Better controlling controls.

A graph provided with the article "Think like a scientist" (SN: 6/20/09, p. 20) presents data about the understanding of control of variables by students from two different groups, those who received explicit instruction about the concept and those who were left to learn by an exploratory approach. However, the graph also appears to show that the students in the direct instruction group had a much higher mean score on prior knowledge of the concept. Especially for research about learning the concept of control of variables, I am surprised that better-matched test groups were not chosen for this research.

Greg Skala, Nanaimo, Canada

The difference between average scores of the two groups of children on a written pretest that's apparent in the graph was not statistically significant. The statistically significant difference between groups on control-of-variables knowledge emerged on a test administered shortly after each child received either explicit or exploratory instruction.--Bruce Bower

Intentionally or not, the article "Think like a scientist" clearly illustrated the problem of teaching scientific reasoning. The statement about the control-of-variables strategy, "Researchers hold constant all changeable features in an experiment except for one of interest," does not make sense to a young person. While some engineers and medical researchers do single factor experiments, a single factor experiment is only marginally better than a no factor experiment. The issue is to model a process and, as noted in the article, the concept of modeling (using math and experiments) is critical for the understanding needed to make predictions.

When teaching people, trying to present information and methods that, on the surface, contradict observation will not be successful, unless one clearly notes how the experimental model is correlated to reality. And reality involves more than one variable and lots of interactions. I would suspect the results in the "discovery" versus "direct" instruction groups had more to do with the presentation and reasoning ability of the teacher than the characteristics of either method. My personal opinion is that some of both are required to engage natural curiosity.

David Sweetman, Dyer, Nev.