Apparently, the APA is standing by its meta-analysis of 59 studies of the effects of homosexual parenting, with a couple of very interesting quotes:
The author of the report, University of Virginia psychology professor
Charlotte Patterson, fully noted the limitations of the studies, said Mr. Herek,
"which was that when you compare children raised by lesbian or gay
parents … they really don’t look much different from kids raised by
comparable heterosexual parents.”
Now without getting into the debate itself, the limitations of the APA study are that few if any of the studies they cited had either sufficient sample size or a control sample. The claim, then, that the outcomes for the children of homosexual parents are similar cannot be sustained for a very simple reason; the hypothesis has not, by the APA's own admission, ever been tested.
If this is representative of the APA's thinking at its highest levels--preposterous claims that can be refuted by a simple look at their methodology--we may be entering a dark age in terms of helping the mentally ill, to put it mildly.
Ire On Cue
-
A “plan” for a “masjid [mosque]-centered development” in Lino Lakes dropped
into the chat this week. And it’s generating the usual clouds of ire from
the u...
16 hours ago
2 comments:
The reason the APA is standing behind its meta-analysis because of a bedrock principle of how science is done: absent evidence that disproves a null hypothesis, the null hypothesis is assumed to be true.
If the question is "does the sexual orientation of the parents impact outcome X in childhood development," the null hypothesis is that it does not.
Now, Loren's criticism that the studies may be underpowered is fair--insofar as that is a potential criticism of any study with a negative result. However, this proves nothing. It negates nothing. And it certainly doesn't prove the opposite of what the meta-analysis indicates.
Brian, Loren's criticism goes beyond an insufficient sample size; it also notes a failure to establish meaningful controls, a failure to share data, and a failure to establish concrete measures.
Yes, you keep the null until (Regnerus) contradicted statistically, but in this case, the APA owed it to its members and the public to state "the studies listed are such a mess, we do not dare to make any conclusions."
Post a Comment