Reforming Graduate Education in the Sciences
By Congressman George E. Brown, Jr.
For a number of years, I have been advocating a two-track agenda for the science and engineering community. I have repeatedly urged scientists and engineers to become more involved in the political and policy process and I have stressed the need for a reassessment of our national research and development (R&D) enterprise in light of the major changes that it faces. The first element of this agenda is well underway amidst numerous signs of a heightened political presence by the science and engineering community.
Unfortunately, progress on the second agenda item is not as visible. And nowhere is this reassessment needed more than in the area of academic research and graduate education.
The need for a reexamination of our National R&D efforts has been well discussed. With the end of the Cold War, the clear and simple justifications for much of our R&D efforts, found in our competition with the Soviets and the need to contain global Communism, have disappeared. No clear mission has emerged to take the place of that previous, easily communicated goal. Pressures to balance the budget have eroded funding for R&D programs, clearly displayed in the recent budget agreement that outlines about a 14 percent decline in civilian R&D over the next five years.
All of these changes will force us to reassess our past policies and goals for academic R&D and graduate education, where such policies and goals exist at all. This new funding reality means that we can no longer accommodate the limitless aspirations of every institution and discipline in the science, engineering, and academic community. The stress of this change is already becoming apparent and is producing a higher education system that is financially strapped while the public is treated to news stories of excessive tuition increases. And the worst is due to come in the next decade when the children of the "Baby Boomers" begin to enter college in large numbers and further complicate the situation.
Our science and engineering (S&E) graduate education efforts are already a mess and will suffer further disruption unless we openly address the problems we face. The direct and simple motivation for our national science and technology efforts during the Cold War obviated the need to develop a sophisticated set of goals and justifications for the linkage between academic research and graduate education. What passes for policy in this area is really a set of simple assumptions that we have accepted as truths for forty years, assumptions that are being challenged by the complexities we face today.
For nearly twenty years, we have linked academic R&D grants and graduate education, assuming that some portion of our R&D funding will go to support graduate students. But we never clearly spelled out our higher education goals separate from the performance of R&D. This unthinking linkage of R&D to graduate education means that the number of Ph.Ds produced reflects the availability of academic R&D funding, rather than having a relationship to a set of national goals for S&E graduate education. And, we do an inadequate job of monitoring conditions in the graduate education system and do not conduct sophisticated national analyses of the ongoing situation with S&E advanced education. The predictable result of this haphazard system is a series of surprises such as the current "overproduction" of S&E Ph.Ds.
When federal R&D spending rose through the 1980s, this prompted a rise in the numbers of graduate students. But the full nature of this relationship is not understood and there is no government source able to document how much of the $12 billion in federal R&D funding for academic research (1995 figures) went to support graduate student education. Without this information it is impossible to tell if our current levels of S&E graduate education support are too high, too low, or about right. And, there is no way of knowing what we will do to S&E graduate education as we cut federal R&D programs in order to balance the budget.
But being able to measure levels of support tells us little without some idea of our goals for S&E education. We have seen some discussion of these goals in recent studies citing the need to diversify and broaden graduate education to prepare S&E graduates for jobs outside of academia. These discussions reveal the consequences of our past simple linkage of graduate education to academic R&D: the creation of a system of training and rewards that is myopically focused on academic R&D as a career choice. It is no wonder that this system resulted in an "overproduction" of Ph.Ds seeking academic appointments as federal R&D increased and then fell. But this revelation is a very superficial one and begs further discussion of national S&E education goals.
This revelation also prompts an examination of the role of academic institutions in contributing to our current graduate education problems, for if there is an "overproduction" of Ph.Ds one is lead naturally to an examination of "overcapacity" within the academic R&D system. While there has been growing discussion of the narrow career focus of S&E graduate students there has been little discussion of the reward system of academic institutions that is just as narrow.
With the federal government's steady funding of R&D through the 1980s, we rewarded higher education institutions that were able to capture some of that R&D funding. This encouraged schools to work their way up the Carnegie Classification of Academic Institutions ranking system "ladder," to Research II status (receiving between $15.5 million and $40 million annually in federal support) and then to the top category of Research I status (at least $40 million in annual federal support and awarding at least fifty doctoral degrees). Receiving more R&D money meant expanding the size and prestige of the institution and allowed more graduate students to be supported off of those grants. The results are predictable.
Between 1988 and 1993, the number of Carnegie I research universities jumped 30 percent, from 68 to 88. Given that these schools graduate 65 percent of the science and engineering Ph.Ds, the structural capacity of the system to produce Ph.Ds may have expanded as well. But while this fact gets passing mention in the NSF 1996 Science and Engineering Indicators, the reasons for and details on this expansion are not fully understood or analyzed. But it seems evident that the federal government's linkage of R&D and graduate education sent clear signals to higher education institutions that they should move into research and expand their research and graduate education capacity.
We do not understand how this dynamic plays out in individual disciplines, because neither the federal government nor the National Academy of Sciences collects and analyzes relevant data by discipline or profession on an ongoing basis. For that information you have to go to the various scientific societies, such as the American Institute of Physics, which does have good statistics and can provide analysis. But this ad hoc process does not meet our current need for a more sophisticated review of graduate education. Again, we are so used to the old, simple paradigm that evolved during times of plenty that we are flying blind into a time of limited resources.
But in an odd and ironic twist, it may be that the availability of resources is to blame for some of the problem we face. In a study submitted for publication by two researchers at the University of California, San Francisco, there emerges what seems to be a direct relationship between increases in extramural funding at the National Institutes of Health and numbers of doctorates awarded in the biological sciences. Other trends include an flattening grant size, lower success rates for extramural R&D applications, advancing aging at the point of first grant and first position, and lengthening time spent in postdoctoral positions. In short, it may be that high funding for biomedical R&D, combined with our simplistic model for R&D and graduate education from the Cold War, is creating the same stress in the biological sciences that we have seen in physics, chemistry, and mathematics. If true, this means that restoring funding for federal R&D will not only not fix the problems in graduate education, but may make them worse. If true, this data indicates that broad S&E graduate education reform is needed before we can discuss levels of funding.
But this reform requires the active involvement of the higher education community, a group that has not provided much public information and analysis on these issues. This reform will also require the candid participation of the scientific and engineering professional societies and focused discussion, such as we see in this publication. And we will need open and painful discussions by leading scientific institutions, involving not only the esteemed Ph.Ds from the 1960s and before, but some people who have recently minted doctorates and are living through the current situation.
The time is long past when we can placate taxpayers and parents with reassuring anecdotes about physics Ph.Ds who were lucky enough to get jobs modeling stock prices on Wall Street. Serendipity is an inadequate public policy, even in times of plenty. There is a need for more sophisticated answers than, "Trust in market forces to rectify the situation." The federal government, institutions of higher education, directors of research labs, and scientific societies have all passively conspired to create the current market forces that are producing a dysfunctional situation. We must all work together to change the market forces.
George E. Brown, Jr., is serving his 17th term as Representative of the 42nd district of California. He is the ranking Democrat on the House Science Committee.
©1995 - 2017, AMERICAN PHYSICAL SOCIETY
APS encourages the redistribution of the materials included in this newspaper provided that attribution to the source is noted and the materials are not truncated or changed.