Here's an article on a study that points out the rather interesting circular reasoning that can be found beneath the lot of splashy headlines about scientific studies...
http://www.physorg.com/news5617.html
Studies are usually trumpeted in the press in a very simplistic way. 'This equals that' is the usual theme. However, especially in the kind of sociological research studies that are normally used with regard to promoting various behavioral points of views or lifestyles, you can find that the simple soundbite conclusion is often actually far from even a basic understanding of what went on in the study.
They found that heavy drinkers and nondrinkers had lower IQs than moderate drinkers. However, people who live lifestyles of heavy drinking may have been less intelligent on average to start out with. Is it not possible that the choice and the ability to choose to drink moderately is indicative of more intelligent behavior? Therefore, is moderate drinking produced intelligence or is the action of drinking moderately the likely result of one being intelligent to start with?
Of course, as always, when dealing with sociological studies, they are brought to generalities and are bound to contradict specific realities in many cases. So, this is not to say that there are wildly intelligent people and wildly stupid people (comparatively) who engage in all three behaviors. This is discussing tendencies rather than creating concrete categorizations and, I think, is more useful when viewed as such.
Food for thought at any rate!
Be Well,
Michael Perez