I just read the actual study.

These guys list on page 14 the variance ratios (i.e. male variance divided by female variance) of 31 countries on 5 different tests (1 PISA, 4 TIMSS). 28 test-results are missing so there are in total 127 measurements.

I don't have SPSS but the following should be illuminating enough:

In 7 tests the female variance is higher than the male variance. On 6 tests they are equal. But on 114 tests the male variance is higher than the female one.

In the Netherlands and in Marocco the average variance ratio is around 1. In Indonesia female variance seems to be greater than male variance. But in 28 other countries male variance is on average higher than female variance.

It's true however that the average score of men in some countries is lower than the female one, so maybe the greater male variation is due to the very low scores of some boys. It's important to note however that most participants were younger than 15, and girls tend to score higher on IQ-tests than boys when they are young, whereas this trend reverses as they are getting older. (Oh, and if you look at page 16 you'll see that they only list 16 countries out of 86. Is there some cherry picking going on? )

Maybe I'm missing something huge but these results seem not that promising. Not to mention other studies which showed greater male variance, publication bias and stuff :-)