Federal Underinvestment in Education Research

The federal government has a bigger role to play in understanding how the "knowledge industry" of higher education works. Move over Big Science, it's time for the era of Big Education.

By Ralph W. Kuncl

We are a knowledge-based society. But the knowledge business has a problem. It doesn't know enough. When General Electric or Microsoft has a problem, it spends several percent of its revenues—perhaps billions of dollars—on research and development. It does so despite enormous demands on the resources that drive its profits. Historically, we as a society have pumped about 5 to 10 percent of total federal expenditures for defense and health into research and development—in spite of serious pressures to fund the delivery of national defense and health care. It is a great irony that education, a field that values new knowledge so much, lags severely behind in the proportion of outlays devoted to research.

Why invest in education research? Put simply, we have vast areas of ignorance in education. A small sampling of research questions applicable to all educational settings includes (a) How do students learn best? By experience? By drill? In play or sport? (b) What's the optimal class size (especially in K-12 education)? (c) How does ethnic diversity enhance learning, and what evidence shows that it does? (d) How do we achieve fewer dropouts? (e) What techniques work best for retention of knowledge? (f) How can basic cognitive science be translated into the classroom?

Some of these questions are gaining new attention. In 2000, the National Academy of Sciences and the National Research Council published How People Learn: Brain, Mind, Experience, and School to "explore the critical issue of how better to link the findings of research on the science of learning to actual practice in the classroom." In 2003-04, the National Science Foundation (NSF) launched a program to establish some fifteen to twenty-five national "science of learning centers." The program aims to fund basic scientific research into how people learn in order to help inform educational practices and policy. The centers might explore, for example, robotics, artificial intelligence, cognitive and behavioral neuroscience, and the biological basis of learning. These trends represent the basic research side of research and development.

The translational side must necessarily follow. Smart research programs such as the NSF's science of learning centers recognize that as long as basic researchers have no ties to the applied researchers and practitioners in their field, their research will be effectively meaningless. So the NSF will fund outreach to users and assessment and dissemination of the research its centers produce.

Like the study of learning, research on the educational efficacy of a racially and ethnically diverse environment has also advanced over the past several years. The American Educational Research Association's Panel on Racial Dynamics in Higher Education published its findings in 2003 in Compelling Interest: Examining the Evidence on Racial Dynamics in Colleges and Universities. Similarly, the American Council on Education's Office of Minorities in Higher Education continues to publish research annually in special reports. These efforts are interesting starts, but they are not nearly enough.

In higher education, an array of research issues face us, such as (a) How can college students best be encouraged to graduate in a reasonable time? (b) How do student enrollment patterns (full time, part time) affect the likelihood of completion of degree, subsequent degree attainment, and outcomes in the labor market? (c) What role does financial aid play in determining who goes to which institution and what happens to them? (d) How do different postsecondary curricula affect initial degree attainment, subsequent degree attainment, and labor market and citizenship outcomes? (e) In the competitive, stratified environment of U.S. higher education, how does wealth inequality influence the ability of institutions to fulfill their missions? For example, are some institutions forced to focus more on market demands than on their core missions? (f) What role, if any, does high-stakes, criterion-referenced school testing play in linking K-12 and postsecondary education? How, if at all, would the role of testing change college aspirations in a K-16 model? (g) What role does institutional or departmental size play in producing nurturing cultures for degree attainment? (h) What are the determinants of learning? That is, what kinds of college and university cultures seem to produce the most efficient attainment of critical knowledge (baccalaureate as well as graduate)? This question is related to that of diversity but goes far beyond it, to organizations of residential life and graduate student employment patterns (getting a teaching fellowship at Harvard differs from teaching courses in four different boroughs in New York City while at the City University of New York Graduate Center).

One could go on and on. But these questions have to do, first, with the neuroscience of how we learn and, second, with whether and how institutions reproduce the social structure, even as they promise untold opportunities. There are many other research questions that have to do with access (about, for example, merit-based versus need-based financial aid and ways to achieve socioeconomic diversity most effectively) and organizational or administrative arrangements (the efficacy of decentralized management or responsibility-centered budgeting, for example) that relate less directly to learning.

Barriers to Investment

The need for research in education is huge, but the structural barriers are formidable. Comparing the United States and Europe is instructive. Many more of our young people go to college than do Europe's. At the same time, many European universities offer free, or nearly free, tuition. The European system of higher education is much more centralized and homogeneous than that of the United States. Because ours is nothing of the sort, it is that much more difficult to study. Of course, that means we should be devoting more, not fewer, resources to understanding its nature, dynamics, and outcomes.

From the federal perspective, there are fifty different state systems of higher education (many of which themselves encompass university, state college, and community college systems), plus the private sector, which includes independent colleges and universities and for-profit entities. It is difficult even to design a study of a particular phenomenon in higher education, when its settings differ so greatly.

Methodological and historical barriers to educational research often make generalizability of results difficult. Further, educational research, unlike science, does not have a rich culture of controlled experiments. It has instead a culture of "best practices," which are inherently anecdotal. Often, poorly funded empirical reports and experiential follow-ups by educators describe the "effects" of a latest "innovation," forgetting the basic lesson that association is not causation. Education in this country has seen one fad after another, each with a five- to ten-year lifetime. Rarely do we critically examine the evidence that validates one or another new practice. This soft approach to educational research has led to redundancy and inspired cynicism and existential fatigue among education journalists, policy makers, and the public. In the process, education has become a low-status field within the academy. Witness the decaying number of large, top-notch colleges of education at our universities.

In his 2001 National Academy of Sciences presidential address, Bruce Alberts challenged us to "make a science out of education." In October 2002, Congress did the same when it passed the Education Sciences Reform Act, replacing the Department of Education's Office of Educational Research and Improvement with an autonomous and, one hopes, more rigorous Institute of Education Sciences. The act was a small step in the right direction. But "big education" (large groups of interdisciplinary teams studying major problems) has not emerged in educational research in the way that "big science" brought us high-energy physics and the Human Genome Project, or in the way that the "military-industrial complex" delivered technology-rich peace keeping, smart weapons, and a downsized modern army. That is the central point of this article: we have not invested in educational research and development in the way our society has committed to health and defense.

Federal Budget

It is perhaps a truism that no one person understands the federal budget. It is a daunting document to discern. Another truism, however, is that, no matter what we profess to believe, what we really believe in is our budgets.

To begin, one needs to look at federal outlays for selected agencies as a percentage of total U.S. outlays. Outlays are the amounts issued through checks and cash payments during a given period, regardless of when the funds were appropriated or obligated.¹ 

As Figure 1 shows, between 1965 and 2004, the proportion of total federal expenditures allocated to defense and health changed radically, while outlays for education remained relatively flat at about 2 to 2.5 percent of the total federal budget. Most federal expenditures on education go to the Pell Grant program, federal family education loans, and federal direct student loans. These programs are all necessary. But in the current debate in Congress over reauthorization of the Higher Education Act, discussion of financial aid virtually drowns out line items for research. If we were forward looking, small expenditures for research might have the power to revolutionize the delivery and experience of education and ultimately reduce the need for aid and loans. If efficiency in graduating students in a timely manner can be improved, for example, overall costs of higher education may decrease.

Figure 2 reveals, however, that in the thirty-seven years since research data have been available, the percentage of the total federal budget assigned to research and development has remained fairly steady at 10 to 12 percent for defense and 4 to 8 percent for health care (although the percentage was briefly higher in the 1960s). Education has brought up the rear for decades, devoting less than 1 percent of its budget to research.

In the 1960s and 1970s, however, the federal government considered education research to be such a critical part of society's effort to fight inequality that it spent 2 percent of its education outlays on research. The Reagan administration cut such efforts, and the spending curves have been flat since then. Although education was raised to cabinet status in 1979, our commitment to education research has hovered at a half percent throughout several successive administrations and congresses.

This flat progress occurred during a period that saw economic cycles of inflation, stagflation, Reaganomics, the waxing and waning of federal deficits, and the technology boom; the global political impact of the Vietnam War and the end of the Cold War; the staggering sociological significance of the "graying" of the population and the growth of Medicare; the revolution in biomedicine as symbolized by the delineation of the human genome and AIDS research (which occurred despite escalating health care costs); and the elevation of education to federal departmental status and a top presidential campaign issue. Both education and health care have evolved as rights, yet education remains research poor.

Business Practice

How does higher education compare with Fortune 500 companies regarding the percentage of total outlays devoted to research and development? Of course, the amount that firms spend depends on the nature of the industry; the share expended on research varies, from zero to as much as 15 percent of total revenues. Perusal of the 2002 financial reports of the fifteen most profitable companies in the United States is instructive. As table 1 shows, Merck, the tenth most profitable U.S. company in 2002, spent $2.677 billion, or 5.2 percent of its revenues, on research and development. Even so, it was outdone by Pfizer, Johnson and Johnson, Microsoft, and IBM.

What determines such investment behavior? One might consider risk taking. Pharmaceutical firms take huge risks with research and development (spending up to 16 percent of their revenues) because of potential enormous gains, not only for themselves (profit), but also for society (disease management, better quality of life). But investing in research and development is not merely a phenomenon of the pharmaceutical industry. Table 1 shows significant expenditures, in the multi-billion-dollar range, by technology giants IBM, General Electric, and Microsoft. In both kinds of industries, the concept of a research pipeline directly connected to future productivity is salient.

By contrast, Exxon Mobil, the fourth largest company in the world and the fourth most profitable U.S. firm in 2002, spent only 0.3 percent of its revenues on research and development: $621 million (excluding exploration costs) on revenues of over $204 billion. This proportion is identical to the amount devoted to research and development by another global oil company, the Royal Dutch Shell Group. In such companies, the salient task is to churn out more of the same old product, ever more refined. Financial corporations spend so little on research that an exact amount does not appear as a line item in consolidated financial statements.

By these comparisons, the federal budget for education behaves more like the product-oriented oil business than the discovery-oriented technology and pharmaceutical firms. But higher education is not in the extraction business; its goal is not to dig students out of high schools and efficiently refine them into laborers. A growth- and future-oriented oil company would see its future in solar, wind, and hydrogen enterprises and understand that it is really in the energy business. If education seeks to be more discovery and risk oriented, it needs to see its calling as creating opportunities for lifelong learning, producing engaged citizens, helping learners and workers make multiple transitions between universities and the workplace, and developing intellectual breadth.

Clear Conclusion

The conclusion the data suggest is woefully obvious: we undersupport research on education. It is shortchanged compared not only with business, but also with public spending on health and defense. Education is not, however, so different from health and defense. All three enterprises involve massive amounts of services and huge personnel costs and occupy an essential, high place in the panoply of human needs in a democratic society. Education is a part of what we mean when we revere "the blessings of liberty."

In her well-regarded 2000 book, An Elusive Science: The Troubling History of Education Research, Ellen Condliffe Lagemann, dean of Harvard University's Graduate School of Education, tells us we may not be any more able now to "make science out of education" than was philosopher Josiah Royce in the inaugural issue of Educational Review, published in 1891. There, he editorialized his unwillingness to "apply so pretentious and comforting a name as 'Science' to any exposition of the laborious and problematic art of the educator."

The jaded among us will claim that throwing more money at educational research is wasteful, because its quality is so low. It is probably true that qualitative methods, heuristic approaches, and phenomenological research are more common in education than in the physical and biological sciences.

Rigor can, however, come simultaneously with new funding commitments. There is no reason national expert review panels assembled to evaluate grant applications in educational research cannot raise standards at the same time that the federal allocation to such research rises. That happened with science funding at the National Science Foundation, the National Institutes of Health, and the National Institute of Mental Health, and it helped fuel the biomedical era in which we live.

Research methodologies are moldable. Interpretive methods like participant observation, interviews, discourse analysis, and archival analyses are weak tools. Educational research surveys can be a cut above that, provided that sampling techniques ensure in-ternal reliability, external validity, and relative freedom from bias, and provided that the statistical errors of estimates are clear. Direct comparisons of experimental interventions as independent variables will raise the standards, as will use of control groups and randomized trials to help to overcome the problem of selection bias. In that new future of American educational research, historical controls or no control group at all would rarely be acceptable.

James Duderstadt, former president of the University of Michigan, mused in the February 4, 2000, issue of the Chronicle of Higher Education that the twenty-first-century university that becomes a virtual laboratory for educational research "would propagate a risk-tolerant culture, in which students and faculty would be strongly encouraged to 'go for it,' and in which failure [in research] would be accepted as part of the learning process, associated with ambitious goals rather than poor performance."

The spirit behind some Bush administration initiatives and the creation by Congress in 2002 of the Institute of Education Sciences suggest that it is now education's time. The institute's goal of pooling research data and disseminating them to educators, administrators, policy makers, and the public is valid. The institute itself, however, constitutes a mere reorganization of feebly funded structures. Reorganizations of this kind are not a new trend. The politics of recurrent education mandates with weak funding is traceable from the establishment by Congress of the Office of Education in the Civil Rights Act of 1964, through the creation of the National Institute of Education in 1972, the formation of a cabinet-level U.S. Department of Education in 1979, and the reorganization of the National Institute of Education into the Office of Educational Research and Improvement in 1985. This history appears rather lame when viewed from the perspective of the past forty years of actual federal funding. To reprise, no matter what we profess to believe, what we really believe is in our bud-gets. The need now is simple. We need to fund rigorous research.

What is the payoff? What might we actually get if we increased the educational research and development budget? Fewer high school dropouts? More college graduates? A more efficient workforce? A more engaged citizenry? Perhaps all of the above. In addition, many unintended positive consequences of increased education funding will probably emerge, just as they did in defense and health. In health, for example, the emergence of primitive magnetic resonance imaging seemed truly remote in 1973. But the benefits arising from its eventual development have been incalculable. Applications of the technology range from gated cardiac imaging and brain tumor detection to routinely imaged rotator cuffs and knee cartilages. Because of the myriad possible payoffs of investing in research, we should take the gamble and increase the budget for education research.

Challenges

We've passed through the era of Big Science, and it's time to move into the era of Big Education. The portion of federal dollars for educational research should rise initially from the current 0.5 percent of the total federal outlay for education to at least 2 percent. Such an increase would be revolutionary. An eventual goal of 5 percent can be envisioned.

It's time to believe in the huge effect research can have on social ills. Continuing our exclusive focus on serving the ills (paying for health care, paying for peace keeping and war making, paying for educational services) will not succeed in the end. The message of that approach is that "the poor are always with us," "war is inevitable," and "ignorance is bliss." Research will lead to cures for our social ills. Just as important, every dollar spent on specific research is secondarily amplified in the general economy.

The public's confidence in education, now that higher education is becoming a universally accessible commodity, will be increased by valid and reliable research as opposed to our current reliance on cyclically faddish models and anecdotal evidence. Through research, we can become more efficient transmitters of knowledge.

As historian Stanley Katz has pointed out, philanthropic foundations targeted university research a hundred years ago and became the dominant funders of it precisely because they found they could not directly solve the root causes of the social ills they sought to conquer: ignorance, poverty, and illness. They chose to amplify their effectiveness by investing in research.

Unfortunately, many foundations are now moving away from higher education, with a certain ennui that the world does not need one more academic conference or published proceedings. Individual donors are now often more interested in bricks and mortar. Governments are inherently more interested in providing services. In an era when the total federal budget outlay for education and educational research is so small, one cannot hope to cure ignorance until one targets research on learning. At the same time that we increase our investment in research, we must be accountable by assessing the outcome of educational interventions and measuring the effects of that new research spending.

Today, the need to inculcate mentored research throughout undergraduate curricula and the desire to make inquiry-based learning the norm are high on the agenda of groups seeking to reform higher education (for example, the Reinvention Center, a consortium of research universities seeking to re-envision undergraduate education). Inquiry means asking questions. Research means gathering evidence to test hypotheses. There will be no conclusions to draw from cyclical faddish movements in education, no matter how impassioned or innovative or logical they seem, unless we can really believe what we see—and unless we professors profess through our budgets.

Nota Bene

To calculate total federal outlays for selected agencies between 1965 and 2004, it was necessary to clarify certain terms and data with staff at the Office and Management and Budget (OMB) responsible for preparing relevant portions of the budgets. Total U.S. outlays were derived from the portion of the Budget of the United States Government for fiscal 2004 titled "Summary of Receipts, Outlays, and Surpluses or Deficits 1789-2008." Total health outlays were calculated from "Outlays by Super-function and Function 1940-2008," which incorporates Medicare starting in 1970. "Outlays by Agency" was used to calculate the total outlays for education and defense, the latter including military and civil defense budgets. All figures were updated in July 2003, using revisions estimated in February 2003 by the OMB. The figures for fiscal 2002 are actual and finite, whereas those for fiscal 2003 and fiscal 2004 are estimates. The estimates were done prior to the war in Iraq, taking into consideration its possibility rather than its actual costs. More recent figures from July 2003 (called "Mid-Session Budget") were not used because doing so would have led to inconsistencies in comparisons with prior years.

To ensure a consistent reckoning of numbers, the various departments housed within the Department of Health and Human Services throughout the years were uniformly tracked and included. From 1967 to 1978, health was included under Health, Education, and Welfare. For the purposes of this article, and for those years, health figures for research and development included the following entities: Alcohol, Drug Abuse, and Mental Health Administration; the Centers for Disease Control and Prevention; the Environmental Health Service; the Food and Drug Administration; the Health Care Financing Administration; the Health Resources Administration; the Health Services Administration; the Health Services and Mental Health Administration; the National Institutes of Health; the Office of the Assistant Secretary for Health; the Office of Human Development Services; the Offices of International Health; and Social and Rehabilitation Services.

Starting in 1979, the Department of Health, Education, and Welfare was renamed the Department of Health and Human Services. From 1979 to 1985, for the purposes of this article, health figures for research and development were included in the following agencies' budgets: the Agency for Healthcare Research and Quality; the Alcohol, Drug Abuse, and Mental Health Administration; the Centers for Disease Control and Prevention; the Centers for Medicare and Medicaid Services; the Food and Drug Administration; the Health Resources and Services Administration; the National Institutes of Health; and the Office of the Assistant Secretary for Health.

From 1996 to 2003, for the purposes of this article, health figures for research and development were included in the following agencies' budgets: the Agency for Healthcare Research and Quality; the Agency for Toxic Substance Disease and Registry; the Centers for Disease Control and Prevention; the Centers for Medicare and Medicaid Services; the Food and Drug Administration; the Health Resources and Services Administration; the National Institutes of Health; the Office of the Secretary; and the Substance Abuse and Mental Health Services Administration.

Note

*The figures can be found in the print edition. Back to text.

1. One obvious complication was the change in names of federal agencies over time. Between 1967 and 1978, for example, education was part of the Department of Health, Education, and Welfare (HEW). Starting in 1979, however, it became the Department of Education, when HEW was renamed the Department of Health and Human Services (DHHS). Unlike education and health, defense figures were easily tracked from Office of Management and Budget line items during the period of study. Delineation of the components of research and development within health and education, however, required strict comparability of agencies with changing names over time. Specific agency name changes were as follows. The Agency for Healthcare Research and Quality: in fiscal 1991, it became separate from the Office of the Assistant Secretary for Health, having been previously named the Agency for Healthcare Policy and Research. The Agency for Toxic Substances and Disease: before fiscal 2000, data reported for the Environmental Protection Agency included research and development data for the Agency for Toxic Substances and Disease Registry. Alcohol, Drug Abuse, and Mental Health Administration: because of the reorganization of the DHHS, this agency ceased to exist as of October 1992, and its research and development functions are reported with the National Institutes of Health (NIH) data beginning fiscal 2000. National Institutes of Health: owing to the reorganization of the DHHS, the Alcohol, Drug Abuse, and Mental Health Administration ceased to exist as of October 1992, and its research and development functions are reported with the NIH data beginning fiscal 2000, at which time the NIH classified all of its development activities as research. Office of the Assistant Secretary of Health: in fiscal 1991, the Agency for Health Care Policy and Research became separate from the Office of the Assistant Secretary for Health. As of fiscal 1995, the Office of the Assistant Secretary for Health merged with the Office of the Secretary. The merger created a new role for the assistant secretary, who became head of the Office of Public Health and Science, a new division with the Office of the Secretary. Back to text.

Ralph Kuncl is provost of Bryn Mawr College. He wishes to acknowledge the assistance of David Karen, Sona Armenian, and Nancy Collins.

 Printer-Friendly Page