Evidence-based lecturing?

This week’s Science includes a paper by M. K. Smith and colleagues, which assesses undergraduate learning in an introductory genetics course that uses “clickers” and small-group discussions.

The paper’s authors include Carl Wieman, whose educational philosophy was a topic of one of my posts last summer.

In the current paper, Smith and colleagues examine the role of small group discussion as a component of the “interactive” strategy of lecturing:

Some instructors who use clicker questions skip peer discussion entirely, believing that instructor explanation of the correct reasoning will be more clear and accurate than an explanation by peers, and will therefore lead to more student learning. Although our current work does not directly compare the benefits of instructor versus peer explanation, research in physics has shown that instructor explanations often fail to produce gains in conceptual understanding (15). We have shown that peer discussion can effectively promote such understanding.

I wrote about physicist Eric Mazur’s teaching philosophy in 2007, when he was profiled in the NY Times. He has an editorial accompanying the new report in Science, again emphasizing the small-group hands-on approach as a superior alternative to lectures:

I now structure my time during class around short, conceptual multiple-choice questions. I alternate brief presentations with these questions, shifting the focus between instructor and students. The questions address student difficulties in grasping a particular topic and promote thinking about challenging concepts. After posing the question, I give the students 1 to 2 minutes to think, after which each must commit to an individual answer. They do this by submitting their answers using handheld devices called "clickers" (see the figure). Because of the popularity of these devices, questions posed this way are now often referred to as "clicker questions." The devices transmit the answers to my computer, which displays the distribution of answers. If between 35% and 70% of the students answer the question correctly, I ask them to discuss their answers and encourage them to find someone in the class with a different answer. Together with teaching assistants, I circulate among the students to promote productive discussions and guide their thinking. After several minutes of peer discussion, I ask them to answer the same question again. I then explain the correct answer and, depending on the student answers, may pose another related question or move on to a different topic. This approach has two benefits: It continuously actively engages the minds of the students, and it provides frequent and continuous feedback (to both the students and the instructor) about the level of understanding of the subject being discussed (Mazur 2009:51).

I like many aspects of this approach, and I actually teach with a similar style. My lectures are divided into relatively short segments. I ask questions of the students, frequently calling on individual students for answers, and occasionally polling the class by a show of hands.

But what I really don’t like in Mazur’s description is the “multiple choice” formula. I know that the clickers don’t really give an “open-ended” option, but many of my most interesting class presentations have come from using a completely open-ended approach. I react to student questions, bringing topics into the lecture that wouldn’t otherwise have been covered. An open-ended approach allows me to show the application of methods outside the usual textbook canon. And it allows the engaged students to help direct the class content. An episodic series of multiple choice questions has to be completely stultifying for the bright students in the room.

Well, that’s my opinion, anyway. I can’t really imagine coming up with a series of multiple choice questions for each lecture. When I teach laboratories, I have an extensive set of questions ready to challenge students at each lab station, and I certainly have questions ready at the beginning of lectures. But the multiple choice format is demeaning. Or hokey – kind of a “Who wants to be a millionaire?” approch.

And I sort of think that students could spend five minutes outside of class discussing the content, instead of being cheated out of five minutes of my time in class.

UPDATE (2009-01-06): A neuro-potent reader writes:

Evidence-based teaching (direct instruction/programmed instruction) has been around for fifty years and been misunderstood almost as long.
Clicker-training is based on the same principles, but has much more narrow applications and should only be used with animals. There are much more powerful methods available for humans. The science behind the methods was initiated by B.F. Skinner at Harvard in the fifties and have evolved into Direct Instruction (www.zigsite.com) and Precision Teaching (www.celeration.org). There's a lot of little used knowledge out there that can make learning of even complex subjects relatively effortless.

I guess, leave it to the physicists to teach students with methods appropriate for animals!

The e-mail provides some more resources for those interested in the topic. The appearance of this stuff in Science and the widespread press profiles of its advocates is a clue that the topic is trendy. If, like me, you follow the trends as a check on your own teaching style, you might well want to look into these alternatives.


Mazur E. 2009. Farewell, lecture? Science 323:50-51. doi:10.1126/science.1168927

Smith MK, Wood WB, Adams WK, Wieman C, Knight JK, Guild N, Su TT. 2009. Why peer discussion improves student performance on in-class concept questions. Science 323:122-124. doi:10.1126/science.1165919