Friday 8 December 2006

I'm not missing, I'm just good at hiding.

(Warning: statistics talk ahead.)

The newspapers would have us believe that there is a "missing men" problem. The "sex ratio" is the number of men per 100 woman, and it is less than equality. This is true, but which number to believe? I've seen statistics place this at near 95 or 96 males per 100 females , although experts believe it to be around 98/99 males per 100 females.

According to the Census counts, there are 1,965,621 men and 2,062,328 women, or, if you like, 96,707 more women than men. Oh no! There are 5% more women! (The sex ratio here is 95.3.)

First note: we can only survey so much, with error. Consider the number of people overseas when we take the Census, for example. If lots of men are away and not many women...
Second note: women live longer, so there are more older women than men.
Third note: You have to live here a year to be considered a resident of New Zealand. If women come and stay (but men don't) then the number of women, of course, increase. (Although this should be observable in drops in women counts, or increases in men counts for that matter, in other countries, and this isn't true either.)
Fourth note: men count funny. Consider Mr C. He leaves to live in the UK, and so when he goes, he fills out the departure card to say "permanent stay overseas". So we subtract one man. However, he comes back within the year (as a "temporary visitor"), but leaves again for a "permanent stay overseas". Subtract one man.

As you can see, the numbers won't necessarily add up to reality. The "missing men" problem is a problem with the counting, not with actually missing men. It's an interesting question of statistics about how to measure it properly, but not one to cause great alarm to the nation (despite the attention grabbing headlines).

(Also, there obviously aren't enough missing men, 'cos no women are throwing themselves at me...)

No comments: