AA

[Brooks Fudenberg]

Forgive a naive question, but I wonder what is the measure of success
for AA?  More precisely, what is the population from whom the
success rate is determined?

Someone told me once that AA has a 100% success rate for those
who follow the program.  "Following the program" was defined as (among
other things) not drinking.  By this criterion, I would think that every
anti-drinking program is 100% successful.  (Similarly for anti-drug
programs, anti-smoking, etc.)

Certainly(?) my colleague was mistaken, or in the intervening years, newer
definitions have been proposed.

But I would think the definition of the population may be a difficult
question.  If we include everybody who goes to AA, that would *seem* to
include many people who were not "serious" about ceasing to drink, and
our "success rate" would be artificially low.  If we
only include those who refrain from drinking, we are tautologically at
100% success.  If I had to guess, I would guess the basis for current
statistics is "success rate is defined against all those who attend even
one meeting."  But I could well be wrong, and there could be more
sophisticated measures.

On Tue, 23 Dec 1997, Robert O'Brien wrote:

> Paul Salamanca offered:
>
> >  But AA has a fairly strong record of success,
> > particularly in light of its cost and its accessibility.
>
>
> I do not claim to be up-to-date on the scholarship, but my impression is
> that the success rates of those who attend AA and those who do not are
> about the same.  I am not prepared at this point to cite studies.  Can
> Professor Salamanca cite disinterested scholarship to support his claims?
>


Brooks R. Fudenberg
University of Miami School of Law