On an entirely separate note, all this talk of inter-gender differences is useless without some consideration of their scale relative to intra-gender difference

On an entirely separate note, all this talk of inter-gender differences is useless without some consideration of their scale relative to intra-gender differences. But isn't that exactly what the author is suggesting? That the within-group variance tends to be higher in men than in women, even if the differences in averages between the groups are not significant?