In Risk Management we use statistics a lot. This is something we have in common with people involved in medical and other areas of scientific research. Unfortunately, however, we are also in the company of politicians, journalists and salesmen: groups generally regarded, rightly or wrongly, as untrustworthy. That is probably why so many people do not trust statistics.
Now, there are two reasons why I tend to be sceptical of statistics or anything else, until I have good reason to believe.
- Firstly, I am not a very trusting person in my nature. My patron saint is really St John the Evangelist but some people think it should be St Thomas the Doubter. Well, that is fair enough, because, like Thomas, I am usually sceptical, but am open to being convinced by reason and evidence. (Remember that Thomas eventually believed in the Resurrection when he had seen the risen Christ with his own eyes).
- Secondly, I have spent many years in Internal Audit and also many years dealing with liability claims. I have come across many frauds, scams, and dishonest claims. Wherever there is an opportunity to make a dishonest penny, or several, there is always someone there to make the most of it: and if every claim I have seen for tripping on the pavement was true, you ought to see someone fall down nearly every time you go out!
However, I do not distrust statistics in particular. Dishonest people use anything they can: facts, words, pictures, quotes, the Bible, and of course statistics. There are also many people who use statistics lazily and mislead others without intending to be dishonest, because they do not think clearly about what they really mean, or they have misled themselves into jumping to conclusions which the facts do not really support.
Yet all these things can be used rightly and in a way that informs rather than misleads. The Victorian Prime Minister, Benjamin Disraeli is supposed to have said there were “Lies, damned lies and statistics”, but the writer Andrew Lang said of someone “he uses statistics as a drunken man uses a lamp-post: for support rather than illumination”, thus blaming the user rather than the “lamp-post”. Well, I want us all to gain illumination as we use statistics whether in Risk Management or in anything else. They can help us to:
- establish facts
- reduce guesswork
- gain a sense of proportion
- identify the most important issues
- find what works and what does not.
Fortunately, we can avoid being misled when we know about a number of giveaways which I learned about a long time ago and have always found very useful. So I am going to write a series of articles showing how statistics can be, and often are, misused, so as to put you on your guard and enable you to sift the truth from the lies, and especially from the half-truths, and to help you avoid inadvertently misleading yourself and others. I hope you will find this fun, but it will also give you power.