![[Table of Contents]](/icons/up.gif){kind=icon}
II. Summary of the Proceedings
As it is with conferences such as this, the comments and recommendations cannot be readily documented, validated or even necessarily agreed to by all participants. They are nevertheless the considered views of well-informed, often expert, leaders and participants in educational research; in many cases, they are mutually reinforcing, almost consensus statements. They should be considered authoritative but not final; weighty but still needing further deliberation, evidence and argument.
This report is neither a stenographic nor even a chronological recapitulation of the statements made in the day and a half of conversations. It is, rather, one participant's authorized rendering of the principal emphases and tentative agreements of the sessions, paraphrasing cogent interventions, making connections and drawing conclusions that were implicit in the discussion but could only be discerned in retrospect, with the stenographic record and group session reports in hand.
Both at the conference and in this account, most comments were and are meant to reflect upon the history and performance of the educational research system, albeit from the vantage point of one or another institution or experience. Thus, the comments are not about any particular enterprise within the system.
The clear sense of the conference was that the educational research system has had powerful constraints and limitations on it, which have hindered, sometimes crippled, numberless researchers and projects. The wonder is that educational researchers have been able to accomplish what they have. The prospect is that more confidence and opportunity could pay off handsomely, if the support is strategically provided. The past century's record buttresses this claim. Educational research as been used time and again, at critical junctures, to improve teaching and learning. Important examples range from John Dewey through constructivism, to Edward L. Thorndike through behaviorism and educational testing, to the diverse social scientists who influenced the design and evaluation of Great Society education programs. In our time we have seen the impact the students of cognition and organization have made upon pedagogy, assessment, and other school reform strategies. At the same time, there have been many hard lessons to be learned. Educational improvement occurs slowly and in small increments no matter how powerful the research base behind it. Deep disagreements among prominent researchers are continual and perhaps inevitable. Professional educators have never become enthusiastic consumers of research, and weak designs and measures, combined with these doubts and disputes, have produced too many research results whose values and political implications are more prominent than their scientific validity. There are, moreover, important contemporary proofs of the capacity of educational research to make an important difference in education. Recent important examples include:
The instance of reading research seemed to the conferees particularly pertinent. Here is a field transformed by a high-powered, highly-focused, speedily accomplished and clearly communicated research planning effort undertaken by federal research managers in the early 1970s, and then developed over 25 years in federally funded reading research centers. It was spurred on by the International Reading Association's A Nation of Readers in the 1980s, and nurtured by new discoveries in cognitive and developmental studies. Over the past decade, it was revisited by the research program at the National Institute of Child Health and Human Development (NICHD), implemented in Success for All and other successful early reading programs; and synthesized and applied to policy, practice, and the next generation of research, in the NRC's recent report.
The current moment of opportunity and risk for educational research has distinctive new features. The quest for measurable, accountable results will be a permanent policy feature, at every level, sought by policymakers, parents, and the public. And this is not a trend confined to education. Many realms of policy are becoming increasingly results-oriented. The Government Performance and Results Act (GPRA) at the federal level, the growing demand for international comparative data, and the broad insistence upon assessment and accountability at all levels of education reflect a broad and persistent public policy shift. Education may have some unique characteristics that must be taken into account, but it is not and will not be immune from this profound trend. Educational research will be rewarded if it contributes (and is seen to contribute) new knowledge and insight in this arena, and punished if it does not. It will take particular courage and some diplomatic skill to succeed: to support and sustain appropriate developments while at the same time standing, with solid research knowledge, against wrong headed, damaging proposals. And the tolerances will be slight. The pace of social and educational change is accelerating and the stakes are growing. Insights ranging from cognitive science (in the case of pedagogy) to political and economic theory (in the case of vouchers) raise questions about the basic structure of schooling. Policy paradigms are moving away from centralized control mechanisms to reliance on incentives and capacity building, as the significance of education grows in the worldwide economy.
Even so, there are still too few resources available for educational research to do all that it could to improve education. There are strong logical arguments for more resources, ranging from the link to economic development to the inability so far to design more effective approaches for poor and minority students who need them most and whom our society and economy need as well. There are also arguments by analogy from other fields, like health and science, where sustained research expenditures, both public and private, have paid off in better lives for all.
At the same time, there are hidden consequences of long-term patterns of scarce funding: the compromises researchers make in their designs to accommodate limited resources; and the inability of research managers to amass resources for sustained, necessarily multidisciplinary, attack on important, slow-yielding problems. One of two equally unsatisfactory outcomes usually ensues: too few resources get spread too thinly and for too short a time on too many topics; or, by contrast, very large investments are made in a few highly-visible, large-scale, overly-ambitious, uncoordinated programs of research and evaluation which fail to find significant effects or to attribute them to causes. The net result is to weaken the repute of programs, investigators and, by association, educational research itself. Rarely is there the opportunity for persistent, cumulative pursuit of questions from theoretical work and hypothesis-building, to small-scale experiments or other controlled trials, to demonstration designs, to large-scale field trials using randomized assignment or other methods of evaluation.
Inasmuch as progress in children's learning is inherently a long run proposition, this situation will take a long time to fix, even if more resources become available. The 1997 report of the President's Council of Advisors on Science and Technology (PCAST) makes the most recent and persuasive case for the swift and steady build up of resources for educational research to $1.5 billion per year ultimately. PCAST suggests that the necessary investments must encompass methodological enhancements based on new technological and statistical techniques, the production of a large new generation of researchers, and the strengthening of the institutions that will carry forward the enterprise.
There are, of course, numerous obstacles to progress. The most prominent is that educational research has yet to find a way to relate effectively with the teachers and other educators or policymakers whom it serves. It is perceived by these partners to run the narrow gamut from nonconsultation to tokenism at every stage, from agenda setting to research design and execution to interpretation and implementation of findings. To those in the field, educational researchers seem not to be united around the theme of student learning, but seem rather to be working within discrete disciplinary boundaries and not communicating across them or with the field.
Viewed from the opposite perspective, the deepest problem may be that many schools are not learning organizations with a commitment to continuous improvement; and most teachers have not thought of themselves lifelong learners and knowledge-based, reflective practitioners. This is the demand side of things: education is not a research-hungry system.
To some extent, these disparate perspectives are inevitable. Researchers expect to solve problems on a generalizable basis over extended periods of time, while teachers and school people are naturally preoccupied with the here-and-now of classroom instruction and individual learning; and the incentives and rewards researchers and teachers face differ accordingly.
These perceptions of disparity are changing, but slowly. Most teachers have not been trained or encouraged to be constructive learners from research. With heavy workloads and scant time or opportunity to make a serious effort to absorb and incorporate research findings into their daily routines, they settle for second hand reports or, alternatively, existing lore and practice. Thus we have a vicious cycle education has not consistently valued research knowledge nor fostered the culture and institutions to strengthen it. As a result, it often cannot get the knowledge it now seeks, and so it continues to undervalue research. Researchers, for their part, neglect to shape their inquiries to respond directly to the problems facing teachers, devalue lore when they should honor and learn from it, do not train teachers to be participants in and consumers of inquiry, and take their norms and rewards from academic disciplines rather than the profession they serve.
A similar tale can be told about the relationship of educational research to policymakers and administrators. Most educational policymakers are simply not used to making research-based decisions to the extent occurring in other realms where the evidence is from "hard science." When they do make such decisions, as in the cases of, say, class size or the adoption of specific reading programs or the construction of entire programs of school reform, they may either ignore research findings or go beyond what the shaky research evidence can bear. Administrators face similar dilemmas, as they decide whether research results should guide decisions or if administrative convenience or simplicity should carry the day.
The reportedly shocking lack of sustained research on the learning problems of poor and minority youngsters seemed to be a case in point, especially considering that major federal programs of long standing, like Title I and Title VII, have been promoted as knowledge-based and frequently evaluated. The opportunity presented by the Obey-Porter program for comprehensive research-based school designs in Title I might well be missed. The problems seemed to be twofold: policy and evaluation research that proceeded ahead of adequate understandings of how children learn and thus produced little sustained program improvement; and the inability of researchers, policymakers and educators to create together a program of studies that would, over time, yield the basic knowledge and instructional designs that might improve learning. As often as not, neither practitioners nor policymakers are able to judge the significance of any particular research enterprise or result. Educational research on the one hand has not established clear, widely acknowledged standards for the design and conduct of research. On the other, it countenances work that is value-driven and neither significant nor rigorous. Well-meaning researchers sometimes pull their punches when their findings are discouraging, not wanting to damage programs or the children they serve with results that they know are partial at best. Dubious findings and implications can be peddled, while high-quality work is not adequately appreciated or used.
Two recent initiatives will help alter and diminish these long-standing patterns of miscommunication and mistrust:
[III. Implications for Educational Research Planning]