This article is devoted to the study of the possibilities of using methods of demographic analysis in order to do research on the educational mobility of college students, on the basis of administrative statistics. The standards that are demanded of present-day scientific analysis determine the necessity of bringing data of a new type into play, by means of which the situation can be described with a high degree of detail. That potential is afforded by information gathered in institutions of higher learning and is being used for preparing consolidated reports for a limited range of indicators and searching for individual data. Converting administrative data into statistical data creates fundamentally new possibilities, already used in statistics. In particular, prospects are opened up for the study of the educational mobility of college students, by means of methods that are not commonly used in that field. Methods of cohort analysis make it possible to arrive at a multidimensional assessment of the process of determining the influence of external and internal factors on its dynamics (including the conditions under which a group is formed and structured). In a number of countries, the necessity of cohort analysis of educational mobility became apparent in connection with the reform of higher education. The rise in the number of institutions of higher learning, and the increasing strain imposed by problems of financing, made it essential to address the rational use of available resources. It is proposed that the rate of student attrition from higher educational institutions be taken into account in calculating the average costs of educating a student. The use of the cohort method makes it possible to determine the actual duration of schooling, which must be taken into account when calculating the costs and to find approaches to estimating the losses involved in uncompleted education, as well as to determine the categories of students susceptible to the risk of dropping out, for the purpose of putting together programs and measures to help students. This study of students at Moscow State University shows that social background factors influence the likelihood of graduating. Cohort analysis of data that higher educational institutions have collected for administrative purposes could yield valuable insights into the dynamics of educational mobility. [This article was translated by Kim Braithwaite.] (Contains 3 figures, 4 tables and 11 notes.)