Problems with conventional teaching

by Stephanie Chasteen on September 21, 2007

Since I’m now employed as a science education researcher, I’m learning a lot about how people are taught, and how that affects how they learn. People have found some really interesting stuff in this field, and here are a few of them.

For one, in traditional physics lectures, it’s found that students learn, on average, about 25% of the concepts that they didn’t already know about force and motion (Hake, using Force Concept Inventory — PDF here). Wow. That’s not as much learning as we’d like.

Eric Mazur at Harvard was skeptical of these results. Not at Harvard, he thought. But when he measured his students’ learning, he found similar results. Why was this? He found that students could do the calculations just fine, but they didn’t really get the concepts. In fact, when he phrased a problem in a conventional way (like calculating the voltage across a resistor in a complicated circuit), about 69% of the students got it right. But when he showed them a similar circuit — but with lightbulbs instead of resistors — and asked which lightbulb would get dimmer… only 49% were able to answer that question correctly. So they can do the calculations, but don’t know the concepts.

One traditional model of learning is that students heads are empty and can be filled by good explanations and content. This is the “transmissionist” way of thinking about things.

Another way of thinking about it is that students construct their own meaning through active engagement. What that means is that students need to be really working through the ideas for themselves by talking about them with each other and thinking about them on their own. One way to do this is through Mazur’s Peer Instruction, which I’ll write about in another post. Basically, he asks students concept questions throughout lecture, and they vote on the answer using radio-frequency “clickers,” and then discuss the answers with their neighbors until everybody understands the right answer. This kind of instruction has been shown to increase the amount of learning that takes place in lectures (more than that pitiful 25%).

Comments on this entry are closed.

Previous post:

Next post: