This study uses nationally representative data from the National Education Longitudinal Study of 1988 to show that information about postsecondary educational institutions (PEIs), guidance, and essential preparatory actions taken by secondary students influence whether an individual will attend a PEI within 2 years of high school graduation, and, if so, what type of PEI he or she will attend. Multinomial logistic regression is used to model PEI enrollment as a function of critical explanatory variables, controlling on an array of background and contextual characteristics including socioeconomic status, race and ethnicity, and gender. The conceptual framework is embedded in research on talent loss, which is described as the occurrence of promising students not reaching their full educational potential. Increased levels of information, guidance, and critical actions taken were positively and significantly associated with initial enrollment in a 4-year PEI, as opposed to enrollment in a 2-year PEI (full-time or part-time), or no enrollment. Further, aspects of information, guidance, and action affect the contrast between each of the 2-year enrollment categories and the never- enrolled category. These results remain significant after controlling for background and contextual characteristics. The effects on talent loss and implications for policy and practice are discussed. Three appendixes discuss the construction of variables, the estimated parameters of the logit model used, and the calculation of predicted probability profiles. (Contains 14 tables, 2 appendix tables, and 35 references.) (Author/SLD)