The attendance rates of welfare recipients at self‐help group meetings was compared when supplementary reinforcement was given for attending and when it was not. Reinforcers included donated items, such as clothing and household goods, and information about welfare services. It was found that the attendance at meetings averaged three recipients per meeting without reinforcement and 15 recipients per meeting with reinforcement. There was evidence that attendance was associated with participation in other self‐help activities. It was concluded that practical forms of reinforcement can be found to maintain the participation of severely disadvantaged populations in self‐help g