Being gay has nothing to do with the propensity to abuse children
A child in a straight foster home with an adult male is much more likely to be abused.
"Sexual abuse of children is almost always committed by men"
Logic dictates then that men should not be allowed to care for children.
the majority of pedophiles are heterosexual males.....

Such a display of statistical illiteracy in one thread (though in TIA's case the error may have been made tongue in cheek to expose it).

(1) You should make clear if you speak about absolute numbers or percentages. Absolute numbers are often very misleading. Of course, trivially the majority of pedophiles are heterosexual males because there are so much more of them than homosexual males. Trivially true but completely irrelevant (no wonder, Ake didn't feel like pointing out the obvious reasoning error implicit in that statement).

(2) You should not mix up relative risk increase and absolute risk increase. Relative risk increase can be extremely high on a very low absolute level. That error is made (pointed out?) by TIA ("logic dictates then..."). The probability that a child is abused by a male (foster) parent is much higher than it being abuse by a female (foster) parent. That's a relative risk increase by a factor of ten or more. But this alone is not a good basis for a policy if the absolute risk increase is extremely low. In other words, the probability given that a child was abused that the perpetrator was male, is very high. The probability that given a person was male that he'll abuse children is very low.

I see some disputing numbers because they fear if they don't dispute them that a policy they would not wish for would be adopted. These are two extremely different things. One is an argument on the basis of facts and the other is an argument about what should we do knowing facts. Let's assume for the sake of the argument that male homosexual couples are relatively more likely to abuse children (assumed fact). Should follow from that assumed fact that we should generally disallow male homosexual foster parents. Not at all, if the absolute risk increase is still very low.

For instance we still allow males as teachers though we know that the relative risk increase is very high compared to female teachers for the absolute risk increase is still low. By far, most teacher perpetrators (sexual abuse) are males but also by far most male teachers are not perpetrators.

So even a higher relative risk alone is usually not used as a basis for discrimination (in this example, of male teachers). I have seen no good argument from Ake why it should be used in this particular case (male foster couples). But some of the counter argumenters have assumed without good (or any) reason that there is no relative risk increase (Being gay has nothing to do with the propensity to abuse children) at all. The motivation I see for such statements is the fear that if that was wrong a wrong policy would necessarily follow.

There is no good basis for LilyFestre's claim. I link to a fairly recent abstract here. There are several similar data of that kind. But keep in mind that this is on a very low level of absolute risk increase.

There is one counter argument to such data which may or may not convince you: Being homosexual is defined as being attracted to same sex adults. A male abuser of a male child is quite often not interested in adult males. Therefore he is not a homosexual (in the above definition). People using this definition of homosexuality of course can make the statistical claims repeated by Lilyfestre. But such statements are not very helpful in this context. If one would define heterosexuality in a similar way (attracted to other sex adults), some of the statements in this thread also would have to be considered as completely wrong for male heterosexual (in the broader sense) child abusers are often not interested in sex with mature women.

Wolfgang