Sounds Great: ideas, explanations and facts.

First, a fact that is generally well know and true: teenager drivers are about twice as likely as older drivers to have a car accident.

Now, a question: do you agree that it would be a good idea for schools to provide driving lessons to teenagers before they get their drivers licence so they know how to drive properly and learn useful driving skills? Yes or No?

In our discussion this morning, Mary asked an important question: if the facts are surprising, how do we explain them? (These may not have been her exact words - I wasn't taking notes.)

In my country, Australia, and a number of other countries such as the US and the UK, politicians, parents and educators correctly realized that younger drivers had more car accidents than older drivers. They decided that it would be a good idea to teach driving skills in high school to reduce their high rate of accidents.

Now, what did you answer above? Is this a good idea? It sounds pretty obvious to most people that it must help if teenagers are taught to drive and given useful skills before they get their licence to drive a car.

But like many great ideas that sound obvious, it is wrong.

When researchers analysed the results of three large sets of data, they found that providing driving lessons in high school actually slightly increased the number of accidents for the teenage group (Roberts & Kwan, 2001).

What's going on? What I think happens is that we often get an idea about what must be true because it seems obvious, so we never actually test it. We just assume it is true because others tell us or because it just seems so obvious. We believe our explanation, for example that driving lessons will reduce accidents, and assume that the facts must follow our explanation. Unfortunately, facts do not always fit themselves to our explanations. What we need to do is check the facts, and if they don't follow our explanation, it probably means our explanation is wrong, and we need a new one.

A more likely explanation of the effect of driving lessons in high school on teenage car accidents is that because they have had lessons, the new drivers think they are better than they really are and so are less careful and take more risks, hence leading to slightly more accidents. However, whatever explanation is given must fit the facts; we can change our explanation to fit the facts, but we can't change the facts to fit our explanation.

So, when you're presenting your ideas in a talk or an essay, if you make a statement about what people do, it will be a lot stronger if you have some supporting statistics. Just saying, "It's obvious," isn't very strong support. It's especially dangerous to your presentation if someone does have statistics or facts that are against your idea. It always helps to get some reliable statistics or other facts to support your idea.

For some more examples of some "obvious" ideas that turned out to be false when tested, including a reference to Roberts and Kwan's study, see the editorial ("Scientific testing") and "Science rules OK: Running societies the rational way" (Muir) in the May 21, 2008 issue of New Scientist.

__________

References

Muir, H. (2008, May 21). Science rules OK: Running societies the rational way. New Scientist. Retrieved August 31, 2009 from http://www.newscientist.com/article/mg19826571.900-science-rules-ok-running-societies-the-rational-way.html

Roberts I. G. & Kwan I. (2001) School-based driver education for the prevention of traffic crashes. Cochrane Database of Systematic Reviews. Issue 3. Art. No.: CD003201. DOI: 10.1002/14651858.CD003201. Retreived August 31, 2009 from http://mrw.interscience.wiley.com/cochrane/clsysrev/articles/CD003201/frame.html

Scientific testing can prevent policy disasters (Editorial). (2008, May 21). New Scientist. Retrieved August 31, 2009 from http://www.newscientist.com/article/mg19826574.000-editorial-scientific-testing-can-prevent-policy-disasters.html