Stand and deliver ... or let them discover?

Picture this: While using a new computer application, you're stuck. As you start experimenting with key commands, a friend looks over your shoulder and says, "Just press that button." Onward, to the task at hand.

A study published in the journal Psychological Science shows that the same thinking can be applied to science teaching. Discovery learning, widely accepted as the best road to true understanding of scientific phenomena and procedures, can be an effective option. But in certain instances, direct instruction may be better.

The study contrasted discovery learning, with only the teacher's suggestion of a learning objective, and direct instruction, which included teacher-controlled goals, materials, examples, explanations and pace, in getting 112 third- and fourth-graders in four elementary schools to learn good experiment design.

The discovery learners were asked to find out what they could about designing a good experiment to see how far a ball rolls down a ramp, explains co-researcher David Klahr, a psychology professor at Carnegie Mellon University. The direct instruction group was told how to design a good experiment, and then asked to do so. Afterwards, students designed four more experiments. Prod a week later, the effectiveness of the two teaching methods was tested in a more authentic context, with children evaluating science fair posters.

The findings? Discovery works for some kids, "but many more learn in direct instruction," Klahr asserts. Procedures, for example, are tough to discover. "If you want to get kids to a mastery level real fast, then you can tell them what they need to know.... Given the tremendous amount of stuff [science teachers] have to cover, I'd like them to see that direct instruction can be good science."

Given discovery learning's popularity, education listservs were abuzz about the study. Klahr says he saw it "quoted, misquoted and counter-quoted." He could see future studies on when discovery learning is best. Kids also need to feel "how wonderful science is [through] discovery."

www.psy.cmu.edu/faculty/klahr