A model for evaluating effectiveness in preschool and school social work programs is proposed. The paradigm has two major dimensions: criteria for evaluating success, and modalities for assessing whether the criteria have been met. Three criteria are suggested including improved views of significant others, and seven modalities including test scores, observational data, and self-reports. The conditions under which each criterion and modality is optimal are discussed and examples offered of instruments which can be used for each modality. The difference between effort and effectiveness evaluation is highlighted along with the procedures which can be followed in conducting each type of evaluation. Among the other issues explored are the use of the interrupted time series design, unanticipated consequences, the need for both short-term and long-term objectives, and the advantages of contrast groups over control groups. Suggestions are offered to individual social workers for modifying this model to assess the effectiveness of their own practice. Appendices include criteria for judging program effectiveness, tests of parent attitudes, the manual for the Toy Sorting Task, a bibliography of self concept tests, and tests involving locus of control, interpersonal perception, and teacher behavior. (Author/RD)