Evaluation is a process of systematically collecting, analyzing, and using data to aid program planning through a feedback mechanism. It involves the application of social science methodology to programs. The basic questions asked are what was intended, what actually happened, how the intended relates to that which took place, and why there was similarity or disparity between the intents and events. The last question is in the realm of evaluative research, involving scientific techniques of controlling variables, and may be beyond the capability of program staff. Exploratory evaluation is done by external researchers, while internal evaluation, involving the collection of descriptive information, status checks, and comparison of program effectiveness against standards, can be performed by program staff. Outcome evaluation compares program goals to results, while process evaluation assesses program activities as a means to the stated goals. An evaluation design requires a feedback loop mechanism that leads from evaluation back to planning. Data collection should reflect realistic information needs and should involve colaboration between evaluators and decisionmakers. Information collected should be timely, and the evaluation design should provide for broad-based participation of those affected. It is essential that evaluation be planned at the time of program planning and be viewed as an intrinsic component of program implementation. Participant discussion concerns responses to and interpretation of evaluation results. For discussion of action planning and management, see NCJ 83284.