The sample plan for this study was a single-stage, stratified, proportionate cluster scheme. The resulting sample was one of randomly selected clusters (in this case undergraduate classes) and stratified across class level (the selected classes represent a progression from freshman to senior representation).
Although it may seem that freshman students would be overrepresented because they made up two sampling quadrants of the frame (Figure 2), the percentage of selected freshmen was held as proportional to the population percentage. The freshmen participants were volunteers from two randomly drawn courses rather than one. The proportions were matched as closely to the university percentages as possible. While a perfect match was not possible, this method allowed greater representativeness for freshmen, who alternate in taking the two courses in question, English 101 and History I between semesters. That is, freshmen who take English 101 one semester take History I the following semester, and vice versa.
Because of the nature of this study, the sample consisted exclusively of males. The focus of this work is on male sexual aggression as a function of personality dimensions and socialization