Reporting on Science

http://www.dailymail.co.uk/news/worldnews/article-1148611/Drinking-just-glass-wine-day-INCREASE-risk-cancer-168--finds-shock-new-study.html 

This is what amuses me about scientific reporting to the lay masses. The presumption of clear cause-effect. We have correlations, but the correlations are not explanations of underlying mechanisms. Years ago it was a correlation between heart HEALTH and red wine, so it was assumed by the lay masses that a little red wine must therefore be great for you. Now there is a correlation between alcohols and cancer.

However, it is IMPOSSIBLE to isolate the mechanism due to the way the science was performed - these are large studies of populations, and each individual in that study has added more nuisance variables to their own lifestyle than could ever be rigorously accounted for in the science. Correlations are not causes, they are simply that - correlations. RISK increases, not direct cause and effect.

Note, when I say IMPOSSIBLE, I mean from within the study and methodology. Of course mechanisms can be isolated with protracted direct studies, but this is not the way this kind of stuff is being done - it is more of a broad survey of behaviors and trying to isolate behaviors and health from these cross-sections of populations.