A distinction is made between program evaluation and project evaluation. The former is used to determine the effectiveness of a particular program or to compare programs. The latter is used in the initial stages of program development to define and monitor progress; to identify problems at an early stage; to record what is done and why; and to provide decision-making data. Test selection criteria and procedures for establishing external and internal validity are defined. Program evaluation consists of these steps: (1) defining objectives; (2) identifying information to document attainment of objectives; (3) selecting or developing instruments to collect this information; (4) establishing success criteria for these instruments; and (5) developing a management system for the administration and analysis of evaluation forms. Project evaluation consists of four steps, each having its own evaluation form. Steps include specifying tasks; monitoring progress toward completion; documenting completion; and evaluating completed tasks. The project evaluation approach follows Provus' Discrepancy Evaluation Model. A role distinction is made between project evaluator as manager of activities and the program evaluator as impartial planning consultant. (CP)