The prevalence of invalid reports of cigarette smoking initiation (reporting smoking at one point and denying it at another) was studied, with 2,597 seventh- and eighth-grade students questioned. About 4% of the sample provided inconsistent responses, a rate that justifies concern about self-report accuracy. (SLD)

An example is provided of an evaluation in which the formulation of explicit decision rules was a central activity. The case for and against the use of decision rules in program evaluation is considered, and the appropriate context identified. Implications for the rights of decision makers are addressed. (Author/CTM)