Sunday, November 23, 2008

Junk Science-Fuelled Triumphalism Interrupted

Bad conclusions drawn from bad science

In what some fervent demagogues continue to insist is a "war" between religion and atheism, more and more "warriors" are continually in search of what they believe is the metaphorical magic bullet that will win the battle for them once and for all.

In the form of a 2005 study, certain individuals seem to think they've found it.

In a recent study published in the Journal of Religion and Society, anthropologist Gregory Paul identifies what he believes to be a correlation between religiosity and poor social health -- marked by higher rates of crime, STD infection, teen pregnancy, abortion, and early adult and juvenile mortality.

Unfortunately for the demagogues, however, it turns out that there are several problems with Paul's study that should take the wind out of the sails of those using it to denigrate religion.

First off, the study itself is inherently and fatally flawed. It relies overwhelmingly on statistics as opposed to any kind of experimental research or case studies. In particular, the statistics used for violent crime rates and STDs are particularly problematic for this study -- such statistics frequently turn out to be far less reliable than Paul touts them to be, as many violent crimes are recorded in crime rate statistics only when charges are laid and a conviction attained. Many violent crimes (exempting, of course, murder) go unreported in western democracies. Likewise, STDs are only recorded when those suffering from the disease seek treatment. The shocking number of carriers who remain unaware they have an STD or choose not to seek treatment cast serious doubt over the results of Paul's research.

Moreover, Paul's research seems to ignore some of the world's most religiously devout states: no mention at all is made of countries such as Turkey, Iran or Saudi Arabia.

If the conclusions being taken away from Paul's work contained the merit some are crediting them with, one should expect theocracies such as these to have even higher rates of social dysfunction than the United States. Yet when examined by the outsider, the rates of "social dysfunction" (at least as defined by Paul) are much, much lower in these countries.

While the often-oppressive nature of these states demonstrates their own particular form of social dysfunction, the case of these countries actually offers significant cause for doubt in Paul's conclusions. Naturally, there are other factors underlying the low levels of social dysfunction in these particular countries. As such, while their low levels of social dysfunction (as it is defined by Paul) correlate with their high level of religiosity, the causational relationship between the two is extremely ambiguous.

Which brings one back to the number one reason why the conclusions being drawn from Paul's study are so comically remiss: correlation is not causation. In fact, this is one of the primary rules of any science, be it a social science or physical science.

Paul's study compares rates of religiosity to phenomenae of what he defines as social dysfunction. Yet other factors -- such as alcohol and drug use -- have been shown through various research methods, ranging from statistical research to individual case studies -- to be a very strong precipitating factor in each and every social phenomenon Paul identifies.

Likewise, factors such as poverty and level of educational attainment have proven to have stronger correlatory relationships with these forms of social dysfunction than does religion.

It doesn't take a PHD in any social science to find all the holes in Paul's research.

Unfortunately, many of those touting his research as evidence that religion is "corrosive" are simply too motivated by the prospect of an ideological triumph to closely examine his work, blemishes and all. Consider this particular article from Slate magazine wherein Paul Bloom makes himself utterly transparent when he describes the more secular countries of Europe as "atheist societies".

If correlation were causation -- or at least if the forms of social dysfunction Gregory Paul chose to study were even relatively simple matters -- some of the triumphalism may be justifiable.

But unfortunately for the aforementioned demagogues, facts are quite different on each count.

Pretending that atheism makes on intellectually superior is one thing. But when such obnoxious triumphalism is based on a study as flawed as Gregory Paul's, it simply leads one to question the intellect of these individuals altogether.

No comments:

Post a Comment

Post your comments, and join the discussion!

Be aware that spam posts and purile nonsense will not be tolerated, although purility within constructive commentary is encouraged.

All comments made by Kevron are deleted without being read. Also, if you begin your comment by saying "I know you'll just delete this", it will be deleted. Guaranteed. So don't be a dumbass.