Jim:

> Conjecture is a guess. 

Or you can say it is an untested theory.

> This rises from mere conjecture to a
> theory

There is little difference between conjecture and untested theory.

> (which may be fact) proposed not by me, not by the BBC
> writer, but by the researchers of the
> British-Medical-Association-published Bristol study. 

And like all researchers they are inclined to defend and promote the
importance of their own study. They don't want to believe their results were
merely a statistical anomaly, and so they speculate about a theory that
might explain the apparent anomaly. Nothing wrong with that... It could
inspire someone to do a controlled test of their theory. But unfortunately
in the meantime that theory is no more substantiated than the theory that
life exists on Mars. 

> They found no health benefit from the consumption of moderate
> alcohol, that the U graph is more like a J.
> 
> There may be many theories to explain away this finding.  I
> cannot think of any and am asking you to enlighten me.

It is in the nature of statistical testing that sometimes differences
between two samples will not be found to be statistically significant even
when the two samples are in reality intrinsically different. In this case
those two groups are "non-drinkers" and "moderate drinkers". These Brit
researchers found no statistically significant difference between the
non-drinkers and moderate drinkers they studied, but the majority of studies
have found a statistically significant difference between non-drinkers and
moderate drinkers. 

So what are we to conclude? If we are going to accept one hypothesis or the
other then obviously we should accept the hypothesis supported by the
majority of studies, which show health benefits from moderate alcohol
consumption.

The important thing to remember is that these are studies about large groups
of people, and we cannot infer that what is true on average for a large
group of people is always true for everyone in the group. Each individual is
different. 

-gts