Mr. Chairman and Members of the Committee:

Thank you for the opportunity to testify before you today regarding the reauthorization of research functions within the Department of Education.

The shared understanding of the Congress and the Administration about the role of research in educational reform was evidenced vividly in the recent reauthorization of the Elementary and Secondary Education Act of 1965 (ESEA). In that bill, passed by overwhelming majorities in both chambers and signed into law by the President on January 8, the phrase scientifically based research appears 110 times.

Scientifically based research will also be a component of reform in the upcoming reauthorization of the Individuals with Disabilities Education Act (IDEA). The ESEA and the IDEA account for approximately $30 billion in annual federal expenditures within the Department of Education. Mr. Chairman and members of the Committee, $30 billion is a lot of money. We all recognize that, historically, the huge annual investment in the education of disadvantaged students and students with disabilities has not achieved everything that was expected of it. For instance, in the most recent National Assessment of Educational Progress in reading, 40 percent of white 4th graders read at a proficient level, compared with only 12 percent of African-American students. In some urban school districts that serve predominantly disadvantaged children, 70 percent of 4th graders cannot read at even the basic level. Nothing has changed in the last decade in these statistics, and the overall gap between the highest and lowest performing students has actually increased in some subjects.

If scientifically based research is going to be the key to reform of our most important federal education programs, then we had better make sure that the federal office with the principal responsibility for generating that research has the tools it needs to get the job done. That is what we are here today to address.

In facing that task, I want this Committee to understand that we are dealing not only with gaps in student achievement, but also gaps in scientific knowledge. Consider some of the major program areas in the ESEA in which Congress instructed that funding decisions and practice should adhere to scientifically based research. These include the core academic subjects of reading, mathematics, and science, schoolwide reform models, early literacy programs in preschools, professional development of teachers, supplementary educational services, education of gifted and talented students, character education, educational technology, and programs for safe and drug-free schools, among others.

We have a substantial and persuasive research base in only one of these topics, learning how to read. However, even within reading, the research becomes substantially thinner when we move down the developmental range from learning to read in early elementary school to getting ready to read in the preschool period, and up the developmental range from learning to read in elementary school to reading to learn, otherwise known as reading comprehension, at later points in schooling. In the other core academic subjects of math and science, research has not progressed to a level at which it is possible to make strong statements about which approaches produce the strongest effects on academic achievement for which children in which circumstances. In the education and professional development of teachers, we don't have research to answer dozens of fundamental policy issues about how to best train and sustain teachers in order to enhance student learning. The ESEA authorizes supplementary educational services, such as after-school tutoring, for children in failing schools. Which tutoring programs work best for which types of academic skill deficits? Sorry, we don't know. How about comprehensive school reform? The ESEA instructs local educational agencies to consider successful external models and to develop an approach to reform of their school that is derived from scientifically based research. By one count, there are well over 100 comprehensive school reform models from which a local educational agency might choose. Which of these are successful? That is hard to say, because only a few have been subjected to research, and much of that research isn't sufficiently rigorous to permit strong conclusions about the effects of the models compared to business as usual, much less compared to each other.

My point, and I apologize for making it repetitiously, is that there is a lot we don't know about how learners learn and how to deliver instruction effectively.

The extent of our ignorance is masked by a "folk wisdom" of education based on the experience of human beings over the millennia in passing information and skills from one generation to the next. This folk wisdom employs unsystematic techniques. It doesn't demand scientific knowledge of mechanisms of learning or organizational principles or social processes. It is inefficient, and it is hit or miss. It lets us muddle through when the tasks to be learned are simple, or in a highly elitist system in which we only expect those with the most talent and most cultural support to learn advanced skills. But it fails when the tasks to be learned are complex or when we expect that no child will be left behind. The tasks to be learned in a 21st century economy are without a doubt complex, and we have rightly decided that our education system must serve all learners well. We have to do better than we have done in the past.

Consider the analogy of medicine. For thousands of years, folk remedies have been used to cure disease or relieve symptoms. But the successes of modern medicine have emerged in the last 75 years and derive from advances in the sciences of physiology and biochemistry that allowed us to understand the mechanisms of disease, and from the wide use of randomized clinical trials to determine which prevention and treatment approaches drawn from these sciences work as intended.

Or consider the analogy of agriculture. For thousands of years, humans barely managed to avoid starvation by using agricultural methods that were passed from generation to generation. The abundance of inexpensive and nutritious foods that can be found at any neighborhood grocery store today results from agricultural practice that has moved from reliance on folk wisdom to reliance on science.

When we come to education, the picture is different. The National Research Council has concluded that "the world of education, unlike defense, health care, or industrial production, does not rest on a strong research base. In no other field are personal experience and ideology so frequently relied on to make policy choices, and in no other field is the research base so inadequate and little used." At the same time, the National Research Council has concluded that scientific inquiry in education is at its core the same as in all other fields. In other words, the core principles of scientific inquiry are as relevant for education as they are for medicine. There is every reason to believe that, if we invest in the education sciences and develop mechanisms to encourage evidence-based practice, we will see progress and transformation in education of the same order of magnitude as we have seen in medicine and agriculture. I believe we are at the dawn of exactly that process, and it is very exciting.

How quickly will the transformation of education into an evidence-based field occur? The actions of this Committee and the Congress as it considers the reauthorization of the research functions in the Department of Education will have a lot to do with the answer to that question.

A number of significant changes are necessary so that we can operate consistently with the standards of a science-based research agency and so that the research, evaluation, and statistical activities we fund lead to solving problems and answering questions of high relevance to education policy.

Before assuming my current position, I spent 31 years conducting research on children's learning. I am proud to say that some of that research has proven useful to educators and parents. For the last 15 months, I have focused exclusively on OERI, first as a consultant to the Department, and since July of last year as Assistant Secretary for Research and Improvement. My testimony today is informed both by my background as a practicing scientist and by my experiences to date in leading OERI.

I believe that we have made substantial progress in OERI over the last year. To be specific, we have launched three major new cross-cutting research initiatives--in reading comprehension, preschool curriculum, and learning in the classroom; we have hired a number of key personnel; we have brought the responsibility for the evaluation of the impact of federal education programs into OERI and have designed a new generation of evaluations that will use scientifically rigorous randomized trials to provide definitive evidence of what works and what doesn't; we have helped the Department move towards a greater reliance on evidence in its delivery of programs; we have implemented new procedures for peer review of applications for research funding that are modeled on those used at the National Institutes of Health (NIH) and that are working very well to help us select only the very best proposals for funding; we have established a What Works Clearinghouse, which will vet the evidence from research on education and make it available to decision-makers in easily understood forms; and we have put forward to the Congress as part of the President's FY '03 budget request an unprecedented and badly needed 44 percent increase in funding for our research functions and a 12 percent increase in funding for our statistics functions. I believe we have also created a positive buzz in the research community about the new OERI that helps us attract strong applications and that enhances the participation of distinguished scientists in our planning and review processes.

If you are willing to take my description of these successes at face value, you might be tempted to draw the conclusion that the current OERI statute doesn't need fixing. Why not report out of this committee a bill that is pretty much the same as current law?

Let me tell you why not: A lot of what we have accomplished in the last year has been much more difficult than it should have been because of the current statute. Further, I have been operating with a remarkable degree of support from within the Department, the Administration, and Congress, and from many non-federal organizations that are eager to see the federal education research agency revitalized. Appreciative as I am for that support, it is natural for enthusiasms to wax and wane. Further, I'm quite concerned that the alternative to progress will be backsliding and entropy rather than the status quo. We need an authorizing statute under which the Department's research agency can develop and sustain a cumulative research program, and we need it this year.

Here are some major problems in the current statute that should be corrected in reauthorization.

Organizational structure

Institutes. OERI is currently divided into four principal operational arms: 1) the National Center for Education Statistics, which conducts surveys and assessments to determine the condition of education; 2) the Office of Reform Assistance and Dissemination, which monitors ten regional educational laboratories and administers a large number of programs funded under the ESEA; 3) the National Library of Education, which manages a physical library in the Department of Education as well as an electronic repository of documents in education called the Educational Resources Information Clearinghouse; and 4) the National Research Institutes, which are five administrative units that manage research centers at universities and field-initiated grants to individual researchers.

This administrative structure is seriously problematic. The five national research institutes have overlapping responsibilities, redundant personnel functions, and statutory restrictions on funding that do not permit the agency to pursue a focused agenda or to support significant programs of research. To be specific, the statute requires that an equal amount of the funds appropriated for research be made available to the Achievement and Assessment Institute and the At-Risk Institute; that each of the five national research institutes use at least one-third of its share of the research appropriation to fund university-based research and development center(s) and at least one-fourth to fund field-initiated research (the statute does not permit the agency to specify even broad topic areas for field-initiated research -- individual investigators choose both the topics and methods of study); and that not more than 10 percent of the total research appropriation (and not more than 33 percent of the share for any particular institute) be used to fund crosscutting research. Crosscutting research is research that is germane to more than one institute and may be carried out jointly by two or more institutes, or by one or more institutes jointly with other offices in the Department or other agencies within the Federal Government.

Each of the initiatives we have launched this year is cross-cutting. Take our new program of research in reading comprehension as an example. Should this be the responsibility of the At-Risk Institute or the Achievement and Assessment Institute? And isn't it also an initiative of relevance to the Early Childhood Institute and the Postsecondary and Adult Learning Institute? It is difficult to assemble staff outside the Institute structure to focus on cross-cutting issues. And, most critically, we have only been able to move ahead with our new programs based on bill language in our appropriations statute that exempts the funds for new initiatives from the statutory requirements for apportioning funds under the institute structure. The appropriators have done this since our statute expired in 1999 based on the assumption that these funding strictures would be removed in reauthorization. I hope their assumption was correct, because it would be impossible to do the new work that needs to be done under current law.

Centers. Another facet of this same problem lies in the current requirement that at least one-third of institute funding go to research and development centers located at universities around the nation. Centers are the major mechanism by which OERI supported research prior to my arrival. Currently, there are 11 R & D centers. Several have been funded for over 15 years. Some of the centers have performed well and the center mechanism is one we intend to continue to use. However, centers have failed as the principal mechanism of supporting field-based research. Why? First, an effective center needs to have scientists who work closely together and interact frequently with the goal of solving a particular problem or closely connected set of problems. Too many of our centers end up being mail drops that serve scholars scattered across the nation. Center support is parceled out to these scientists for individual projects that are only loosely connected to each other, if connected at all, and the goal of the work--the point at which success could be declared--is undefined. In effect, such centers become intermediate funding agencies. We give them money, and then they give it to other people under conditions that are much less competitive, much less strategic, and involve much more overhead than would be the case if we skipped the center mechanism entirely and parceled out the money ourselves. Second, centers as the sole mechanism of support freeze out all those researchers who could be doing important work but aren't part of the club. In the recent history of federal funding of education research, if you were not connected with a center you had scant prospects of continuous funding for a serious program of research. We need much more capacity in the education research community than we currently enjoy. To get there, we need to open up our funding process to all interested and competent parties, including those who are not a part of the existing education research community and center structure.

Creating a culture of science

The recent National Research Council report on scientific research in education concluded that building a scientific culture within the Department's research agency is a prerequisite for all else. This is my view as well. It is very important to understand that successful research agencies, such as the NIH, embody a scientific culture because the people in the principal program management roles share the dispositions and training that characterize scientists. It is this shared culture, much more than statutes, rules, and regulations, that supports high-quality research. My experience in trying to increase the number of qualified scientists at OERI highlights the importance of our excepted service authority, which allows us to hire scientists for limited terms outside the regular civil service. OERI has had this authority for its entire existence, as do our sister research agencies. It is critical that it be continued so that we can rotate scientists through the agency for limited terms and under flexible conditions.

Building a scientific culture at the Department's research agency also requires stability in leadership and the shared sense that the organization can pursue its agenda over the long term.

Regulatory burdens

The Department of Education is required, by section 437 of the General Education Provisions Act, to take public comment on priorities for grant competitions before funding announcements can be published. This can add up to 6 months to the time necessary to make grants, and can push our grant-making under into the final half to third of each fiscal year. The provision makes sense in Education for programs that deliver funding to State and local educational agencies where broad areas of the public may have an immediate stake in the funding program and be motivated to comment. However, our funding announcements are technical documents directed to scientists. The period for public comment required under rulemaking has historically generated very little in the way of comment. Letting us use the same exemption that the NSF and the NIH use would have no downside that I have identified, and would allow us to be much quicker on our feet in getting funding out the door.

Another regulatory burden is the possible application of the Federal Advisory Committee Act (FACA). Our Office of General Counsel has informally advised that OERI may be subject to FACA for the purposes of peer review if a panel that has fixed membership, meets regularly and advises me or the Secretary. Such a panel would have to be chartered as a federal advisory committee. Because applications for funding for scientific projects often include proprietary and privileged information and because FACA requires open committee meetings, we do not want to charter our peer review panels as federal advisory committees. As a result, historically OERI has not had standing peer review committees. Further, when peer review panels meet just once they cannot provide a summary judgment on the quality of applications. Standing review panels are an important tool in the competitive funding process in a science agency. It is critical that we be exempted from FACA for peer review committees.

Budget flexibility

We need more flexibility in authorization and appropriations. When there are separate authorizations for particular, narrow components of our work, there are two predictable consequences. The first is that we are not able to move quickly into a new area of activity that is important. The second is increased pressure to fund work of lower than desirable quality so that we do not have to lapse funds. It would be very helpful if legislation gave us the flexibility to direct funds among program areas in response to project quality and national needs.

Nonpartisanship

The research activities within the Department have sometimes been seen by the outside community and Congress as more subject to political involvement than would be the case for research conducted by the NIH or the NSF. Regardless of the accuracy of that view, the perception that politics is driving research needs to be avoided if we expect the Department's research activities to have the force of scientific findings.

There are a number of ways that legislation could increase the perception and reality of nonpartisanship of the research process. A consolidated budget would help because it would isolate the agency's budget for personnel and supplies from the core Department budget for those items. An agency staffed predominantly by scientists, who are committed by virtue of their training to the integrity of the research process, will contribute significantly to the goals of nonpartisanship and objectivity. Placing the responsibility for evaluation of federal education programs in a center for evaluation within the agency will provide useful distance between the program evaluation and program management functions within the Department.

Research, statistics, and evaluation activities need to be based on sound science and be independent of undue partisan influence. We look forward to working with the Committee towards legislation that supports that goal.

Separation of research and program delivery functions.

Our current statute, as well as previous administrative decisions, have led OERI to be responsible for a large number of non-research programs. These include a number of ESEA programs such as character education, a large number of earmarks through the Fund for the Improvement of Education, technology programs such as Star Schools, and the regional education labs. We believe it is critically important to separate the research agency from the responsibility of delivering educational programs and technical assistance. Over the years, those activities have been assigned to OERI in increasing numbers to the point that over two-thirds of our budget is devoted to non-research programs. The agency responsible for evaluating program effectiveness and upholding high standards of evidence cannot fulfill its role if it is directly delivering the very educational programs and technical assistance that it is supposed to evaluate. Further, the culture of science that is so important to establish within the agency is impeded when we need so many staff to engage in activities such as monitoring FIE earmarks that do not require scientific training. Also, far too much of my and my senior staff's time has to be spent in overseeing these non-research activities. We need a solid intellectual connection between scientific research and technical assistance, but in keeping with the recent National Research Council report on scientific research in education, we believe it is very important to keep these types of activity operationally distinct.

The Regional Educational Laboratories

A very important instance of this general theme has to do with the role and function of the regional labs. I have spent considerable time over the last year getting to know the labs and their work. It is a mixed picture. Some of the labs do work that is considered quite valuable by their customers. Other labs are weaker in the quality and value of the work they conduct. So, one issue with the labs is this substantial variability in quality and relevance. A second issue is defining their core function. Is it applied research and development or is it technical assistance? Applied R & D in any field, including education, means the development of products that are intended to address needs and then doing research on their effectiveness until a final product is developed that successfully addresses the problem it was designed to solve. Some of the labs are actively involved in developing products and programs. Others develop few products. However, even for those labs that do a lot of product development, the research half of the R & D process usually gets short shrift. None of the lab products I have examined has gone through the cycle of development, research-based evaluation, and revision that constitutes the full R & D cycle. Instead, the products are developed and put into the field. Whether they work, or how well they work, is never assessed in a rigorous way. From my perspective, we do not need to use federal funds to sponsor the development and dissemination of unproven educational materials and products. Education is plagued with that from the commercial and non-profit sectors. We don't need to support the expansion of the large evidence-free zone that already exists in education through the regional lab structure.

The labs have a unique and critical role to play in regional technical assistance. The No Child Left Behind Act imposes a new and challenging set of requirements on State and local educational agencies. States and schools need a lot of help in designing and implementing assessment and accountability systems, in training teachers in how to teach reading and math, in selecting curriculum and aligning it with State standards, in recruiting and retaining highly qualified staff, and so on. The Department has been engaged in a Herculean effort to help States and schools understand and implement the new law through a wide variety of meetings, workshops, printed materials, and web sites. However, the Department has few troops on the ground to provide the follow-up and local assistance that educational agencies will need when the unavoidable problems and questions arise. The labs represent a resource for that assistance that could be extremely valuable if focused and aligned with the implementation requirements of NCLB and other federal programs, and if driven by the expressed needs of the State and local educational agencies within a region. The Department's research agency is not the organizational component that should be overseeing regional technical assistance, but it will be important to write legislation that takes advantage of the labs' presence and expertise in each region to provide technical assistance that meets local needs and that structures the labs functions so that the unevenness in the quality and relevance of their work is addressed.

Funding

The entire research and statistics budget of OERI for fiscal year 2002 is less than 1/2 of 1 percent of the Department's discretionary budget. The core research and dissemination budget for 2002, leaving out statistics, is only $122 million. The education research agency needs adequate resources in order to support a sustained and cumulative research effort in its areas of responsibility. I am very pleased that the President is committed to investments in education research. Accordingly, he has proposed a 44 percent increase for fiscal year 2003 in our core research budget and 12 percent in our statistics budget. This is an unprecedented increase. We need the support of Congress in making an appropriation consistent with the President's request so that we can move forward on the important work that needs to be done.

In an effort as large, complex, and important as this, informed, well-intentioned individuals and groups will differ on details. Let us talk about those details and compromise on those that seem to represent different routes to the same goal. However, we cannot and should not compromise on the end points. We need an invigorated agency that is capable of carrying out a coordinated, focused agenda of high-quality research, statistics, and evaluation that is relevant to the educational challenges of the nation, and that has sufficient flexibility to adjust to new opportunities and problems when they arise. This is a unique and unparalleled opportunity to begin a process that will make American education an evidence-based field. If we succeed in this task, historians may look back at our actions in the next weeks and months as building the foundation for a new era in learning and teaching, an era that propelled the United States into another century of preeminence.

Thank you again for this opportunity to testify.

Top