Hang out the flags!
The great day has arrived. The first UN World Statistics Day, that is, which naturally falls on 20.10.2010. The bunting is out, and statisticians everywhere will emerge from their data-mines, blinking through their pebble spectacles. They will throw off their drab grey suits and dance in the street, discard the row of pens in their top pocket and trim their unruly hair, their dull monotone will turn to song and they will stop staring at their shoes.
All this is fantasy of course. As usual, a meagre story such as a spending review will grab the headlines and statisticians will sigh and retreat to their customary obscurity. They are used to getting little credit: the recent Times Eureka list of 100 UK scientists did not feature anyone for their statistical work.
But the UN are right to hold a celebration of statistics. Statisticians such as Richard Peto have saved countless lives through working tirelessly on testing medical treatments and investigating causes of diseases. Put “D R Cox” into Google Scholar and see the 20,000 citations to a single academic paper on methods for analysing data on length of survival, whether the data are about patients with cancer or failing businesses.
Medicine is one discipline that treats statisticians as equals, and it is no coincidence that health is years ahead in the appropriate use of evidence to make decisions. For over 60 years doctors have been testing medical claims by randomly allocating patients to competing treatments, but such ideas are still struggling to get established in education, criminology and other areas which desperately need a proper basis for policies. An inquiry criticised the climate change modellers at the University of East Anglia for their statistical naivety, while if economic modellers had fitted their theoretical ideas to the real world, instead of trying to do things the other way round, we might not have needed today’s spending review.
But statistics should be more than celebrated. Some basic statistical skill should be something we all aspire to, something to help us sort the meaningful from the meaningless. Today, the Royal Statistical Society is launching a 10 year campaign for widespread, public ‘statistical literacy’. The data that is collected and the manner of communication are all human choices, and this can be done either well or badly. If we can deconstruct statistics, absurd newspaper headlines about the latest cause or cure for cancer will provoke the derision they deserve. I always teach that any health scare story in the paper should always be taken with a pinch of salt – but not too much because I read somewhere that it increases the risk of heart disease.
So as World Statistics Day passes almost unnoticed, spare a thought for those labouring away at their lonely task before eventually retiring to Dunsummin, the Home for Distressed Statisticians, their days finally numbered. But their time will come. Probably.