Dennis Epple, Richard Romano, and Holger Sieg," Peer Effects, Financial Aid, and Selection of Students into Colleges and Universities: An Empirical Analysis", Journal of Applied Econometrics, Vol. 18, No. 5, 200, pp. 501-526. In our empirical analysis, we use both university-level data and data for a representative sample of students. Our primary data source is the National Postsecondary Student Aid Study (NPSAS) obtained from the National Center for Education Statistics (NCES). The NPSAS contains extensive information for a sample of students. Of particular relevance for our work, the NPSAS contains the student's performance on standardized tests (either SAT or ACT), information about income of the student's family, and information about the financial aid received by the student. We secured from the NCES a restricted-use version of the NPSAS that contains student-level data for 1995-96 and links each student in the sample to the school the student attended in academic year 1995-96. Since this is a restricted-use database, we are unable to archive the data. However, other researchers may request access to these data from the NCES. In addition to data for individual students, we use data for colleges and universities. Peterson's conducts a survey of all colleges and universities, obtaining information on faculty resources, financial aid, the distribution of standardized test scores, and a host of other variables. We have purchased their database. Since this database is also proprietery, we are unable to archive these data. Researchers interested in accessing these data may contact the Thomson Corporation. We have supplemented the above data sources with information on educational expenditures and endowments from the NSFWeb accessible Computer-Aided Science Policy Analysis and Research (WebCASPAR) database.