Gender, pay and 'misleading' stats
Equalities Minister Harriet Harman has been accused of over-stating the plight of women in the workplace: using misleading statistics to make it look as though female workers are having a tougher time than they really are.
After the debacle over the use of knife crime statistics last year, one would have thought that ministers might have learned their lesson.
But I am reliably informed that when the National Statistician Karen Dunnell went to the Government Equalities Office last November and told them that their way of calculating gender pay differences might be confusing and potentially damaging, the GEO ignored her and published anyway.
So instead of the Office for National Statistics (ONS) pay-gap figure of 12.8% (hardly something to crow about), the department which stated that women are paid, on average, 23% less per hour than men.
Now by Sir Michael Scholar, the chair of the UK Statistics Authority (the official watchdog on the use of government stats), saying that her use of the 23% figure "may undermine public trust in official statistics" and "risks giving a misleading quantification of the gender pay gap".
The GEO's version of events is rather different to that of my source. They claim that they "ran the 23% past the ONS and they approved the calculation".
They may have ticked the maths, I pressed, but did they approve the use of the figure? "As far as I am aware," said a spokesperson, "no-one from the ONS has ever suggested we should not use the figure." However, the official promised to check.
What the GEO and ONS agree is that the UK's official statisticians are currently reviewing how they can best present the gender pay gap - a review which is ongoing.
Both the ONS and the GEO number come from the same data source, but the equalities department and Britain's top statisticians interpreted them very differently.
The 2008 Annual Survey of Hours and Earnings (ASHE) attempts to provide the detail needed to get a true picture of how men and women fare in the workplace. For me, this is a key table:
Comparing men and women who work full-time, the gender pay gap is 12.8%, the ONS preferred measure.
Looking at part-time workers, women actually do better than men: their hourly rate is 3.4% higher than their male counterparts.
But when you add the two together, because part-timers get paid less than full-timers and because there are nearly four times as many part-time female workers as there are male, the gap appears to jump to 22.6%, which the GEO rounds up to 23%.
The GEO justifies its approach to me in a short statement:
"The 23% gender pay gap figure used by the Government Equalities Office includes both full and part-time employees. With women representing over three-quarters of the UK's part-time workforce, we believe this figure gives the fullest picture of the country's gender pay gap."
Nowhere in the press release, though, is the point made that those part-time workers are actually outstripping men who work part-time.
There is clearly some quiet fury at the ONS that Harriet Harman should have apparently rebuffed the country's foremost statistician. My source tells me: "The most important point is that the GEO has no statisticians inside it." The question, then, is: who did approve the 23% figure? One of Ms Harman's officials promised to find out for me, but did say that the decision was made "across government".
If one accepts my ONS source's version of events, this was not a professional difference of opinion between two statistical experts. Harriet Harman's officials preferred their in-house interpretation of the data to the independent and professional one because, one might assume, it made the case for their controversial Equalities Bill look a little stronger.
Attached to the letter from Sir Michael Scholar are the notes from the Monitoring and Assessment team at the authority which investigated the case. This suggests that it is not just the GEO which may occasionally get political with the numbers.
The Equalities and Human Rights Commission refers to a gender pay gap of 35.6% for women working part-time. It comes to this conclusion by comparing the mean hourly earnings of female part-time workers with those of male full-time workers.
As the Statistics Authority document so delicately puts it:
"While we see value in providing a range of measures to present the differences between the earnings of women compared with men, a gender pay gap that compares the hourly earnings of women part-time employees with men full-time employees needs particularly careful explanation and justification if it is not to mislead."That there is an issue about the pay gap between men and women is not disputed. But there must be a danger, as Sir Michael Scholar says in his letter, that throwing around unofficial and misleading numbers is "likely to confuse the general public" and to "undermine public trust". It also makes it more difficult for people to understand what is really happening in the workplace.
I'm 
Comments
or to comment.