Click for next page ( 53


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 52
FIT ] h eS u D a PRIZE :10H What organizational structure for a SERP enterprise would allow it to successfully carry out the broad mission described in the opening chapter? The key design features we propose emerge from two overarching goals: 1. Developing and steering a research and development (R&D) program that is coherent, high quality, use- inspired, and cumulative and Attracting stable funding and support. After considering key design features that we believe would support those goals, we propose a SERP governance structure and a broad organizational design. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ DEVELOPING A PROGRAM When the committee began its work, we had no common vision of a SERP organization. Rather quickly, however, agree- ment emerged on two central design features we judge to be critical to developing and executing the SERP program: (1) an organization with a strong center capable of steering the re- search and development program and assuring program utility and quality, and (2) dispersed field sites in which collaborative, use-inspired research can be carried out as part of a coherent program. The rationale for each is described below. 52 STRATEGIC EDUCATION RESEARCH PARTNERSHIP

OCR for page 52
S E R P H E A D Q U A R T E R S Tapping existing resources to improve student learning, we argue, will require ongoing assessment of the opportunities presented by existing research, teaching practices, and innova- tions as well as making hard choices about where investment in further research and development is likely to have a high pay- off. Once investments have been undertaken, carrying research and development through all of the stages necessary for utility in practice will require an ongoing assessment of program re- sults and an active steering of the program over time in produc- tive directions. This assessment and steering, if they are to lead to a coherent and cumulative program of research and develop- ment, will require a strong central SERP operation an entity that we shall refer to as SERP headquarters. Requirements for maintaining program quality further sug- gest the need for a strong center. We envision two mechanisms for supporting quality: one is oversight of research designs by a scientific advisory board to assess whether the questions asked can be adequately addressed by a proposed study. Much has been made in recent years about the methodological weakness of education research, and the debate about research methods has even been carried into the halls of Congress. This committee concurs with the National Research Council report on scientific standards in education research (National Research Council, 2002b) in believing that there is no one best methodology for education research. Rather, the method must be matched to the question. Oversight by a scientific advisory board would allow SERP to conduct and support research using a variety of meth- ods, while at the same time addressing the problem of weak research design that has plagued the field. The second venue for quality control we envision is over- sight of the research products through rigorous peer review. While the peers who carry out the reviewing would be outside SERP in order to avoid a potential condict of interest, the orga- nization of peer review would need to be coordinated. Both the scientific advisory board and the coordination of the peer re- view would be the responsibility of SERP headquarters. Finally, we have argued that to create a body of research that accumulates both across research projects and over time, uniform data collection efforts and common research protocols T H E S E R P O R G A N I Z AT 10 N 53

OCR for page 52
will be required. Moreover, since the mission of SERP is to improve student learning, the research program must concern itself from the outset with specifying and measuring educa- tional outcomes and promoting the requisite integration of data across studies. Questions about what works, for whom, and under what circumstances are difficult to answer; the dimen- sions of context that are potentially important are many and are not simply measured. But prospects for progress on these criti- cal issues are improved if the definition of "what works" in- cludes outcomes carefully measured over time, if "for whom" includes a large array of relevant individual and group charac- teristics, and "under what circumstances" includes information on the schools, teachers, administrators, and resources that con- tribute to context. Education research to date has been characterized by a marked absence of the sorts of shared standards, measurement protocols, and other techniques that are necessary for research- ers to replicate and cross-validate findings. Without such re- search protocols and common measurement instruments, we cannot be sure whether positive results redect attributes of a given intervention that can be generalized. Common measure- ment and research protocols are far more standard in other fields, like medicine, where advancing knowledge efficiently has been a priority for some time. Generating an atmosphere of productive collaboration among researchers through require- ments placed on grant funding, as well as establishing and maintaining standards for data collection and for research and measurement protocols, are all critical activities for a SERP head- quarters. FIELD SITES The defining features of the SERP research program are that it is collaborative and use-inspired, being research for, and often on, practice. It will therefore require the participation of teach- ers, administrators, and policy makers, as well as access to classrooms, schools, school districts, and possibly teacher edu- cation programs as "field sites." These are the equivalent of the teaching hospitals in medicine, places of practice that serve as sites for research. While we propose a strong SERP headquar- 54 STRATEGIC EDUCATION RESEARCH PARTNERSHIP

OCR for page 52
ters, the locations envisioned for most of the actual research and development would be widely dispersed. Field sites occupy a central place in the SERP plan. They will provide a test bed for iterative research on learning, on teach- ing, on changes in school organization and culture to support learning, and on the development and use of tools and curricu- lum. In addition, they will be the ground for observation, articu- lation, and modeling of effective practice. They will demon- strate the effects of incorporating research-based programs, pedagogy, and organizational changes to practitioners and policy makers. The field sites might well take many different forms, and, indeed, different types of sites would be suited to different research questions. An entire state may want to join the SERP partnership and itself become a field site for study of state-level policy change. One or more school districts, perhaps in collabo- ration with a local university, might constitute a site. Or a set of school districts dispersed across a state or across the nation might join together to work on common interests that fit within the SERP agenda. While flexibility about the configuration of field sites is im- portant, they must be willing and be judged able to work as a partner in a SERP Network on a mutually agreeable local ver- sion of the SERP agenda. This will require buy-in from both the teaching and the administrative staff. Among the most impor- tant factors are access of SERP research teams to schools, ciass- rooms, teachers, student records, etc.; a sufficiently long-term commitment (and arrange- ment for continuity if change of leadership occurs) to justify the initial investment; released time for teachers to work with research teams, for professional development, and for reflection and interaction with other teachers; adherence to research protocols and sampling re- quirements; adoption of SERP assessments and performance measures; and effective collection of district, school, classroom, and student data to support evaluation. T H E S E R P O R G A N I Z AT 10 N 55

OCR for page 52
Field sites should be committed to promoting the transfer of fruitful findings and applications throughout the school, dis- trict, and wider system. The ultimate aim of SERP is to foster widespread improvement, not just change in one school or 10 SERP field sites. This will require replication of promising inter- ventions in the range of environments in which the research- based change might be applied, as a necessary part of the pro- cess of bringing the most successful innovations to scale. Field sites engaged in the original research can play an important role in the process of supporting the "travel" of effective practices to new locations. This contextual emphasis will make possible the kind of follow-through that has seldom characterized education re- search. For example, understanding how interventions work in classroom settings requires that the research and development be carried out in a pre-determined range of settings and with a variety of students. Understanding the needs of teachers for support in using educational interventions, as well as the char- acteristics of schools that support or undermine change, also requires that schools be available as sites for investigation. Learn- ing more about the personal and environmental characteristics that distinguish excellent teachers from their less effective peers calls for observation and analysis in situ. Are there alternatives to the mode! of a strong central office and dispersed field sites described above? The committee did consider whether a strong SERP center was necessary, given that many of the researchers and practitioners SERP would need to attract are located in dispersed university and school settings. Without a strong center, however, we envision an out- come that would mimic what we see today: a situation in which even the best work is not carried forward, there is little coordi- nation and accumulation of research across sites and centers, the opportunities to coordinate data collection and analysis are routinely missed, and the dissemination of findings is haphaz- ard. Moreover, dispersion of activity across many research loci without a coordinated mechanism for review, assessment of progress, and course correction would make quality control difficult. As a result, the agenda of research centers and indi- vidual researchers might be constrained relatively little by the strong program goals of improving educational outcomes. Similarly, the issues of whether field sites are necessary was considered, with the alternative being one of funding research- 56 STRATEGIC EDUCATION RESEARCH PARTNERSHIP

OCR for page 52
ers who themselves have made arrangements to work in school settings. But the transaction costs of arranging for that work, for both the researcher and the school, are high, and the projects become highly dependent on personal relationships. The com- mittee concluded that to efficiently undertake work on the scale envisioned and to draw many more researchers and practitio- ners into such collaborations, much of these costs would need to be borne at an institutional rather than an individual level. ORGANIZING THE RESEARCH AND DEVELOPMENT PROGRAM How should the program be organized to effectively bring the resources of the research and development community to bear on the problems of practice? Should it be structured around problems as they are manifest in the world of practice, like improving mathematics performance in urban school districts, or problems as they are defined in the world of scientific re- search around sets of issues that share theoretical underpin- nings and research paradigms, like the core conceptual compo- nents of mathematical proficiency? Persuasive arguments can be made for each if the goal is to improve student learning. On one hand, current understanding of mathematics learning will serve as a constraint on the poten- tial of a reform effort. The best intentions for improving math- ematics performance are unlikely to produce impressive results without an understanding of how children learn mathematics and the nature of the problems that result in failure to learn. If the goal is to advance the effectiveness of the education we provide to students, defining a scientific research program that pushes the boundaries of current understandings would seem to be an essential component. At the same time, however, those engaged in practice are only too aware that research on learning will make little differ- ence to schools in the absence of effective organizational sup- ports and public policy incentives for change. Because research- ers by tradition address questions of learning and instruction, organizational change, and policy separately, it is difficult to build on research to improve practice where all are at play simultaneously. We have concluded, therefore, that this is not an either/or proposition. Because the organization of knowI- edge does not map the realities of educational practice, we have T H E S E R P O R G A N I Z AT 10 N 57

OCR for page 52
designed the SERP research program to be responsive both to the strengths of current scholarship and the complexities of the educational system that we are aiming to improve. We provide an argument for one organizational arrangement, but note that others are possible and may be preferred by those who ulti- mately are responsible for building a SERP enterprise. S ER P Ne two rks The first level of organization that we propose is a set of networks broadly defined by fields of research: for example, learning and instruction, schools as organizations, and educa- tion policy (see Chapter 4 for further discussion of networks). While each network will be interdisciplinary (e.g., learning and instruction would bring together cognitive scientists, develop- mental psychologists, education researchers, and discipline ex- perts in reading, science, mathematics, etc.), research traditions and methods are likely to be shared to a much greater extent within these networks than across them. We use the term "network" to refer to a dispersed commu- nit~v of participants whose work in a broad topic area is actively linked through productive collaborations. The networks will promote synergy across researcher-practitioner teams working on common agendas to extend the boundaries of current under- standing in each broadly defined field. For SERP to achieve the kind of programmatic strength that, for example, the National Cancer Institute did in the 1990s, the networks will need a central organizing intelligence to give coherence and direction to the program. The design plan an- chors the planning process for each network in the SERP head- quarters; thus, we expect that the network headquarters staff would take a leading role in conceptual framing, strategic plan- ning, and coordination. In tra m u ra ~ R e s e a rc h We propose that some network research be conducted at the headquarters as well, and that an effort be made to develop a field site in close proximity to the headquarters. Effective steer- ing of a large-scale program will require a primary commitment to SERP, and obtaining that primary commitment from top researchers to an institution that is not conducting intramural 58 STRATEGIC EDUCATION RESEARCH PARTNERSHIP

OCR for page 52
research would be unlikely in our view. The committee consid- ered the National Institutes of Health to be an instructive mode! in this regard. The intramural research program we envision would be a relatively small share of the SERP portfolio per- haps 20 percent of the total. In the early years, however, the intramural program might play a considerably larger role in an effort to create program coherence. The headquarters research group will be well positioned to convene periodic meetings of the teams that make up each network in order discuss the progress of the work and assess the need for revisions, course corrections, or a strengthening of the links between the parts. The challenge of coordination is to maintain a creative tension between coherence and growth of understanding on one hand, and dynamism and innovation on the other. In the interest of the latter goals, we recommend that the scientists and professional experts who staff the headquar- ters research and development program join the enterprise on a fixed-term basis, renewable upon critical outside review. Extramural Research As a network's program begins to take shape, SERP will look increasingly to external teams of researchers and practition- ers to carry out a large proportion of the research activities. Team members might be located in one place or, if not, be linked through a SERP web site, virtual communications, and regular face-to-face meetings. Some teams might be nominated by the network leadership; others would be chosen through a competitive process in response to a network request for pro- posals; selection of yet others might be made in response to field initiated proposals. The creation of extramural research teams when and as needed is intended to lend the SERP enterprise a high degree of flexibilit~v and malleability, while the coherence and continuity required for a strategic program is served by the presence of a core internal research group. Use of external research teams permits SERP to search methodically for and exploit externally generated ideas and a wide array of expertise. It also enhances the organization's responsiveness to the level and kind of effort needed at any given stage of the research program. While some of the research teams are likely to exist only briefly to do limited T H E S E R P O R G A N I Z AT 10 N 59

OCR for page 52
or highly specific tasks, others will have Tong-term involvement in the network. Cross-Network Coorclination While each of the networks would have a separate research agenda, the work of all networks would be brought together at a second, higher level of organization: the research and devel- opment department or unit. It is at this level that advances in knowledge from each of the component networks would be integrated for purposes of improving education. Department management would include a department director and the lead- ership of each of the networks. One of the functions served by that group would be the active coordination of strands of re- search that are carried out within a network or that cross net- works. To illustrate, we can imagine a scenario in which a learning and instruction network has been conducting a strand of re- search on the acquisition of reading skills in language minority children. Simultaneously, the network on schools as organiza- tions has been working with a school district that is undergoing a major reform perhaps like the decentralization reform in Chicago. The school district decides to introduce a research- based intervention for language minority children and wants to study the effects. Two existing strands of research in two differ- ent networks are then very relevant to the study, and the combi- nation of expertise from these two networks will strengthen the research design and study of this new strand. Strands of research that cross networks would, in our judg- ment, represent a considerable portion of the SERP portfolio. These strands would be managed in an office of internetwork research and development. The participants in these cross-net- work strands would continue to have a home in a network, allowing for growth and continued contributions in the area of dominant expertise. But network members might spend much (or little) of their time in a particular period working on internetwork strands. This type of organization would take ad- vantage of the gains to be made by defining problems both from a scientific perspective and from 60 a ~ practice perspective. STRATEGIC EDUCATION RESEARCH PARTNERSHIP

OCR for page 52
Network Fielc' Sites Each field site will become a partner in one or more of the SERP research networks. As partners in SERP, the field sites will be involved in shaping the network agenda. They may have ideas or interventions that warrant study, or they may want to use the collaboration as a way to become more systematic about rejecting on their practice. They may be committed to closing the performance gap between majority and minority students or to finding better ways to educate children whose home lan- guage is not English and see opportunities to advance such goals in the network research program. The SERP organization will provide several kinds of sup- port for the on-site research program, including outreach and public support, research support, on-site research support, and financial support. Outreach and Public Support The SERP director will provide information about the initiative to state political, education, and business leaders on an ongoing basis, to encourage a continuity of policy interest in, and support for, the work going on at the field sites. This will help the school districts directly involved in the initiative command the staying power needed to bring about real learning and change. Research Support The research arm of each SERP network will include a multidisciplinary team of senior researchers, se- nior practitioners, and SERP fellows (midcareer, postdoctoral, and doctoral candidates in education or education-allied fields). When SERP establishes a research relationship with a field site, this team will work with the site practitioners to negotiate the specifics of the research program and the terms of the collaboration; formulate the specific local version of their joint research agenda; develop the research protocol and map its imple- mentation; provide instructional materials and protocols; provide (or work with the site to develop) needed tools, instrumentation, and data systems; T H E S E R P O R G A N I Z AT 10 N 61

OCR for page 52
provide training/professional development for par- ticipating teachers, principals, and others involved; and engage in discussion, feedback, formative evalua- tion, and course corrections through e-maiT networks, face-to-face meetings, and summer training institutes. SETTING R&D PRIORITIES Who should make the decisions about broad program pri- orities? As with the organization of research, there are different answers that can be supported with compelling arguments. Identifying knowledge bases and research topics in a field that have potential to advance understanding of effective educa- tional practice will require people with deep knowledge of re- search in the areas of relevance, as well as people experienced in research methodology (including the complexities of conduct- ing research on educational practice). But while the knowledge base of researchers is critical to identifying and developing a productive program, they are not uniquely qualified to deter- mine which questions, if answered, would be of most benefit to teachers, or most feasible to implement, or which type of educa- tional improvement would have greatest payoff for society. Teachers, administrators, and policy makers will have impor- tant and very different contributions to make in these regards. If we take the area of early reading as an example, on one hand, researchers may be best suited to identifying weakness in the current knowledge base that could productively be addressed through research like opportunities to better understand the problems of children who do not learn to read even with inten- sive individualized instruction. A teacher, on the other hand, may point out that existing knowledge cannot be effectively used unless there is an investment in research and development on instructional strategies for working with multiple groups of children in a single classroom who require different levels of instruction. School administrators struggling to put well-pre- pared reading instructors into classrooms may argue that the most important investment is in effective professional develop- ment to give teachers of early reading access to current knowI- edge on effective instruction. In contrast, a policy maker may favor evaluation of major instructional programs to determine what works and what doesn't. 62 STRATEGIC EDUCATION RESEARCH PARTNERSHIP

OCR for page 52
numbers of language minority students, or severe teacher short- ages could effectively combine their resources to support ef- forts of particular importance to them. To develop the capacity of the states to operate a compact that can promote the long-term effectiveness of SERP, resources (human and financial) will need to be set aside from the start for that purpose. Some of the required investments in research and development may take years to bear fruit. Staying the course with long-term investments poses a challenge to policy makers whose success is measured by short-term outcomes. If they are to see a new role for themselves in steering the education re- search and development enterprise in directions that are pro- ductive in the Tong run, support for playing that role will be required. More broadly, SERP must attend to building state capabilities to participate in helping to frame, carry out, use, and evaluate research and development. A FLEXIBLE FUNDING STRUCTURE While we propose that core, stable funding for the SERP program would eventually flow from a compact among the states, we think the vitality of the SERP program, and the reach of the SERP infrastructure, will be enhanced if the organization is receptive to participation from a variety of other sources as well. We see two alternative models. The first would invite participation from federal agencies, foundations, regional com- pacts, and businesses at the level of specific research initiatives or specific SERP field sites. This would provide the flexibility for them to simultaneously enhance the impact of their own efforts and the SERP program. Support from the National Sci- ence Foundation for a SERP study of implementation of model science curricula, for example, would provide the opportunity for NSF to fund independent field site testing and evaluation. The results of that work would further both the NSF and the SERP agendas. A second alternative would incorporate the first, but it would also allow for a Tonger-term commitment from the federal gov- ernment agencies or private foundations to the annual SERP budget that would earn them a place on the governing board of the organization. To the extent that the potential partners see the success of a SERP infrastructure as enhancing their own goals and their ability to carry out their own work, the more 68 STRATEGIC EDUCATION RESEARCH PARTNERSHIP

OCR for page 52
extensive participation provided for in this second option would be desirable. ~ . ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ T H E S E R P G OV E R N A N C E S T R U C T U R E Success of the SERP enterprise will require strong leader- ship. To attract that leadership, the director of SERP needs to be invested with broad authority and accountability. Above all else, attracting a highly competent director to SERP will require minimizing political influence on that appointment. The inter- ests of the states, other funders, and the schools and school districts that participate in SERP will be best served, in our judgment, if the quality and effectiveness of the institution are the primary criteria on which leadership is judged. The director would serve at the pleasure of the SERP gov- erning board, being hired (and fired) by this board. It will therefore be necessary for the governing board itself to be com- posed of individuals who can be entrusted to promote and protect the quality and efficacy of the SERP work: individuals who command public respect across party lines for their service in the public interest. The membership and size of the governing board will be decided by those who fund SERP, and will no doubt need to be adjusted as SERP grows and prospers. The committee's image of the governing board is one of perhaps a dozen members, who might be drawn from the ranks of former governors known for their commitment to education in their states, industry leaders with a personal involvement in educational improvement, presi- dents or provosts of institutions of higher education, and those who have served as effective public spokespersons for the im- portance of educational improvement. Fiduciary responsibility for the institution would rest with the governing board. Approval of general program direction, budgets, large initiatives, and oversight of quality in personnel and program would be its responsibilities as well. While the board would approve creation and cessation of research and development networks and the allocation of financial resources across networks, that ultimate responsibility for agenda setting in SERP must reside with the director. The director would be responsible for setting policy, planning, managing, and coordi- T H E S E R P O R G A N I Z AT 10 N 69

OCR for page 52
eating programs. The director's decisions will be informed and guided by his or her key appointees, including the leadership of the research and development networks. Since the networks are themselves collaborations of researchers and practitioners, those communities will influence decision making indirectly. The committee thinks that two advisory boards are also important to bringing an external perspective from the scientific and practice communities to the director. The first is the scien- tific advisory board described above, a board with responsibil- ity for oversight of the research design and the peer review process. We envision this board as a group of paid agents of the institution appointed for their methodological expertise, serv- ing on a part-time, fixed-term basis. A second agenda-setting advisory board would bring perspectives to the director on the pressing problems of policy and practice that should be consid- ered in setting the SERP agenda. Unlike the governing board members, who would be prominent figures, these advisory board members would consist of individuals whose everyday experiences keep them in close touch with the problems of classroom practice and policy making: teachers, principals, cur- riculum directors, superintendents, and legislators. These, too, are envisioned as paid, part-time, fixed-term positions. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ SERF ORGANIZATIONAL STRUCTURE Ultimately, decisions about the structure of SERP will be made by its funding partners and by its original governing and management team. We envision a substantial gestation period, during which the number of initiatives grows and the SERP program matures. Here we lay out a potential organizational arrangement that can serve as a point of departure for discus- sions of a SERP launch. We have not tried to anticipate or resolve every issue of organizational design that an entity of this complexity can expect to encounter. Rather, we have tried to identify a limited number of key features. Like parents caring for a newborn that they hope to nurture to a vital and produc- tive adulthood, we can make some choices to encourage healthy development right now. But many decisions will be best made only after observing and responding to growth as it occurs. In the design we envision, SERP functions are carried out in 70 STRATEGIC EDUCATION RESEARCH PARTNERSHIP

OCR for page 52
four operational units that report to the SERP director (see Figure 3.1~. These are research and development, quality assur- ance, communication and public liaison, and management, bud- get, and administration. We refer to operational units and of- fices as a means of both identifying critical functions that a SERP organization must carry out and indicating lines of au- thority and responsibility that will promote a healthy organiza- tional dynamic. The major challenge of the SERP management team, however, will be one of creating a flexible and dynamic organization. This will require primary attention to the func- tions to be performed rather than the offices that perform them, as well as to the links that keep the functions productively and responsively connected rather than the boundaries that distin- guish them. 1. A research and development unit would coordinate the re- search and development networks and the internetwork initia- tives. The number of networks should be small at the start no SERP Structure Governing Boarcl Executive Committee Scientific Advisory ~ Board r FIGURE 3. I SERP structure. T H E S E R P O R G A N I Z AT 10 N Director Key func tions I r Communication Research and Quality and Development Assurance Public Liaison Agenda Setting Advisory Board Budget, Management, and Administration 7

OCR for page 52
more than three but that number might grow over time. Each network would include intramural and extramural research and development efforts and would partner with one or more field sites (Figure 3.2~. The SERP commitment to partnership would find immediate and visible expression if each network were led by a duo of one practitioner and one researcher. Ideally they would be chosen for their ability to address the network's hub question and their skill at leading research teams whose mem- bers are diverse in professional background and expertise. Above all, they must be open to the possibilities and ready for the challenges of collaborative work. These leaders should be very talented people at the height of their careers. They will oversee the network's overall research plan, nurture productive interac- tions and collaboration among the parts, and provide intellec- tual coherence to the whole. They will also be responsible for seeing that the network provides carefully developed research protocols and the other tools and instruments needed for work in the field. It is crucial that the leaders be closely connected R&D Networks FIGURE 3.2 R&D networks. 72 Office of the Directors 1 1 Extramural Research research protocols Tools and instruments Field sites intramural research STRATEGIC EDUCATION RESEARCH PARTNERSHIP

OCR for page 52
with the central functions of SERP and also intimately familiar with the field sites where SERP is operating. An office of internetwork initiatives would house research that draws on multiple networks and would carry out the same functions of research management as the networks. Field sites would join one or more networks, providing the context for carrying out much of the research on practice. The research and development unit would house five other functions as well. The other functions are: planning and evalua- tion, communications, research services, fellowships and ap- pointments, and data collection (Figure 3.3~. Planning and evalu- ation will operate like an internal consulting group, consolidating knowledge about instrumentation, research protocols, and evalu- ation of research and development projects. Ultimate responsi- bilit~v for each of these tasks would lie elsewhere: in the net- works and the quality assurance office. However, developing capacity, particularly in the early stages, may be facilitated by providing expert support. Although there will be an operational unit responsible for communications and public liaison, we nonetheless propose a communications research and development office housed in the research unit. Effective communication will be the key link be- tween the research and development program and improving student learning. A central message of the report How People Learn: Brain, Mind, Experience, and School (National Research Council, 2000) is the importance of the organization of knowI- edge. It is in that organization that meaning, and a foundation for informed response, is created. How knowledge can be orga- nized to maximize learning and support for knowledge utiliza- tion must itself be the subject of study if SERP is to have an impact on student and teacher learning. Where that study can best be undertaken is a somewhat difficult question. Communi- cations has not been an area of high status for researchers. If SERP is to attract high-quaTity researchers to this critical study, it will have to set out to elevate that work. We propose that this be done in part by infusing communications questions through- out the research program. When teacher learning is the subject of study, communicating findings effectively to teachers and schools of education should be on the agenda. When policy incentives are the topic, communicating effectively with policy makers should be a subject of study. We do not propose that the communications functions themselves be carried out by the T H E S E R P O R G A N I Z AT 10 N 73

OCR for page 52
a) Hi Q a) a) Q o a) Ct S Ct a) En a) 74 ~5 ~ in in ~ Q Cl) S ~ it._ O O = Q Cl) Q IL ~ id O ~ o o Cal Ct O ~ ~ Cal .O Ct O ~ O 'Cal _ _ in O ~ C' 8~ Q o - ~5 c' in G' 2 fir .o lo, o CO ~ .2 CO M :e o . ~ o fir u v) STRATEGIC EDUCATION RESEARCH PARTNERSHIP

OCR for page 52
networks, but that the research to support communications be done here. Research services would house a variety of functions to facilitate the process of working in field settings. It would handle terms and conditions of participation, contracts and agreements, negotiations for time release and job protection, and human subject protections. It would also handle library and informa- tion functions, and other services common across research and development initiatives. Since the SERP proposal entails a substantial increase in capacity to carry out research in field settings, our design as- signs the task of capacity building to a fellowships and appoint- ments office that would recruit and provide learning opportuni- ties for postdoctoral fellows and arrange for visiting positions for university professors and for teachers. There are many Ph.D.s in science, mathematics, and engineering who are not employed by universities, and whose expertise might productively be chan- neled to education research (National Research Council, 2002a). SERP would provide an opportunity that does not now exist for training in education research. This function is housed in the research and development unit of the organization to ensure that capacity is developed where it is needed to support the programs. Each of the research efforts would presumably involve data collection, but we propose a separate data management office that would build a longitudinal dataset that incorporates a com- mon set of measures for all participating schools. The key con- stituency for this office will be the networks themselves. This office, like each of the networks, may undergo dramatic change through the gestation period. If the states in the compact wish to invest in a common data collection system, for example, they could decide to undertake their existing, regular data collection and reporting in a common format and to have that data aggre- gated at SERP headquarters. This would greatly expand the size and function of the data office, giving states a far greater capac- ity to understand and guide the education system. Those responsible for each of the functions described would report to a director of research and development, who would hold major responsibility for coordinating the efforts. The direc- tor would monitor if and when projects in different networks overlapped and required coordination, or when work in one T H E S E R P O R G A N I Z AT 10 N 75

OCR for page 52
network should inform that of another. This individual would also be responsible for monitoring the work of other education research institutions, actively looking for opportunities to lever- age and coordinate existing efforts. 2. A communications and public liaison unit would provide a bi-directional link between the SERP program and schools and teachers across the country. The vision of SERP as a place for teachers, administrators, policy makers, and researchers to turn for well-organized information and access to the current knowI- edge base will be realized only if a substantial investment is made in that function. Beyond making relevant knowledge usable, a SERP organi- zation must attend to getting usable knowledge used. Accumu- lation and sense-making, in other words, are not all that is required. Effective communication and support for knowledge utilization including navigation through the knowledge base, technical assistance, and connections to others inside and out- side the SERP organization are essential. For example, SERP needs to be a place to turn when a fifth grade mathematics teacher is looking to understand and solve her students' problems learning long division. We see it as a place where a superintendent can go to get performance data on curricula that have been studied in SERP field sites or informa- tion about the in-service preparation requirements that accom- panied successful interventions. Exploitation of new technolo- gies will give SERP a platform to engage the research, practice, and policy communities in dialogue about what the research findings mean, or the kinds of experiences teachers are having as they implement a particular curriculum, or the latest perfor- mance data in districts that have implemented one of the SERP- sponsored programs. This office will itself conduct or commission from outside sources research and program evaluation aimed at ascertain- ing the effectiveness of communications with the diverse audi- ences that SERP intends to reach. Close coordination between the research and development networks and the communications unit will be critical and will warrant the creation of positions in both with responsibility for that coordination. The appointment of an associate director for 76 STRATEGIC EDUCATION RESEARCH PARTNERSHIP

OCR for page 52
communications who reports to the SERP director would en- sure that the function of communications is elevated. The functions of the communications and public liaison unit would include handling public information, news and media, technical assistance to schools, and online information. Over a period of gestation, we can imagine the efforts of this office expanding greatly, creating, for example, online communities of teachers who are working with a particular curriculum, topic, or problem. This office could also provide support to the SERP director in undertaking capacity building for those involved in the state compact. 3. The quality assurance unit would have two major func- tions: outside review and SERP program evaluation. The two functions are quite distinct. The outside, peer review process would provide assurances regarding the quality of the SERP work. Scientific standards would be central to this evaluation. The SERP program evaluation function, in contrast, would focus on the value added by the SERP endeavor. The charge of this office from the outset would require the development of measures of SERP program effectiveness. These would prob- ably involve evaluative feedback from the SERP user commu- nit~v and funders, as well as efforts to measure the impact of SERP on education policy and practice. This last goal without question will be a major challenge, but one the committee thinks warrants a serious effort from the start. The education enter- prise is huge. A research and development effort aimed at im- proving its productivity, if it were eventually allocated just 1 percent of the education budget, would itself be huge. As in the private sector, the decision about how much to allocate should depend on the contribution that research and development makes the productivity of the investment.3 Unless an effort is made to develop measures of that productivity, SERP could become a set of bureaucratic commitments rather than a vibrant research and development enterprise. 4. Finally, a management, budget, and administration unit would house financial, personnel, and facilities management functions. 3More precisely, the allocation should depend on the expected productivity of the investment based in part on past returns. T H E S E R P O R G A N I Z AT 10 N 77

OCR for page 52
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ S U M MA RY Our proposed design will create, in our judgment, the con- ditions for a newborn SERP to thrive. Its essential features can be summarized as follows: A governance structure that ensures accountabil- ity, broad representation of the views of the variety of education stakeholders, a long-term commitment to cen- tral goals, and the scientific integrity, quality, and use- fulness of the program. High-quality leadership of the organization chosen for substantive expertise and a commitment to the mis- sion of the institution. Research and development networks designed as collaborations of researchers, practitioners, and policy makers, linked to field sites that allow for the study of practice and policy. An organization of the research and development program designed to simultaneously push the bound- aries of academic disciplines in areas of promise for improving student learning and to bring together mul- tiple fields of research relevant to problems of practice to improve understanding and decision making. Monitoring of quality assurance and program im- pact by those not directly involved in the conduct of the program. The development of capacity building within the institution to carry out the type of program SERP seeks to undertake. Elevating communications to a high-priority func- tion that will include research and development. As SERP grows, we would expect the organizational struc- ture to be reshaped and refined, its areas of expansion respond- ing to the demand of the field and the successes and failures along the way. 78 STRATEGIC EDUCATION RESEARCH PARTNERSHIP