Jim wrote:

> He's reporting the results of the study.  Here's the link: 
> http://news.bbc.co.uk/1/hi/health/377381.stm


I believe that is the same study that you referenced here earlier. It is the
one study that does not show positive health effects from moderate alcohol
consumption (nor does it show negative health effects from it). 

About a half dozen other studies show health benefits from moderate
drinking. But these researchers don't want to believe the outcome of *their*
study is a statistical quirk, so instead they are accusing all the *other*
studies of being statistical quirks. They offer no empirical evidence to
defend their minority conclusion. They offer only their unsubstantiated
conjectures. Meanwhile the preponderance of the evidence, from all
researchers, suggests health benefits from moderate alcohol consumption.

Also the title of the BBC article reveals the bias and/or scientific
ignorance of the reporter. One study does not "debunk" the outcomes of
several other studies. It is the nature of statistical research that one's
conclusions will be wrong about 5% of the time. If the majority of studies
indicate that X is true, but one study indicates that X is false, then, all
other things being equal, it is most rational to believe X is true.

-gts