Friday, April 29, 2005

Another example of "you really should call them on their b.s., or where did they get that number from?"

I had this item forwarded to me by my library director, who has her personal online subscription to the Wall Street Journal. The item in question was published in the April 28, 2005 issue in the Numbers Guy column. Since I got the link from her, posting it here will likely not work after a while, but I am sure any librarian with a good database can trace it back. The citation would be:

Author: Carl Bialik
Article title: "Debate Over Gay Foster Parents Shines on Light on a Dubious Stat."

The article summarizes how Cathie Adams, president of the Dallas-based Texas Eagle Forum, an advocacy group, misused a statistic that is basically flawed and a misinterpretation from a study, that to put it mildly, was done by a biased reasearcher with flawed methodology. Bialik's clear explanation provides a perfect example of why we need to question whenever possible any numbers that get bandied about in a soundbite. We should at least be sceptical when some "advocate" decides to throw around a number to "prove" their agenda. The article explains how the number is flawed because for one, the math behind it is wrong. When Ms. Adams was confronted about her error, she simply said that she did not articulate that as well as she could have. Really? The number in question, that children in foster homes with same-sex parents are 11 times as likely to be sexually abused as those with heterosexual parents, is a bit inflamatory for one. You would think someone would want to make sure it was at least accurate before using it. Then again, it seems the sensationalism of it was the whole point. In the CNN interview, the responder, Randall Ellis, executive director of the Lesbian/Gay Rights Lobby of Texas, conceded he had never heard of the study. The main flaw of the study, among its many, is that the source of information the researcher used did not provide for accounting if households were homosexual or not. In other words, the acts, carried out by men on boys or women on girls, do not automatically follow that the perpetrators were homosexual. A crucial detail that was conveniently forgotten, and that the anchors on CNN simply did not call Ms. Adams on it.

What irks me, besides the fact that most people with common sense would likely question such a number, is that they basically allowed Ms. Adams to slip in her flawed number. Since she was not really questioned about it on the air, it means that most people watching it will likely believe it, and some will likely quote the "lady who was on CNN" in promoting their agenda. They probably fail to notice that various studies on the subject of pedophiles and homosexual behavior found that offenders can be homosexual, heterosexual or bisexual. A quick search in Academic Search Premier revealed various articles on this issue. One particular article, in its literature review, states that "findings suggested discontinuity between pedophilia and adult-age sexual orientation. The factors that determined whether a man was oriented toward children (preferring boys or girls) differed from the factors that determined whether a man oriented toward adults preferred men or women." And this is just one example found on a very quick search. I happen to read around enough to know that pedophilia and homosexuality are not automatically connected, but since I am not a researcher like most "experts," I don't exactly carry the specific statistics with me every where. What I know how to do is research and investigate when something sounds fishy, or when I have a hunch something does not sound quite right, and in this case, it did not. Sure enough. The article I used for my example above was published as follows:

Title:Psychological Profile of Pedophiles and Child Molesters.
Authors:Murray, John B.
Source:Journal of Psychology; Mar2000, Vol. 134 Issue 2, p211, 14p

I just typed a quick search in Academic Search Premier with the terms pedophilia and homosexual* (the asterisk is truncation. For those not librarians, it means it tells the database to search for variants of the word, singular, plural, so on). I am sure a deeper search in PsychInfo or other more specialized database would yield more results. Which begs the question. If I can do a quick search of the literature, why can't they? Who the heck checks up on facts for these anchors and the people they let into their programs? Why do they get a free pass? The problem with that free pass, if you will, is that others then use that flawed information to make decisions such as the bill that passed in the Texas House preventing same-sex couples from being foster parents. At the moment, since the Texas Senate passed a conflicting bill, they are "debating how to proceed."

If readers want a different perspective, they should see the clip from The Daily Show where Jon Stewart looks at the debate between Ms. Adams and Mr. Ellis. Like Bialik, I will warn that the clip may have language objectionable to some. But Stewart does ask as well at one point why the CNN anchors did not question the number while on the air. I think it is worth watching after reading the Wall Street Journal piece. Look for the "Gaywatch" segment once you get to the site. Personally, I try to remain somewhat on the middle when possible, but I totally despise people who not only use flawed research, but have no shame about it, to advance an agenda. It is an example of the misinformation some people use hoping no one will notice, and more often than not, people don't notice. Yet, there are some who do notice, up to us then to tell the rest.