By bad statistics, we mean the incorrect or inappropriate use – either from insufficient knowledge or from a calculated intention to mislead – of statistical data (and data displays) or statistical methods, in support of some arguable proposition, such as a scientific hypothesis, an advertising claim or a political point of view.
Many people find statistical data fascinating – even bewitching – when they understand the context. The long‐time popularity of Guinness World Records and Wisden Cricketers’ Almanack is clear evidence of this. However, even when they do not fully understand the context, the general public are still receptive when statistics are presented. True, that receptiveness is sometimes too trusting but, increasingly, people are resisting the notion – perhaps (mis)remembered from school arithmetic lessons – that ‘there’s no arguing with a number, especially a very precisely stated number’.
That is the bright side for everyone who would like to see published statistics better engaged with in our society.
But there is a darker side. People who know only a little about the principles of statistics may unintentionally mislead the public with inappropriate statistics or erroneous statistical arguments. More troublingly, there are individuals with a vested interest in capturing people’s trust who knowingly misuse statistics to bamboozle others for questionable ends.
How convenient it is that English has a single word that ...