This is definitely a handwaving response, and you would be a shoat short of a litter if you used my opinion as a basis for family planning. For one thing simple statistical theory (probability theory if you like) means that human fanilies are too small for high degrees of confidence. Suppose that the sex ration for a given couple were either 50% or 100% males; they had no daughters, what will the gender of the next child be? OK they have a son. Which of the two types are they? All male or 50%? We don't know. Even a 50% ration should kick up a boy half the time. You come back 10 sons later, no daughters. Now what? Well, it doesn't look good for daughters, but even in families with a 50% ratio, you would expect such a result about once in 2048 10-child families.
Surprising, but not really amazing. You could expect half a dozen such cases from a large football crowd!
And then, just as you were planning on a cheaper-by-the-dozen prospect, or getting odds from a bookie that the next would be a son, along comes the daughter!
But never mind that. All sorts of things could affect the real probability. The problem is that if you can find a variable that every biometrician and obstetrician and geneticist will agree on, you are one ahead of me. In rats, maybe. In humans, tell me when to expect it. Some claim that the stage of the menstrual cycle makes a difference, others deny it. Vaginal acidity is supposed to favour females, or not so much in humans as in rats. Take your pick.
At least some staticians argue however, that there seems to be a slight but significant excess of human families in which all the children are the same gender. That suggests to me at least that it is likely that certain factors can bias the ratio. It does not follow that it is always the same factor. But where there is such a factor, there must be a physiological cause, whether biased by behaviour or not.
So the upshot is that if you want to bet on your next child being a male, that is the way to bet. However, it also suggests that the bet had better not be more than you can afford to lose!
