A message from Tom Glennan, Principal Investigator for the Project on Program-Centered Research Planning in Education.

Late in 1999, the Office of Educational Research and Improvement asked RAND to undertake an effort to build a planning base for new programs of research and development in reading and mathematics education. This Web Site is a part of the effort we have begun to meet that request.

There is today, a new enthusiasm for basing educational practice and policy on sound research. The perceived success of the long-term program of research on early reading supported by the National Institute of Child Health and Human Development, and the synthesis of that and other work on beginning reading carried out by the National Research Council, may help explain this renewed interest in research-based guidance. Perhaps the work of Bob Slavin and his colleagues in Success for All, which demonstrates that a comprehensive school program design rooted in research can help schools serving disadvantaged students to improve their students performance, has also provided greater confidence that policy and practice can be guided by research findings.

If these success stories contribute to a new appetite for education research, we should worry that we have so few of them. In preparing for the effort reported on this web site, RAND reviewed literature and talked with many research leaders, practitioners, and policy makers. In those discussions, most viewed education research as being both low in quality and of limited practical use. Dueling reports evaluating state achievement standards and new curricula fuel a longstanding perception that much of education research is calculated simply to reinforce the philosophical or ideological biases of the people carrying it out. Change in education was seen as a recurring cycle of fad replacing fad rather than an evidence-based process of cumulative improvement. In addition, education research was seen as roving all over the map, taking up exotic topics and using language that made it inaccessible or irrelevant to teachers. While clearly these views did not characterize all research on education, they seem widely shared.

We also looked at the ways disciplines assure the quality of their research. What we found was that, aside from some very general and not very surprising or distinctive principles (objectivity, openness/replicability, experimental controls, and so forth), research quality tends to be established and regulated by rules that are integral to the culture of the particular discipline. Apparently, these rules are learned by participants in the course of their training and socialization into that discipline -- they are matters almost of taste and style that become an indistinguishable aspect of being a member of that discipline. Since education is better understood as a profession rather than a discipline, it is perhaps not surprising that there is greater difficulty in agreeing about what stands as quality work.

We concluded that education research could do well enough at the level of the individual research project if it would simply follow the general tenants of the profession's related or constituent social science disciplines. A much larger problem, we felt, has to do with standards for programs of research. Quality might be substantially enhanced not so much at the level of individual research projects as at the program level where issues such as rigor could be addressed in the context of the specific research goals. And by attending consciously to questions of significance, coherence, and knowledge accumulation, a quality program would be one in which the agency succeeded in organizing a sequence or series of research efforts over time that would build on each other, contributing along the way, to a deeper understanding of important problems.

OERI found this notion of program standards very appealing, for it provided not only a way to transcend the discipline versus field problem but offered specific clues for attacking the quality issue at its core. We agreed that a strong program of research should be concerned with an important problem over a sustained period of time. The needs of potential users should be explicitly taken into account as the work is designed. Questions of evidence and method should be debate as a matter of program design as well. When properly managed, knowledge on the problem(s) in question should accumulate and unproductive lines of inquiry would be eliminated.

In order both to better understand what would be involved in creating such programs and to help OERI push forward with its planning, RAND proposed developing a model of what programmatic, problem-solving research might look like in two of the most important problem areas facing American education. Specifically, we organized two study groups in the areas that are of paramount importance to ensuring the educational success of all students -- literacy, or reading comprehension, and fundamental competence in mathematics. We have begun to explore what will be involved in developing a strategic plan for research, and the organization of research support, which would over ten to fifteen years, produce the knowledge necessary to support teachers and schools in succeeding in bringing most of their students to reasonable levels of competence in these two fundamental areas.

The study groups are now in place and we have been able to attract some very good people to join in producing a first approximation of what such a plan might look like. As you also will see on this site, the study groups have agreed that a crucial element in ensuring the success of such a plan is finding a way to engage and develop a community of scholars and practitioners committed to the goal of developing a cumulative attack on these fundamental problems of knowledge and practice. That, among other things, is where this website comes in, and that is why I hope you will engage with us in making this process work, by giving us constructive feedback and by proposing to join in doing the work.

Tom Glennan