Some time ago, I was acting as the exam invigilator in one of our large first year Physics final exams. We were sharing the grim Saturday morning experience of a three hour examination with other classes in Economics and Sociology. Now in Canadian exams, the students tend to ask more questions than I was brought up to expect , coming from the British exam system, where nobody dares ask a question! This makes the invigilation a bit more stimulating, but only just. Confirming the worst prejudices of most physicists when it comes to sociology, all of the sociology students had left after 90 minutes, leaving the grim faced physics and economics students to sweat it out to the bitter end.
In a moment of boredom, I glanced at the rubric on the front of an abandoned Sociology paper. It was a multiple choice exam, and the instructions stated “Pick the answer which is most nearly right”. I made the comment to a fellow-Physicist that our students would hate that. First year students want “The Answer”, of which there is only one possible. But wait a minute, we are physicists. We understand the difference between accuracy and precision, we understand statistical fluctuations, we know that the Truth, the Whole Truth and nothing but the Truth has to have error bars appended, we quantify uncertainly and we wear our underpants outside our trousers. Sorry, got carried away there. So why do we have this variation between what we actually do, which is assign a certain uncertainty to every result, and what the general public, including first year students, think we do, which is to provide “The Answer”?
We can see this in recent debates and opinions in our local newspaper, the Saskatoon Star Phoenix, where there have been debates on global warming (which is an international socialist conspiracy, apparently), too much salt in the diet and energy saving light bulbs are a bad idea because of the mercury which they contain. Inevitably, the nay-sayers mention to a single article written by a scientist that dissents from the majority view and say, “Look there! A Scientist says this, so it must be correct”. This of course ignores the many other scientific articles which take the opposite viewpoint. As practising scientists we are used to weighing up the evidence, which may come from multiple sources, with a wide variation in credibility and conclusions, so that we can sum up our final position. Actually, non-scientists have to do exactly the same thing in real life too – a similar process goes on when you make a decision to buy a new car, for example. So why does the public think that we approach problems any differently? I think part of the problem comes from the science taught in schools. There are very few examples of problems set which have large quantities of data from different data sets, some of dubious quality or relevance, which then require critical examination. Obtaining “The Answer” for this type of problem may be difficult, if not impossible. Being able to do this sort of problem is very useful in many walks of life such as management, intelligence officers, child rearing, to name but a few. So why don’t we teach people how to do it?
When I came to mark the physics exam, I came across an answer to a problem which I had set on Young’s double slit experiment, where I gave a certain amount of information and required the distance between the two slits to be calculated. A classic example of a problem with “An Answer”! The periphery of the page was covered in a set of hieroglyphics which even Thomas Young himself, a noted translator of ancient languages, would have been hard put to translate. The only thing resembling an answer was near the middle of the page, where the phrase “Pretty close” was visible. Imagine my dilemma. Do I give marks, for what is essentially a true statement, or am I looking for the “Answer”. Fortunately, I was spared this tough decision; in the rubric for my exam, there was the instruction “No credit will be given for a correct answer without supporting working”. Some things are certain, even for the masters of uncertainty.
