How to Teach Science

My last blog post from UNB archives: I never really did get the time to read much science education, it will have to wait until my next chunk of research time, which will be this summer, but I came across two interesting books that I’ll talk about briefly here, because they are yet more food for thought.

One of the big disputes in education is “the child must be taught” (“behaviourism”) versus “the child must experiment and find out for themselves” (“constructivism”). The first approach tends to be thought of as conservative, the second is associated with liberals.

I hereby out myself as a liberal in all ways but education. I was firmly on the constructivist side of the debate until 1992 when I taught some Japanese students: yes, they were very shy, not used to expressing their opinions, unsure about the experimental approach. And then they got the hang of it, and these students—who were with us because they weren’t considered smart enough for university places at home—streaked ahead of my UK students. The reason was stunningly obvious; they had something to think about.

Reading The Young Child as Scientist (Christine Chaille and Lory Britain, 1997) I can’t help by being struck by the authors’ faith that by experimenting children learn things. Now, I don’t wholly disagree with them. They state:

1. Young children are theory builders
2. Young children need to build a foundation of physical knowledge
3. As they mature, young children become increasingly autonomous and independent, both intellectually and morally. (5)

All good so far. I also agree when they say that education can foster children’s dependence, rather than their independence, arguing against the “blind adherence to arbitrary rules”.
Then they go on to give various examples of young children exploring their surroundings, but they make huge assumptions about what it is children learn when they observe, so that a child observing a pendulum, learned about pendulum arcs.
Finally, they argue that if a child sees something they don’t understand, they should be encouraged to work it out, so children finding a skeleton of a dog should be asked, “How do we work out what this is?” rather than told it’s the skeleton of a dog.

now for my thoughts on this


First, they don’t distinguish between requiring adherence to blind and arbitrary rules with requiring adherence to explained, rational rules (as a child of a mother who was both simultaneously much stricter than the other parents, but whose rules were also utterly different and always explained, I’m sensitive to this one).

Second, they assume that a child who observes and makes a judgement about object a) can extrapolate into a general theory. I suspect that if the child was asked what she had learned, she would report not the theory of pendulum arcs, but that this piece of equipment could only behave in a particular way--small children rarely make generalisations. It is also quite likely that while she could describe what she saw, she couldn’t explain it. We saw this in the experiment I think I reported on a while back: children were asked what happened if you kept water boiling. Many of them explained “it gets hotter”. It makes perfect sense.

Third, the open ended, non-directive questioning these authors are asking for ignores what we know about the accumulation of knowledge in the world: scientists stand on the soldiers of giants. Children get ‘aback’ (a Cheshire term) of their parents. While it might be very interesting to force a child to work out why a skeleton is that of a dog, surely they will get just as much out of being told it is the skeleton of a dog and then being encouraged to compare it with other skeletons. To lead children through every stage to recapitulate what we already know strikes me as unnecessarily laborious. These authors are mistaking a “given” answer for a ”closed” answer. This is why my Japanese students could do so well: they had the knowledge there, and when asked to try a new skill, had lots of material to test it against.

The authors have also forgotten that while experimentation is one of the primary means children learn, so too is mimicry (think sushi chefs, or little sea otters watching their mums crack shellfish with a pet-rock).

When the authors go onto curriculum development—and I didn’t read too far, this is a book I’ll need to order later—one couldn’t help but notice that they never suggest when exactly it is ok for the teacher to just give an answer. As I have observed before, science is not precisely a straight line of logic. If a child gets an interesting but wrong result through the scientific method, what is the teacher to do?

Oh, and one last comment before I move on: many years ago I used to baby sit a little boy whose mother was a firm believer in constructivism. She used to tell me off when I pulled her little boy away from the mains socket into which he would try to poke his fingers. I’ve always wondered if he made it to adulthood.

The other book I found is by Stephen P. Kramer, How to Think Like a Scientist: Answering Questions by the Scientific Method (New York: Thomas Y Cromwell, 1979). To begin with, this book is written for children, not for adults. It straddles the divide between the constructivist and the behaviourist approaches, in that it encourages experimentation, but and this is crucial, it explains the difference between simple observation, and the process by which scientists turn observation into method, so it shows why some observations are simply untested superstition. It leads readers through why the set of facts reported by two people can lead to two different conclusions. It says to children, that if they think an adult has something wrong, they should think about why this might be the case, and ask the questions and make the challenges that could reveal this; the book shows how to develop scientific questions, and warns, “often it is easier to find an answer we like than an answer that is correct” (16). There is a superb section—centered on how to get chickens to lay more eggs—that introduces the children to the idea of control groups, of variables, and staged experiments, And the book concludes by explaining that part of research is communicating, allowing others to test your results, and it involves reading what other people have written and working with the knowledge others offer—“It also gives them ideas about new things to study”. What this book makes very clear is that science is not a “pure” process, it is not entirely constructivist and cannot be taught that way, and that there is no reason to deny knowledge to children in the interests of stimulating enquiry: children can get just as excited building on knowledge given them, as they can about an unexplained sand pit.