Can the leopard change its shorts?

If the panthera is Ms. Harman perhaps not. The latest from the Fawcett Society, where Ms. Harman is newly installed as El Jefe:

Women of Bangladeshi and Pakistani heritage in the UK are earning on average almost a third less an hour than white British men, a pay gap campaigners say “should be causing national outrage”.

An analysis of pay data also reveals that mixed-race women and women of Black Caribbean heritage take home a quarter less money than their white male counterparts.

The Fawcett Society publishes the figures on what has been designated Ethnicity Pay Gap Day 2024.

Now, Fawcett is doing something that we really, really, don’t like. Which is issue the press release, get the story into the newspaper, before releasing the actual report. So it’s not possible for people - people like us - to check the figures in their report. Not until the news cycle has turned and the press release has done its job.

However, here it’s obvious enough. Those pay figures are “blended”. That is, they are of all workers, full time and part time. Part time workers get paid less per hour than full time. So, if we observe across sub-populations that have different rates of part time work we’ll get a very skewed understanding of pay differences. So much so that this has even hit Wikipedia:

In June 2009, Sir Michael Scholar, head of the UK Statistics Authority, wrote to Harman to warn her that different headline figures used by the ONS and Government Equalities Office with regards to pay differentiation between men and women might undermine public trust in official statistics. The GEO's headline figure was 23%, which was based on median hourly earnings of all employees, not the 12.8%, based on median hourly earnings of full-time employees only, used by the ONS. Scholar wrote: "It is the Statistics Authority's view that use of the 23% on its own, without qualification, risks giving a misleading quantification of the gender pay gap".

As we’ve noted more than once in fact.

But if even Wikipedia is rapping you over the knuckles for your misuse of statistics then really, perhaps you shouldn’t be doing that?

So, we do not have a definitive answer to the headline question. It might well be possible. But in the case of Ms. Harman it appears that no, new job but no new shorts. She’s still using the wrong, misleading but convenient to her politics statistics. Aren’t we all surprised?