saying a lot about little – another example of how not to use statistics

Ben Goldacre has written an interesting post on a ‘news’ item comparing pay scales for UK workers in the public & private sectors. The original story drew a number of comparisons between the two, several of which turn out, on closer examination, to be spurious. For example, the item comments that public servants work fewer hours than those in the private sector – but as Ben points out, in the UK there is a higher proportion of part-time workers in the public sector, & those doing the analysis hadn’t bothered to distinguish between the two but lumped them all in together. The story also suggests that public sector employees in the UK are paid more than their peers in the private arena. But this comparison is also a tricky one, because many state-employee roles don’t have an exact match with private sector positions (policemen & firemen, for example). And this works both ways: some of the lowest-paid UK private sector employees work in retail – not something the state sector is involved in.

So, another example of the need for caution when interpreting statistical data. (And worth remembering next time private:public comparisons are made here in NZ…)

Leave a Reply

Your email address will not be published. Required fields are marked *