The central purpose of this study was to identify factors that increase the likelihood that high school seniors will plan to attend a four-year college, paying particular attention to variables that are associated with college attendance by low-income students. Logistic regression was applied to data from three sources: telephone interviews with 900 seniors who took the Scholastic Assessment Test (SAT) I, a survey completed by these students when they registered for the SAT, and their combined SAT scores. The findings of the study reinforced those in much of the literature. As many studies have found, socioeconomic status, family support, educational goals, academic self-concept, financial aid, and academic achievement were all significant predictors of plans to attend college for the full sample and, for the most part, for the 300 low-income students in the sample. This study found two previously untested variables to be particularly important predictors: the number of years students took college preparatory courses had a significant positive effect on the probability that they planned to attend a four-year college or university; and the beliefs of guidance counselors about students' post-graduation plans had a strong influence on low-income student outcomes. The interview protocol and survey instrument are appended. (Contains 44 references.) (MAB)