Wednesday, February 1, 2012

There's an error in the uncertainty of your measurements

Science vs Politics series.

This title can be interpreted in two very different ways. And the differences highlight the challenge of talking science to politicians and politics to scientists. It is very interesting and I think a little bit misunderstood. That is to say, more awareness of these differences will foster better communication between these two groups in society. It is a little bit scary to me that only one of these groups gets to make the decisions about our society. I think both (in fact all, of societies vast range of groups) should get a say.

The three words in particular in this title are instrument, uncertainty and error. These three words are very different concepts depending on who you’re talking to.

An instrument, to a scientist, is the thing that you observe with. It could be a microscope, a telescope, a seismometer, a scale etc. To a politician, an instrument is a piece of paper that can be used in court as evidence, or in parliament as a law.

Uncertainty in science gives a level of confidence that a measurement lies within a prescribed range. It tells you that no matter how well you make your measurement, you can not be spot on. For example, the temperature today is 15 ˚C ± 0.5 ˚C. Where 0.5 ˚C is the uncertainty, or the range where the temperature actually is. The temperature in this case could be anywhere between 14.5 ˚C to 15.5 ˚C. In politics, if you use the word uncertainty, it means you are not sure, a dangerous thing to admit.

Error, the last and is probably most volatile of these three concepts. In science, error is the difference between your measurement and the actual value of the thing you are measuring. There will always be error in measurement. Error in politics simply means you are wrong. In science, you need to understand and discuss the error, you are not being a good scientist of you don’t. In politics, you avoid it.

Here is a political translation of a scientific statement:

”The error and uncertainty of the instrument were known” = “Not only where there mistakes and things you weren’t sure about in the policy document, but you knew about it and didn’t tell anyone, shame.”

I’d like to see the day where politicians can use these words with confidence in parliament, without being harangued for being wrong and unsure. I’d love to hear the opposition ask the government in questions time “what was the uncertainty of your measurement?” or “Could you characterise your errors?”, rather than “You are wrong, why are you lying to us?”

I’m uncertain that I’ll ever see this happen, hopefully that's my fault!