by Erik Banks

In our last newsletter Frank Riessman and I examined some if the inevitable methodological difficulties of self-help research. Among these are the fact that groups studied by researchers have to conform to the rigors of the controlled experiment. Members have to be randomly assigned to a control group or to the self-help group. Now although studies of this kind have shown very positive effects, critics say that throwing individuals randomly into a group--whether they wanted the intervention or not--alters the self-determined, self-selected nature of groups and falsifies the phenomenon of natural mutual aid. But if only self-selected groups count, how are they to be studied for effectiveness? Without random assignment there is no way to tell whether members would have benefited the same amount on their own.

One would have thought that the motivation to participate in a group and the group's effectiveness were inseparable. But this is far from clear. According to Hazelden's data, the effectiveness of non-voluntary court ordered participation in groups for alcoholism is surprisingly comparable to self-selected groups. And in the article "Who Talks" in the American Psychologist (THIS NEWSLETTER), the authors reveal that the motivation to participate in mutual aid does not necessarily go hand in hand with effectiveness. For example groups for migrane headache do not attract members despite their effectiveness, while less effective groups may attract more members. This suggests that motivation for participation and effectiveness may actually come apart in some groups for some conditions. It is an interesting result, if verified, because it would mean that self-selection is not necessarily a vital variable for effective mutual aid functioning, and this would remove one of the strongest objections to the effectiveness studies already on the books. Each case will have to be considered carefully to separate the variables, but it is worth doing since controlled studies continue to represent one of the most important elements in convincing policymakers to implement mutual aid referrals, and it certainly forms a cornerstone of any appeal to managed care organizations.