by Janice Fanning Madden
This past summer, the Committee on Science, Engineering, and Public Policy (COSEPUP), a joint committee of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine, published their study on graduate education entitled Reshaping the Graduate Education of Scientists and Engineers. The 200-plus page report discusses graduate education in the physical, life, and social sciences and in engineering. COSEPUP makes three broad recommendations for doctoral education: (1) doctoral programs should be broader or less specialized; (2) doctoral programs should provide better information and guidance about career opportunities; and (3) a national discussion should take place about graduate level human resource policies. While the second and third recommendations are defined so that no reasonable person could disagree, the first is not sufficiently justified by the study. And, to make matters worse, COSEPUP suggests that government should encourage such broadening by funding graduate students through training grants rather than through research grants.*
Ph.D. education at universities cannot be considered separately from research at those institutions. Successful research programs require successful graduate programs. There are very few research institutions in the world that are not also teaching institutions or universities. The reason that research requires graduate training is that the senior investigator requires highly skilled assistants who are trained in the specifics of the research being conducted. For research that is on the frontier of knowledge, the investigator conducting the research is the only person who can provide assistants with the necessary training. These assistants provide intellectual stimulation to the investigator, as well as perform many of the tasks necessary to the project. For these reasons, the training of graduate students is a complement to research. To the extent that graduate training (i.e., the training of research assistants) is necessary to research productivity, it is a by-product of research and uses no additional resources. In that sense, Ph.D. education is "free." The intrinsic complementarity between research and graduate education is best demonstrated by the fact that most research institutions are universities. Simply, basic research without research doctorate programs is more costly or inferior to that produced in the research university.
There are inevitable consequences of this link between doctoral education and research. Doctoral training, as a complement to basic research, is designed to produce researchers. This means that the training provided is specific, not general. As the research is basic and non-proprietary, it is funded by public dollars. There is a mandate to fund such research, and the resulting training, from public moneys because of its nature as a public good. Students pursuing professional study of law, business, etc., are not so subsidized. They are pursuing education that generates productivity that is easily captured by a single entrepreneur and is, therefore, proprietary. This type of training results in profits for private enterprise that private enterprise pays for through higher salaries to these graduates that, in turn, make them willing to incur the necessary tuition.
Due to a leveling, or even cutback, in the extent of the nation's investment in basic research, there have been recent disruptions in the career paths of scientists. The cutbacks appear to have resulted in fewer permanent positions in basic research. Scientists trained for basic research are now seeking employment elsewhere. Private industry employs more of these individuals than in the past. Not surprisingly, private industry demands some different skills from those required by basic research. The question that must be answered, then, is should these changes in career paths and the different skill requirements in private industry affect the way doctoral education is structured? Specifically, should doctoral education be made more general, as advocated by the COSEPUP report, and should the federal government attempt to influence such changes?
The answer to these questions depend on the answers to three sets of more factual questions about the current situation:
What is the long-run outlook for basic research? Will there be long-run cutbacks in employment that merit plans for a permanent decrease in the flow of research doctorates into basic research?
What are the relative costs and benefits of expending resources to provide more general forms of training as part of the research doctorate? What is the value to industry of the more general skills? What is the value to basic research productivity of the more specific skills? What is the cost to the productivity of the basic research enterprise of devoting resources to the production of more general skills?
If more general skills are desired among research doctorates taking jobs in industry, what is the efficient locus for such training? Are research doctorate programs the best providers of this training? or other programs in the university, such as professional masters programs? or private industry itself?
These questions are critical to deciding how to proceed, but are not even contemplated, much less answered, by the COSEPUP study. If there is only a short run mismatch between the supply of, and demand for, persons with basic research skills, then there is no reason to make permanent adjustments in the educational system. The COSEPUP report, in fact, argues that it is not clear that there is any permanent mismatch between supply of, and demand for, persons with research doctorates. If the different skills required by industry either cost too much to produce within research programs, or are more efficiently provided by on-the-job training in industry, then there is also no reason to adjust the educational system. COSEPUP never addresses this issue.
Even if there is reason to change the nature of graduate education, is there sufficient reason for the government to encourage such changes? The COSEPUP report provides numerous descriptions of innovation currently underway in research doctorate programs. None of these changes are the result of traineeships or other government intervention. They arose because the institutions themselves, in their own attempts to attract the best students and to train them well, saw these changes as beneficial. They occurred within a system that closely tied government funding for research and for graduate education, but that allowed the institutions to compete for the best research assistants, in terms of the quality of the instructional programs. As a result, the nation obtained high quality research and the training of research doctorates in the most efficient manner.
Whatever the ultimate decision about the advisability of investing in traineeship funding for graduate students, there has been no case made for the suggestion that such traineeships be financed by reallocations with the National Science Foundation and other funding agencies, presumably at the expense of research funding. Research funding is provided because basic research is essential to the economic, social, and technological expansion of the country and, therefore, the long run growth in the standard of living for Americans. Cuts cannot be made in basic research without decreasing this growth. Up to now, we have been able to produce research doctorates as a "free good" in the production of research. If that is to change because the research doctorate is to change, then the funding for the production of doctorates should come from those who benefit from the change. There is absolutely no reason to look to basic research for this funding.
* At pages 4-5 of the Executive Summary of Reshaping the Graduate Education of Scientists and Engineers:
Most federal support for graduate students is currently provided through research assistantships. Research assistantships are included as parts of grants that are competitively awarded...to support...research.... We recommend an increased emphasis on education/training grants...we recognize that a heightened emphasis on education/training grants could reduce the funds available for research assistantships.
Dr. Madden is Professor of Regional Science, Sociology and Real Estate, and Vice Provost for Graduate Education