KalahariLog

I don't know why, but this story appeals to me on a fundamental level. In the mid-1980's a video called "Private Universe" recorded Harvard grads at commencement answering pretty straight-forward (you would think) questions. What is astounding is that the answers are ones that anyone is likely to give and yet they are obviously wrong. Here is an example: Why is it hotter in summer than winter? Most answers are a variation on the theme "because the Sun is closer to the Earth". Now this may sound reasonable but nearly all the people asked are aware that when it is summer in Boston it is winter in Sydney, so the answer they just gave cannot possibly be reconciled with this fact.

Going back to the video: 23 people were interviewed at random, including some professors and alumni and nearly all of them (21 out of the 23) answered at least one of the following two questions incorrectly: Why are there seasons? Why are there phases of the moon? The survey was a part of a series on science education and it has become a commonly cited example that science is not taught well. Students learn a great deal in high school and in University but when it comes to thinking they may not have acquired the prerequisites.

What the Harvard examples illustrate is that smart people, with 4 years or more at one of the best Universities in the US can hold ideas about basic science that are totally wrong. What is worse, the most common answers to these questions are obviously wrong in that they contradict something that everyone knows. The most common answer for the phases of the moon question, of course, is that the Earth gets in the way of the light from the Sun to the Moon. This is clearly not true as often we can see the Sun and a non-full Moon at the same time. Everyone has seen the Sun and the Moon at the same time and if they held the belief that the Earth's shadow causes the phases of the Moon they should have asked themselves "how come?"

I remember hearing about this for the first time in a speech by Alan Kay, the 2003 Turing Award winner who used to be at Xerox Park and then Apple, and HP, and UCLA, etc. He has been interested in teaching science for a long time and the above example is one of his arguments that we are not doing a good job of it. While I don't for a moment doubt that the way we teach is not the best (be it foreign languages or science) I wonder if anyone has tried to measure how effective different techniques of teaching science indeed are. Rick Garlikov has a very nice description of the "socratic method" to teaching the binary system to 3rd grade school kids -- is this a better way to go that the didactic this-is-how-it's-done approach that is most commonly used?