markmcdo said:
I can see how all 1 and 2 child families are 50/50, but wouldn't 3+ child families have more girls? I know I'm missing something but don't know what.
Yes, all the 12-child families would have FAR more girls. But there aren't many of them. By contrast, there are a LOT of families with only 1 boy that were forced to stop early, which balances it out.
Look at it this way:
A given set of N families are trying to have 4 children.
A) 1/2 of them have 1 boy, and are forced to stop
B) 1/4 of them have 1 girl, then 1 boy, and are forced to stop
C) 1/8 of them have 2 girls, then 1 boy, and are forced to stop
D) 1/16 of them have 3 girls, then 1 boy, and stop because they have their desired 4 children
E) 1/16 of them have 4 girls, and stop because they have their desired 4 children
A) produces N/2 boys
B) produces N/4 boys and N/4 girls
C) produces N/8 boys and 2*N/8 girls
D) produces N/16 boys and 3*N/16 girls
E) produces 4*N/16 girls
Total boys:
= N/2 + N/4 + N/8 + N/16
= (8N + 4N + 2N + 1N)/16
= 15/16 N
Total girls:
= N/4 + 2*N/8 + 3*N/16 + 4*N/16
= (4N + 4N + 3N + 4N)/16
= 15/16 N
In fact, the same holds true of families trying to have ANY number of children-- the total number of boys actually produced is the same as the total number of girls actually produced.
The only way it would affect things is if certain families were actually more likely to produce girls than boys, or visa versa. That is, if it weren't always a 50% chance of having either gender.
DaveE