The study design aimed to examine the stability of perceptions over time and to separate the experiential effect of past behaviors on current perceptions from the deterrent effect of current perceptions on future behaviors. The subjects were freshmen at a major university during the fall quarter of the 1974-75 school year. About 90 percent were white and about half were of each sex. The interviews took place between January and June 1975 and again about 1 year later. The questions covered the subjects' past involvement in specific illegal acts, involvement during the past year, and perceptions of the risk of arrest. The questions on the risk of arrest focused on both the risk for the general population of lawbreakers and the individual's personal risk. The perceptions of the risk of sanction were not extremely stable over time. In addition, more evidence existed of experiential effects of past behaviors on current behaviors than of deterrent effects of current perceptions on future behaviors. A third wave of panel data would allow the estimation of models with fewer statistical assumptions and would permit further examination of the processual logic of deterrence and experience. Figures, data tables, notes, and a list of 23 references are provided.