Last month I read a short newspaper article quoting an insurance company. The piece stated that in the UK dancing teachers and choreographers were the most dangerous drivers on the road covering on average 23500 miles between motoring convictions with dental technicians amongst the safest. The piece prompted several negative comments concerning the interpretation of the data, the definitions of safe and dangerous driving and the use of statistics in general. Thinking about these discussions led me to a blog by the economist Jeff Ely in which he notes "......that statistics is a rhetorical practice. The goal is not just to convey information but rather to change minds. " In an ideal world we would all be capable of taking any amount of raw data, analysing it in our own favourite way and arriving at the same conclusions, job done. But the real world contains people like us with prior perceptions, variable knowledge and limited attention spans. We desire certainty, simplicity and clarity of message or we are distracted and move on. Nuances and caveats undermine the power of conclusion. The inevitable tendency is to push data into the background and interpretation into the foreground.
This leads to the dilemma that in order to maximise the impact of the truth one has to filter it. This line between filtering and distorting is a dangerous one to cross.
Science is founded on the unchanging Laws of Nature. Technology help can address the ever changing socio-economic global challenges that confront us but we rarely face decisions to be made with the clarity of certainty in the evidence. Probabilities are just that, probabilities. Statistics may be the heuristics of extrapolating data but it is a much needed discipline.
So who should control the statisticians? It might be relevant to the entire human race and not just dance teachers and choreographers.