The science storm: some say direct instruction works best, while others argue for the discovery learning method. Surprisingly, the two camps are very close to being on the same team.

My dad taught life sciences to junior high students and college students for more than 30 years. He also taught science in the backyard, from the front seat of his car, and at the family dinner table. His insatiable curiosity and boundless enthusiasm for any scientific subject propelled his need to teach his students and his children in whatever method they could understand. When I was 5, science was all about walking on the beach and picking things up. We would talk about what had lived in those shells, and how they had died. By the time I was 14, the California coast had become a vast labyrinth of treasures waiting to be observed and discussed, from kelp washed up in Monterey to the reproductive habits of intertidal fish. Dad was an endless well of both facts and unanswered questions, and that combination of the ability to explain anything and the confidence to let the student figure something out has been my benchmark for good science teachers throughout my life.

I'm sure my father is disappointed that I did not become a scientist, but the lessons learned in science--careful observation, meticulous recording of data, and the need to know more about anything and everything--continue to stand me in good stead.

Dad, as a good science teacher, knew instinctively that he had to modify his methods to adapt to the child, and that he had to present his students with a combination of instruction and the experience of finding something out for themselves. Good science teachers at every level are challenged daily to do the same. Some rely more heavily on lecture and instruction, some feel strongly that exploration and hands-on discovery are the way to go. New research from Carnegie Mellon University questions whether one particular method of teaching science is better than another, and the ensuing debate has provoked a tempest in a Petri dish.

Which Side of the Spectrum?

What the so-called debate seems to leave out of the equation, however, is that the two sides of the argument--the direct instruction camp versus the discovery learning enthusiasts--are actually very close to being on the same team.

The researcher behind the direct instruction data, David Klahr, professor of Psychology at Carnegie Mellon University, describes science instruction as a spectrum of methods that stretch from nothing but lecture and instruction on one end to nothing but hands-on discovery at the other.

The best instruction happens somewhere on that spectrum, not necessarily at one end or the other. "What is needed is what is effective," says Klahr. "A good science teacher behaves like a good scientist" by using what works in a particular situation, not by relying on a single method for every situation.

Klahr's research showed that children who learned by direct instruction how to best design experiments by limiting the number of variables not only retained the information when tested several months later, they were potentially better able to design experiments in other disciplines.

Gerry Wheeler, a former nuclear physicist and currently the executive director of the National Science Teachers Association, calls the debate around direct instruction a "false dichotomy." He also describes science teaching methods as a spectrum "from completely explanatory to completely exploratory. A good teacher has a bag of tricks and can pull out the appropriate trick based on the topic and the age of the student."

A combination of interaction and explanation are needed for the student to be able to grasp the concept being taught, and the teacher learns where on the spectrum a particular age group falls. "A good teacher knows when to give instruction, knows when to walk away," say Wheeler. Klahr agrees. "Science teachers are on the ground, at the epicenter," he says. "They need to trust themselves" to provide instruction appropriate to the topic and the student.

Go-With-the-Flow?

Exploratory learning--the practice of letting students learn for themselves by conducting experiments--is no more appropriate as a total science-learning system than is direct instruction. "There isn't time to reinvent everything," Wheeler says. Students shouldn't have to rediscover basic concepts; that is what instruction is for."

Too much go-with-the-flow, Klahr suggests, without the structured support of instruction, does not effectively teach some concepts. In his research, Klahr used direct instruction to teach students (second to fifth grade) how to design proper experiments.

The Control Variables Strategy, according to findings from Klahr's research, is "the skill that allows scientists to design unconfounded experiments and to draw valid conclusions from experimental outcomes." Good experiment design is a skill essential to all forms of scientific inquiry and one that, according to Klahr, is not effectively learned through exploration. Direct instruction is necessary to the teaching of this concept.

Yet when dealing with misconceptions, direct learning sometimes has little or no effect in overcoming the misconception. "When you directly tell the learner something that contradicts their misconception, it has no impact," says Wheeler. "If you try to do straight explanatory when talking density with a fifth grader, odds are they won't get the concept.

If you don't let them explore, they don't get it." But if you don't explain the concept and the formula (mass divided by volume, and the fact that mass isn't weight) then all their exploring may or may not lead to comprehension, especially when dealing with students below sixth grade. "Some topics are rich in misconceptions," Wheeler says. A good science teacher has to again range up and down that spectrum of techniques, using both direct instruction and exploratory learning to get the concept across.

Klahr's research emphasizes this. "There is good research there," Wheeler says. "It's a shame it gets lost in the debate." What's important in Klahr's findings--and has been largely overlooked in the arguing, says Wheeler--is that if you can get the student to understand something it doesn't matter how they learned it. A concept is not better understood or more deeply ingrained if it is learned through direct instruction or through exploration. Once learned, it's learned. "If they really learn something in discovery learning," says Klahr, "it is not any richer down the line. It doesn't matter how they learned it. And," he adds, "I believe that's true in a lot of domains. Do you remember how you learned something? No. You just know it."

The Importance of Professional Development

The curriculum supervisor for a school district has a large portion of the responsibility for seeing that the science teachers in his or her domain have access to the spectrum of teaching methods available.

A curriculum supervisor is, to Wheeler, "the culture keepers of the program." They have to have enough knowledge of what good science teaching is to be able to keep their teachers using the techniques appropriate to the topic they are teaching. School districts need to spend more time getting science teachers connected to current research. "Professional development needs to be taken seriously," says Wheeler. "The same way a pilot needs to learn about a new plane he is going to fly," science teachers need to know about the latest methods.

Klahr points out that even though the "debate" around direct instruction has been largely imaginary, it is good to raise issues about how science is taught. Teachers need to be encouraged to consume scholarly articles and apply new methods to their teaching strategies. "Having better consumers of science of any kind-we're better off," Klahr says.

Klahr will continue to study how children learn science. Like any good scientist, he first wants to replicate his work with elementary students and Control Variables Strategy in a larger study to see if the results are comparable to his first study. He also plans to do more research into what is meant by "hands-on" (hands on what? How? Does point-and-click count as hands-on?) science instruction and what happens when children are taught hands-on versus hands-off.

"Classroom research," says Wheeler, "is very messy stuff." Physicists may remove the air in order to drop an apple but you have to deal with the messy stuff, the air, in a real classroom. Further research will probably not reveal the ideal teaching method for every scientific concept, nor should it. Good science teachers will continue to use the spectrum of methods available to them to best teach their students.

Elizabeth Crane is a contributing editor.