DURING the 1980s, higher education came under what Derek Bok calls a" torrent of criticism." 1 The charges against colleges and universities included insufficient attention to teaching, intellectual conformity in the form of" political correctness," financial abuses connected with federally funded research, conspiracy to fix financial aid offers, and irresponsibly high rates of increase in costs and tuition rates. Although the loudest of these complaints originated outside higher education, those within it shared a growing concern over the problem, and more than a few presidents subjected their institutions to serious introspective criticism. The present study arises from this spirit of self-examination on the part of several university officers as well as others closely associated with research on higher education. It is directed especially to those university administrators and policy analysts who must address in one way or another the issues raised in this book. In an effort to make the analysis as accessible as is reasonable, virtually all equations and some detailed tables are relegated to footnotes and appendices, while leaving considerable graphical and tabular material in the text for the reader to digest.

Among those who conceived of a project on this subject were William Bowen, Martin Feldstein, Jerry Green, and Neil Rudenstine. The project would not have been possible, however, without the cooperation of each of the four sample institutions. I am grateful, therefore, to Jeremy Knowles and the above-named officials at Harvard, to Keith Brodie, Thomas Langford, Malcolm Gillis, and Charles Putman at Duke, to Hugo Sonnenschein at Chicago, and to Stephen Lewis at Carleton for permitting me unfettered use of detailed information on their institutions. Beyond securing permission, I had to call on many administrators at the four institutions to obtain data and to receive guidance on the use and interpretation of those data. The study required the collection of several kinds of information for academic years, most of it spaced at five-year intervals over the period 1976/77 to 1991/92, a time period that typically pushes to the limit most institutions' computerized record keeping. The most complex data are detailed financial records of expenditures; other data include records of class enrollment, faculty teaching, and capital spending projects, some limited to a few departments and some collected on an annual basis. Because of the variety of sources from which these data were gener-