Parents, courts, and legislatures have been struggling to define equal educational opportunity (minimum achievement level for all? minimum growth in achievement? differential growths in achievements?). At the same time, economists, sociologists, and educators have been struggling to identify which package of school inputs is required for each type of student to equip him or her for educational growth. Most empirical attempts to identify which inputs matter have concluded that schools barely make a difference. From this conclusion has flowed a prevailing nihilism with respect to schools as an egalitarian force. We conclude, on the basis of a microeconometric examination of Philadelphia School District data, 1) that many school inputs do matter, 2) that disadvantaged students can be helped by particular types of inputs, and 3) that the use of pupil-specific data, and statistical methods appropriate to such data, account for the cheerier results of this study. Little theory, economic or otherwise, is currently available to describe the determinants of educational achievement. Casual observation, combined with the education literature, suggests that achievement (A ) is a function of a student's hard-to-disentangle genetic endowment and socioeconomic status (GSES), teacher quality (TQ), non-teacher school quality (SQ), and peer group characteristics (PG). Thus,