"John H." <j_hasenkam from yahoo.com.au> wrote in message
news:58d35f79-b9da-484b-8325-40dbf63b7491 from b40g2000prf.googlegroups.com...
> Hey Peter,
>> I used to reject the cannabis-psychosis link but the data on that
> seems to firm up.
Hi John,
It is worth asking yourself, when you look at data like these, how big the
effect really is. Remember, as N gets larger, the size of the effect that is
detectable gets smaller and smaller. So, what is often praised (large N)
probably needs to be rethought, though this is very unlikely to happen
anytime soon (though there seems to be growing insistence on within-subjects
comparisons where possible, though the reliance is still on statistical
inference). Also remember, if one person out of 100,000 becomes psychotic
normally, and among pot smokers it is three out of 100,000, the newspaper
headlines will read "SMOKING POT TRIPLES RISK OF PSYCHOSIS!" Obviously,
while true, your chance of becoming psychotic is very small. You probably
know all this stuff (and you do point out, below, that the risk is low) but
it doesn't hurt to be reminded, and I can't help but take every opportunity
to denigrate null-hypothesis statistical testing. And, incidentally, this
does not mean that one cannot do between-groups experiments. In the lab
where I was trained, between-group experiments were nonexistent. Later, they
started asking questions that could only be investigated using
between-groups experiments. They might have six subjects in each group; if
you get five in one group that shows an effect and, say, one in the other,
you start to think you are on to something, though six and zero is better.
Needless to say, an ANOVA would show this to be "highly significant," but at
that point who gives a shit? On the other hand, if you have 25 subjects in
each group and obtain a significant p-value, the "effect" could be quite
trivial, even if replicated.
>Moreover I've known a few people who have had
> sufficiently severe psychotic episodes that they have sworn off the
> weed forever. Another toke and they just sit in a corner losing their
> religion. This does relate to heavy and sustained use, and the risk
> is still rather low. Work and associated stress, living in or working
> in CBDs(pollution), does more health damage than pot ever could. In my
> view a good puff once a week is a very good way to ward off dementias.
> The evidence on this is much stronger than the cannabis-psychosis link
> but of course you won't see that evidence plastered over the front
> pages. Gotta keep those peasants working. As Andrew Schmookler wrote,
> "A man at peace with the world is an instrument of limited utility but
> frustrate him enough and you can bend him to society's ends." (Parable
> of the Tribes).
Marvin Harris' explanation of the witch hunts in Europe relied heavily on
this notion. Better to be standoffish with respect to others with whom you
might otherwise bond and join together to exert "countercontrol"; after all,
instead of bonding they might become pissed at you and tell the authorities
you are a witch.
>> I don't smoke that much now, in the past I didn't mind because, just
> to contradict myself, I actually worked quite well while stoned. Too
> old now, I just wanna sit back and relax but that gets too boring.
> With discipline thinking on difficult problems while stoned can be
> productive.
I have heard this, though, of course, I have never smoked the evil weed!
<snip>
