Willpower

haven’t figured out how to factor out all the confounding variables.
    AA’s defenders note that alcoholics who frequently attend AA meetings tend to drink less than ones who attend infrequently, but the critics wonder about cause and effect. Does frequent attendance make people more likely to abstain, or does abstinence make people more likely to keep attending? Perhaps the ones who fall off the wagon are too ashamed to keep showing up. Or perhaps they simply started off with less motivation and more psychological problems.
    Despite these uncertainties, researchers have found some evidence that AA works. When two things go together and researchers want to know which one causes the other, they sometimes try to track them over time and see which comes first—assuming that causation moves forward across time, so the cause precedes the effect. After tracking more than two thousand men with drinking problems for two years, a team led by John McKellar of Stanford University concluded that attendance at AA meetings led to fewer future problems with drinking (and not the reverse—they found no evidence that the presence or absence of drinking problems affected attendance at meetings later on). Moreover, the benefits of AA remained even after taking into account the men’s initial level of motivation and psychological problems. Other researchers have likewise concluded that AA is at least more effective than nothing. The failure rate among members is high—it’s normal for them to relapse periodically—but they usually resume abstinence. In fact, AA seems to be at least as effective as professional treatments for alcoholism.
    Project MATCH, a large-scale research project in the 1990s, tested the theory that all treatments work, but not equally well for everyone. Presumably, some people should do better in AA, while others should benefit from professional treatment. Some alcoholics in the project were assigned to take part in AA, while others underwent one of two clinical programs administered by experts: cognitive-behavioral therapy or motivational-enhancement therapy. Some alcoholics were randomly assigned, while others were matched to the treatment type that was deemed best for them. Several years and millions of dollars later, it turned out that all the treatments were about equally effective, and that there was very little benefit from trying to match people to the optimal treatment. (In fact, it wasn’t even clear that any of the treatments were better than nothing, because the project didn’t include a control group receiving no treatment, so there was no way to tell if the people would have done just as well on their own.)
    All in all, then, AA seems to be at least as good as, if not better than, professional treatments costing much more. Even if researchers haven’t figured out exactly what it does, we can point to some familiar ways in which AA appears to help. We know that self-control starts with setting standards or goals, and we can see that AA helps people set a clear and attainable goal: Do not have a drink today. (AA’s mantra is “One day at a time . ”) Self-control depends on monitoring, and AA offers help there, too. Members get chips for remaining sober for certain numbers of consecutive days, and when they get up to speak, they often start by saying how many days they have been sober. Members also choose a sponsor, with whom they are supposed to remain in regular, even daily, contact—and that, too, is a powerful boost for monitoring.
    There are also a couple of other explanations for the correlation between attending AA meetings and drinking less. The less-inspiring explanation is “warehousing,” to borrow a term used by some skeptical sociologists to explain what high school does. They see school as a kind of warehouse that stores kids during the day, keeping them out of trouble, so that its benefits come less from what happens in the classroom than from what doesn’t happen elsewhere. By a similar logic, evenings spent attending AA meetings are spent not drinking. We think it unlikely that warehousing accounts for the entire benefit of AA, or even the majority, but it undoubtedly contributes something.
    The other, more uplifting, explanation is that the meetings offer social support. Like everyone else, alcoholics and drug addicts are capable of remarkable feats of self-control in order to gain social acceptance. In fact, that desire for peer approval is often what got them in