the discussion of answers.

This research confirms other findings reported in other research. When faced with conceptual problems, students need the opportunity to practice solving them. The value of that practice is enhanced when in addition to finding the answer, students talk to one another about the problem and how they arrived at their answers. What’s most encouraging in this study is the documentation that discussion not only leads more of them to the correct answer, it improves their ability to explain why the answer is correct.