Making Sense of Statistics
At the moment we are doing a lot of work with careers guidance professionals. And part of this work is around the use of Labour Market Information in the guidance process. What jobs are available? What are future trends in employment? How much can a person expect to earn in any occupation? What qualifications are needed?
Much of this information relies of statistics. There are a lot of statistics provided by governmental and other agencies. In the UK the data.gov.uk web site is providing increasing access to data and encouraging visualisations, mash-ups and reuse.
But statistics require interpretation. By chance this morning I stumbled on a tweet –
DrEvanHarris Exposing dodgy various claims on average start grad pay http://bit.ly/gmjT5x by @Straight_Stats – reported uncritically – even by FT
Following that link took me to the excellent Straight Statistics website. The home page says:
We are a campaign established by journalists and statisticians to improve the understanding and use of statistics by government, politicians, companies, advertisers and the mass media. By exposing bad practice and rewarding good, we aim to restore public confidence in statistics.
The tweet from DrEvanHarris led to an article by Nigel Hawkes entitled Questionable Claims on Government Pay.The recent campaign against rises in student fees in the UK has focused attention on how much graduate earns. And the article suggests that many of the figures quoted in UK newspapers may give an inflated impression of graduate starting salaries because of the way these figures are compiled.
The Straight Statistics website provides a number of excellent and free resources including a simple guide to numerical and statistical traps, Making Sense of Statistics. a simple guide to numerical and statistical traps, Although the guide is “ primarily designed for journalists and press officers”, the web site says it may be interesting to others as well. And indeed it is, providing clear examples of how statistics can mislead.
One of the issues we have looked at in the Labour Market Information for careers guidance is the impact of gender on pay. Nigel Hawkes explains this depends on how the figures are collected:
How big is the gap between the earnings of men and women? According to the Office for National Statistics (ONS), it is 12.8%. But the Government Equalities Office (GEO) says it is 23%. And the Equality and Human Rights Commission (EHRC) says it’s 17.1%1.
The differences in these figures arise from the different methods used to produce them: the ONS includes only full-time employees, excluding overtime and part-time workers. The GEO includes part-time workers because it says more women than men work part-time and it is wrong to exclude them. The EHRC figure uses the ONS data but compares the mean salaries not the median. It justifies this by saying than men are over-represented at one extreme of the earnings range, and women at the other.
Three figures – all of them right – but asking what is being compared and how it was calculated tells us why there is a difference.
Well worth a read!
