It is 20 years now since I looked at the actual frequency distributions of male & female earnings. When I did, I got a surprise. For it is not just the averages that were different, but the whole of the distribution
The female distribution was much closer to a Bell curve, with a very narrow spread & a high peak close to the mean
For men the distribution was positively skew, with a much wider spread, a lower, flatter peak & a very long tail out towards the very highest earnings
The differences between male & female variances are sometimes touched upon, but not nearly as often as the averages. It is well known, for example, that men are more likely to figure in both the highest & lowest classes of Oxbridge degrees. The concentrations of lead in blood have also been found to show a similar pattern, to that of incomes, at least before lead was taken out of petrol
You do not however often see the whole distributions being compared, so it is difficult to say whether the hypothesis that Women are Normal, Men are Lognormal has any legs
