Some of the major problems encountered in retrospectively evaluating the effectiveness of ongoing community intervention programs, and some approaches to their solution, are illustrated through a detailed description of the methods employed to assess the performance of the Newark Childhood Lead Screening and Control Program between 1970 and 1976. A process analysis, along with a limited product analysis, provided some basis for judging the effectiveness of the intervention program, despite the absence of an a priori research design, the changing characteristics of the population screened, and the limitations to the measurements used to assess the impact of the program. Even through rigorous scientific control is often unattainable when evaluating community programs, some level of critical assessment of programs is needed to determine whether or not they merit continuing public support.