Map of life expectancy at birth from Global Education Project.

Thursday, December 08, 2005

The Unscientific Method

David C. Reardon is an anti-abortion activist who has adopted a unique strategy -- to try to convince the world that abortions are harmful to the women who have them. His organization, the Elliot Institute, has a web site called AfterAbortion, which archives numerous "scientific studies" purportedly showing that abortions are traumatic and psychologically damaging, etc.

Reardon managed to get a study published in the British Medical Journal in 2002, finding that women who terminate an unwanted first pregnancy are at higher risk for subsequent clinical depression than women who carry the pregnancy to term. Now Sarah Schmiege and Nancy Felipe Russo have taken another look at the same data and found that no such thing is true after all. On the contrary, women who chose to end unwanted first pregnancies end up with higher levels of education and smaller family size (obviously) and are probably at lower long-term risk for depression as a result.

Why the difference? Reardon and his collaborator manipulated the data in various ways in order to produce their result. For example, they excluded women in the group who delivered who then had abortions later on. Since people with multiple unwanted pregnancies are probably more likely to experience depression, this had the effect of reducing the apparent incidence of depression in the women who delivered. (Got that? It's a little tricky. Depression in women who have an unwanted baby when they are young is likely a risk factor for having subsequent unwanted pregnancies. By eliminating these women from his sample, Reardon rigged the outcome.)

Scientific research, unlike theology, is not a method for demonstrating a preconceived conclusion. It is based on the principle that the truth exists independent of our preferences about what it ought to be, and awaits discovery, not invention. For the modern conservative movement in this country, that definition of science does not apply. Activities intended to promote some convenient belief -- that environmental tobacco smoke is harmless, that burning fossil fuels does not contribute to global warming, that abortion causes abortion or breast cancer -- are disguised as scientific research. This pseudo-science is typically sponsored by entities that will benefit from a particular outcome -- the oil or tobacco industries, or religiously motivated anti-abortion activists. Earlier we discussed the bogus anti-gay research of Dr. Paul Cameron. Then of course there is the Discovery Institute, which promotes so-called "Intelligent Design."

Unfortunately, very few science journalists have either the skill or the motivation to sort this out for us. Real science and tendentious pseudo-science are presented side-by-side, with equal credibility, on the TV and in the newspapers. They call it journalistic "balance." With the federal government now engaged in promoting pseudo-science, real scientists need to be much more engaged with the public.

No comments: