This report looks at the problems in trying to assess the outcome of self help groups. It concludes that there are difficulties in trying to do research as the researcher’s external frame of reference may be at variance with the ideology of the group. Using 2 self help models, the study investigates what, whom and when to measure. (MG)