Damn Statistics!
by Marcus Bachler
Remember Benjamin Disraeli? That 19th century British Prime Minister and champion of classical liberalism, and his now- infamous quote, "There are three kinds of lies: lies, damned lies and statistics"?
I am constantly frustrated by official statistics - the way that politicians and journalists often throw them around in discussions in order to justify any argument or draconian legislation they wish to put in place. Take some examples in the UK over the last month or so. Many journalists argued that we should ban smoking in public places because in a poll 60% of people said they wanted it … or, politicians argued that statistics show us that the public are unhappy with increasing levels of “binge drinking” and therefore special legislation is needed to stop this behaviour ... or a Commons select committee produced statistics showing that there is rising obesity in the general population and therefore new fat taxes are needed.
Therefore, it was a bit of a revelation to me recently when I read a book on economics about the start of statistics. Apparently the word “statistic” is a modern amalgamation of two Latin words. From -icus (= of, relating to, or resembling) and status (= concerning affairs of state). Not only that, but a look in the Oxford English Dictionary shows that a statistician in the 19th century even used to be called a "statist" by trade!
The politicians of today act as if they can prove anything they want by quoting what the probability of something is. We are so used to hearing these statistics that we forget that most statistics are just that, a probability of something. We often fail to ask the basic question whether a statistic that is based on a (random?) sample of about 1000 people (in the better ones) can be an accurate representation of a population of millions of voters. As anyone who deals in statistics will tell you, statistics can be made (i.e. manipulated) to prove almost anything you want them to prove. This is often the case when incumbent governments turn up with wonderful statistics showing voters how much the quality of life has improved since they took office, while the opposition party’s statistics show the exact opposite. They can’t both be right.
Statistics, I believe, were historically a tragic side-effect of the Enlightenment's noble credo of “Don’t just tell me, but prove it,” becoming instead, “Don’t just tell me, give me the probability of it being true.” When you think about it statistics are the antithesis to the philosophy of “Objectivism” – probability and uncertainty becoming the “subjective” tools of pragmatic philosophy. We are constantly told by bureaucrats and lawyers that the “truth” is not known exactly but has a certain probability that is subject to interpretation.
I think that it is time to get back to the true spirit of the Enlightenment - don’t just tell me, prove it to me with “certainty”! It is time to bring politicians, bureaucrats and lawyers back out of the Dark Ages to face the truth. The truth and the facts of reality of the individual – not mystical probable outcomes of sample portions of populations. They need to once and for all face up to the facts of reality and the needs of the individual to have his rights protected. No amount of statistics will ever tell them that!
Discuss this Article (9 messages)
|
