As a practicing clinical psychologist, I have been bothered for sometime about a tendency in my profession to use absolutely abominable research to "prove" a social hyptohesis. For example, when I was in graduate school, we were taught that for psychological research to be considered competent a researcher had to have a well-defined working hypothesis. "Well-defined" means "operationally defined", that is, you had to be able to define the working hypothesis in ways that were objectively measurable. Secondly, social research, as with all research, began with the premise that the "null hypothesis," the hypothesis that the working hypothesis was invalid, was correct. Hence the job of any research was an attempt to disprove the null hypothesis Contemporary social science has turned this on its ear, often working to prove the null hypothesis rather than disprove it. In other words, much research works from the presumption that the experimental hypothesis is correct and the null hypothesis is invalid, the opposite of what should be done. This, combined with poorly defined experimental hypotheses, has created a real problem in social science research.
Nowhere is this more evident than in the area of gender psychology and the psychology of the LGBT culture. Now I'm not saying that the conclusions these researchers are reaching are necessarily wrong. What I am saying is their methodology is so slipshod that we don't have enough of a reliable and valid research base to make an informed judgment. Sadly, this "research" is often used to define social policy in our legislative offices.
Here's a link to an article from a Swedish writer that describes the problem as it is experienced in that country. Perhaps this should give us some pause for serious reflection on the potential impact of the lack of scientific rigor in the work of some social scientists.
For those that are interested, an interesting resource on this and similar issues is:
Wright, R. H., and Cummings, N. A., eds. (2005). Destructive trends in mental health: The well-intentioned path to harm. New York, NY: Routledge.
Also, a classic volume with strong implications for social science research is:
Rychlak, J. F. (1981). A philosophy of science for personality theory (2nd edition). Malabar, FL: Robert E. Krieger Publishing Company.