DEMANDS for greater productivity and increased accountability in America's public schools are loud and ubiquitous. Reform and restructuring efforts have addressed almost every aspect of schooling, and state and federal policy makers are increasingly setting standards and specifying criteria for assessment that schools must meet. In this atmosphere of heightened accountability, a series of questions driving school reform have become central: How can schools help all students meet high standards? Who sets those standards? How is student progress best assessed? Who should do the assessing the state, the district, the school? What is the relationship between the external mandates and student achievement?

Recent accountability reforms have moved both the standards and the criteria for meeting them outside the school. But we have learned that what happens inside a school is key. Some schools, which we refer to as inquiry-minded schools, have already incorporated issues related to standards and assessments into their culture and are improving as a result.1 These school communities recognize that improving teaching and learning is an intentional and ongoing process. They ask themselves important questions and have the courage to act on their findings. Because they recognize that evaluation leads to action and that every action creates new questions, they have institutionalized the process of reflecfive inquiry. These schools have become internally responsible for the creation and maintenance of standards.

At the Annenberg Institute for School Reform at Brown University, we have been studying recent accountability movements and their influence on schools. Our aim is to help schools and their communities use data effectively and systematically and to help policy makers attend to the realities of school life as they craft accountability systems. This past year, we observed the work of 18 Annenberg Challenge schools in six states. These schools had been brought together to share their data-based practices. We also followed other schools participating in accountability projects. All these schools are attempting to reconcile the requirements of state- or district-level accountability systems with their own local needs.

In this article, we draw on the experiences of the most successful schools to offer a picture of what a school can do to take ownership of internal and external standards and to use data from state assessments and other sources to improve instruction. We define accountability, describe the mindset of schools that have institutionalized reflective inquiry, and explicate the inquiry cycle. We also consider the challenges that schools face in addressing standards and that policy makers face in developing accountability systems to support the delivery of high-quality instruction. We begin with three scenarios that describe inquiry-minded schools faced with external accountability requirements.

Monarch School

Students at the Monarch School appear to be achieving at high levels. During the current school year, 98% of the students who were tested passed the statewide tests in reading, 100% passed the tests in writing, and 97% passed the tests for mathematics. These figures exceed the statewide average by at least 12 percentage points in each area. Moreover, the school's passing rate exceeds that of other district schools by even larger margins. Gerry Macia, the principal at Monarch, is tempted to let the school rest on its laurels. Still Gerry brings the summary report sheets to the next school council meeting for discussion.

Gerry: I'm genuinely excited to share our report card and scores with you this year. Notice how much they have come up - even since last year! I see real improvement here. This achievement is remarkable because we are not exactly a "privileged" school. Remember, more than half of our kids are on free/reducedprice lunch; they come from diverse backgrounds; many are considered "at risk." And we're not a small school. We have over 800 students in five grades. It seems we have accomplished what all this reform talk aims for.

Sandra (third-grade teacher): Great! I'd like to End out how we did this. We've brought in so many programs in the past two years as part of the Monarch 2000 Initiative - I wonder which ones are doing the job. And are any not worth the time?

Marty (fourth-grade teacher): It's good to hear that our kids did so well. But I'm still a bit skeptical. I just don't feel that all my students know the material that well. Maybe we ought to look a bit more closely at the score reports. I still have some questions.

Yvonne (fourth-grade teacher): I think we need to ask exactly what 98% passing means. What does "passing" mean?

As the members of the group examine the reports from the state testing bureau, they notice gaps between the high percentage of students who passed and the lower percentage of those who mastered all the objectives. For example, Marty sees that, while 96% of the fourth-graders passed in reading comprehension, only 62% mastered all the objectives. Yvonne notes that the gap is even larger - 96% passing versus 40% mastery - in math. They discover that the standard for passing actually means meeting minimum expectations.

In response to this discovery, they decide to focus their inquiry on the gap between passing and mastery. Soon, they are generating questions about student learning. What does it mean to meet minimum expectations? What does it mean to master an objective? Which objectives do students master? Which are they not mastering? Which students master them and which do not? Does mastery tend to clump along lines of gender, race, socioeconomic status?

These questions on learning generate questions about teaching. What are we actually doing in our classrooms? What programs do we use, and are we following them? Do our standards and objectives match those of the tests? If not, how do we reconcile the test with our curriculum? Are we teaching for mastery or for minimum expectations? How are students responding to each program?

The school council first prioritizes the questions and identifies data that can help answer them. This activity leads them to propose additional data collection and analysis. For example, the council decides to analyze the state's report to understand better what the apparent gap means. Then they can focus their inquiry on a specific group of students or on a specific set of objectives. They also decide to seek data on which children participate in specific programs and to gather evidence on how children in these various programs perform on specif is objectives.

Gerry concludes the meeting by reminding council members that their questions about last year's scores had led to their decision to implement a new literacy program and by suggesting that Monarch revisit that decision in light of the new data. "The scores do show that we have improved. But we can't stop here. I agree that we have to ask some hard questions about what look like perfect scores. I hear you saying that we want to aim for mastery, not just minimum expectations. It seems we aren't reaching every child, so we still have room to improve."

Valley Middle School

Several years ago, Valley Middle School adopted a curriculum and pedagogy that recognize and develop multiple intelligences. Since that decision, the state has written and adopted the New Standards for Achievement, and statewide tests are administered to determine how individual schools are meeting these standards. Valley Middle School is concerned because the state tests take into account only the traditional conceptualizations of intelligence. The school's council meets to consider how Valley might handle the state's release of the annual "school report card," which is based on the school's scores on the statewide tests.

Terry (principal): This year's scores are going to be pretty important. It's our second year using our Many Pathways curriculum. We're asking our kids to do more in class than take traditional paperand-pencil tests, so we may find that they won't do as well on the standardized tests. How should we feel? And there might be parents who agree with us on multiple intelligences but won't be happy if they find their children can't do what they think of as the basics.

Nina (a parent): I guess I'm one of those parents. Did adopting a "multiple intelligences approach" mean that our kids can't meet the state standards? I thought this way would make it easier for all kids to do better in all areas.

Jim (sixth-grade teacher): Eventually, I expect it will. fm concerned because we're still new at this, and that might make the scores go down for a while. I believe in it, but I want more time before I am judged.

Terry: So the test scores may raise a lot of questions, some of them legitimate. Let's think ahead. What are we going to want to know?

Carol (eighth-grade teacher): If the scores show some areas where my students' achievement has decreased, I'll want to know why. I'll want to analyze those scores in light of what I am asking my kids to do.

Maria (fifth-grade teacher): You know, I am asking my students to do things they are not tested on - never have been tested on. Speaking of assessment, I'd like to assess how well they do these things. But they aren't things you can measure on a paper-and-pencil test I care that my kids are achieving; I just want to be sure that the tests get at what we are asking the students to do. We need new kinds of evidence.

Nina: We parents want that too. The curriculum is supposed to be building my child's interpersonal intelligence and his spatial intelligence. Do we know that this is happening? I really do want to know what you are doing and what dif ference it makes for my Carl. The school has been using Many Pathways for two years. Carl moves to the high school next year. Now is the time to drop the program if it's not helping Carl learn.

Carol: The big question for all of us is, What difference is the Many Pathways curriculum making for children's learning? It seems to me that we can use the state's standardized tests along with other kinds of evidence to answer that.

Based on this discussion, the council explores questions about teacher assignments and student work. It focuses its inquiry on a list of related questions. What do teachers assign? What do they expect students to do? What does good work look like? What work are students producing, and how does it match our expectations? What exactly do the standardized test items measure? To answer these questions, the council decides to collect samples of teacher assignments and student work to analyze and compare them to the information from the standardized test.

Uncas School

The school improvement council at Uncas School, along with all such councils in the state, has been charged with amending its School Improvement Plan (SIP) in accordance with the strengths and weaknesses identified through statewide testing. Principal Pat Summers convenes the council in the fall and releases the news that nearly 80% of Uncas School students scored at the proficient or exemplary level in all areas tested. Pat is delighted with the improvement over last year and suggests that the council pinpoint what the school is doing right and support those activities in its revised SIP.

Dale (fourth-grade teacher): We had a miracle here! If I read these reports right, we've come from having only 22% of the students at a proficient level or better in math last year. And from only 9% in writing. That means we reached a lot more kids than we did before. I teach the grade that is tested, and if those scores are right I have only one or two children who can't do the work. Sounds good, but I'm not sure I buy it.

Maria (fourth-grade teacher): I don't think I did anything that different this year. I'd like to think that putting the SIP in place made me a better teacher, but I can't say how.

Leah (third-grade teacher): I can't even tell whether we went up in reading because last year's reading test used "high-medium-low" instead of "proficient" and "exemplary." Anyway, doesn't it say here that the assessment used a different test this year?

Alton (a parent): I think we need to look a bit more closely at all these scores before we make any final evaluation of our progress. Notice how last year, when our students scored quite poorly, the middle school kids did so well? Their proficiency and honors levels were all above 50%. Does this mean that our kids get smart after fourth grade? Or that teaching improves after fourth grade? The numbers alone do not show us anything.

Pat (principal): Only one thing is certain: we really cannot make our judgments based on this single assessment - especially if we don't truly understand it. We can ask the state for further clarification, but we should also see what other data sources we have to evaluate our students' learning.

The council decides to focus its inquiry on student learning and spend the rest of the meeting identifying assessments that are relevant for the specific needs and objectives of Uncas School. The next meeting's agenda is set: to design an evaluation plan that makes sense of the data that have been gathered.

These three brief vignettes illustrate the power of strong internal accountability capacity to incorporate the demands of external accountability systems. The experience of these schools is supported by a study by Fred Newmann, Bruce King, and Mark Rigdon of the accountability structures of 24 restructuring schools. They found that, even if high-quality standards (and incentives) were provided, many schools would be unable to meet them because they lacked the necessary professional knowledge and skills, appropriate curriculum, and adequate materials and facilities. In schools characterized by strong accountability systems, the authors found that:

Staff identified clear standards for student performance, collected information to inform themselves about their levels of success, and exerted strong peer pressure . . . to meet the goals. In some schools, strong internal accountability was accompanied by compatible external accountability, but in others, internal accountability existed without, or even in opposition to, external accountability requirements.2

What's Missing from Accountability Systems

Unfortunately, most external accountability approaches have paid little attention to creating the internal capacities required to carry them out. Instead, they have fostered a tension between the public's legitimate need to know and a school's legitimate need to explore its own questions. The term accountability frequently makes teachers and principals uncomfortable because they see the questions asked and the data collected as originating outside their work. Accountability appears to be public and external rather than a central component of their own practice; they see themselves as "held" accountable, not as "being" accountable.

To be accountable means to be obligated to understand and explain one's actions. Accountability relies on feedback; it links performance with results. Thus accountability in schools is not only about results but also about every aspect of teachers' actions. What are we choosing to teach our students? How should we instruct them? How will we know when they know it? What will we do when not everyone learns? Put simply, practitioners who are accountable evaluate their own practice and then use the information to improve From this perspective, accountability is the foundation of successful practice because it entails knowing what we do and learning from that knowledge.

Often, however, assumptions about the nature of schools and their internal capacities limit the effectiveness of current accountability systems. One such assumption is that high standards + assessment + incentives (or consequences) = higher student achievement. This equation overlooks two key components of effective accountability: capacity and shared accountability. It assumes that schools have the capacity to improve and lack only the standards, assessment tools, and incentives to do so. However, policy makers must ask a key question: Once schools address standards and receive assessment data, what do they do? To answer, policy makers need to understand schools and the conditions that lead to continuous improvement. They then need to develop policies that foster the leadership, collaboration, and skills that are essential to school improvement and to accountability. To strengthen these, schools need both time and money. Accountability systems can then accommodate and support specific practices and conditions as well as provide the structure of standards, assessment, and consequences.

Another assumption that appears to underlie many accountability systems is that schools alone are accountable for student learning. In fact, the responsibility for educating children is shared by many different players, including families, state policy makers, and the business community. This notion of shared accountability in no way releases schools from their responsibility for student learning. However, it does put schools in the active position of being accountable rather than in the passive position of being held accountable. It also places families and the community in an active relationship with the schools and makes clear the need for schools to involve families and the public in the ongoing process of data collection, inquiry, and decision making.

Four Current Approaches

Four approaches to accountability are currently being tried. Two of them - performance reporting and market-based approaches - emphasize external accountability.3

Performance reporting assumes that the collection and public reporting of information on student academic performance, combined with negative or positive consequences, will stimulate improvements in teaching and learning. In reality, no such direct link between performance reporting and school improvement has been demonstrafed. Statewide assessments do not always accurately measure local or even state standards, so schools remain unclear about how to use the data. By the time scores are reported, teachers have difficulty relating them to their instruction.4 Furthermore, high-stakes consequences can lead to a narrowing of the curriculum through teaching to the test and other unintended consequences.5 Finally, the narrowness of measures and the inappropriateness of standards have raised issues of fairness, corruption, incompetence, and technical complexity.6

The market-based decision approach advocated by some representatives of the business community and by some educational theorists holds that, if individual schools are held accountable for overall student performance, consumer choice will force schools to improve. `Bad" schools will close, and "good" schools will thrive. This approach assumes that schools alone are accountable and that parents, as consumers, have the resources and ability to make informed choices. This assumption overlooks data that indicate that parents often make choices for reasons other than performance data.7 It also overlooks the deleterious effects of this approach on equity concerns. Finally, data indicate that this approach exerts little pressure for school improvement.8

Market-based and performance-reporting approaches to accountability fail to take the school itself into consideration in their formulas for improvement. Both rely solely on external accountability and have not shown themselves to be effective in improving the learning environment in schools for all students. External accountability alone does not guarantee high performance if the schools lack the internal capacity - that is, the requisite human, technical, and social resources - to improve.9 Because the external standards and inducements of these approaches seldom align with the specific culture and values of a given school, they have little meaning for teachers and principals working in the school. Consequently, they tend not to affect practice.

Two other approaches to accountability - changes in governance and teacher professionalism - acknowledge the importance of local schools but have also failed to reconcile successfully the relationship of internal accountability practices and habits with external accountability requirements. Changes in governance are promoted as a means to hold accountable those closest to the delivery of services. This approach assumes that shifting to site-based management will allow individual schools to improve student achievement by allowing them to respond to context-specific needs. In practice, however, decentralization is not always clearly linked to student achievement, and local boards and school councils seldom define and measure that for which they will be held accountable.

Teacher professionalism, an emerging approach that is rooted in the assumption that professionals are accountable to their clients, also locates accountability within the school.10 Ideally, when they are seen as professionals, teachers are accountable for delivering a defined body of knowledge and skills to their clients - the students. With professional and accountable teachers, schools are themselves accountable. Teacher professionalism assumes that the profession employs a self monitoring mechanism to ensure that its members collect and use feedback as a means of being accountable for improving student learning. However, the idea - and indeed the reality - of self monitoring remains vague and varied across the teaching profession, for there is no formal monitoring process.

Many teachers and principals seek to be professionally accountable to themselves, but most avoid public inspection and explanation of their actions. Surrounded by data, they often cannot see how such evidence might inform practice, and they often lack workable mechanisms to make sense of it. Those who do try to act on the data usually do it alone. Few structures are in place to facilitate collective analysis and decision making. Thus teachers close their doors, principals buffer the community from the actual teaching and learning that take place in schools, and schoolpeople play it safe by avoiding risks that might lead to improved learning.

Combining Internal and External Accountability Through Inquiry

Most schools lack the capacity to improve student achievement through external or internal accountability alone. Both forms of accountability are necessary components of improvement. Fred Newmann and his colleagues emphasize that both must be interwoven into a school's culture for the school to be successful.11 The key is finding a way in which the two systems can work together.

We believe that the needed bridge is the cycle of reflective inquiry. Therefore, we propose a combination of external and internal accountability practices built on reflective inquiry. Through reflective inquiry, the school community achieves the capacity to define its goals by setting its own standards or by incorporating and taking ownership of external standards. It also achieves the capacity to evaluate progress toward those goals and to take corrective action. As a result, the cycle of reflective inquiry produces accountability and builds greater capacity. We assert that schools themselves - not the state, not the market, not the profession - build in the self-monitoring mechanisms and reflective inquiry necessary to ensure continuous improvement. Successful schools are inquiry-minded schools.

Our earlier discussion of Monarch School, Uncas School, and Valley Middle School suggests that inquiry-minded schools are in the best position to balance reasonable external requirements with their own internal improvement needs. A closer look at the practices engaged in by such schools can help other schools struggling to make meaning of assessment data and other information and can help policy makers who are developing systems that promote school improvement.

The Mindset and Activities Of Inquiry-Minded Schools

As an integral part of their daily functioning, inquiry-minded schools examine their practices explicitly, publicly, and collectively. Each school is guided by an ethic of continuous improvement that makes it willing to take risks that are reasonable and supported by evidence. The mindset or culture of an inquiry-minded school perceives questioning, seeking data, reflection, and subsequent action as the steps that are necessary to improve performance and get things done in the school.

Inquiry-minded schools focus their inquiries on what Grant Wiggins has called "the gap between the ideal and the actual." This gap provides schools with essential questions related to improvement: What does our dismal performance reveal about how we might adjust things? How might we better capitalize on our strengths and minimize our weaknesses? Such questioning, Wiggins asserts, reflects the "trial and (learning from inevitable) error" by which "one's teaching [is] altered to respond to evident need."12 The process first requires definition of the "ideal" and the collection and examination of data to reveal the "actual." The absence of such data makes it "hard for schools [and their communities] to reflect on what they're doing and make adjustments."13

Inquiry-minded schools naturally and automatically engage in six activities that, taken together, we define as the "inquiry cycle."14

Establish outcomes for which we accept responsibility. Participants discuss and define satisfactory outcomes and establish standards for judging the quality or degree of success toward achieving those outcomes. To do so, they ask the following questions: For what exactly are we responsible? What do we expect students to do as a result of our teaching? How will we know whether they are meeting our expectations? What standards will we use to judge our success? What will success look like? How does it relate to our purpose and vision? As we saw in the scenarios at Monarch, Uncas, and Valley schools, each engaged in healthy conversation about the gap between their own and state standards in order to draw a clear picture of what success would look like for them.

Central to this activity is agreeing on who decides on and accepts responsibility for outcomes: Who needs to be at the table when defining the outcome? Does the conversation include all responsible parties? Who is responsible for what activities? Whose outcomes are we discussing? What is the school's role in achieving the outcome? What are the impediments to achieving the desired outcome? What supports do we have? What do we need to do to accomplish the outcomes? What resources do we need, and from whom can we get them? Such questions are easily overlooked because of time pressures. Yet the degree to which the decisions made at the end of the cycle are realistic and supported often hinges on the answers.

Identify important questions concerning student learning. Participants articulate key questions about learning in the school, based on experience, concerns, or data: What do we want to know about our students? Do we have intriguing puzzles or troubling issues we want or need to address? Is something happening that pleases us or bothers us? What do we do well already, and how do we know this? What do we do poorly, and how do we know that?

Identifying questions related to student learning focuses the school's inquiry. This can be important, since schools suffer from a glut of data that can pull them in diffuse directions. Only by identifying and prioritizing questions can schools organize the information they have available. Sources outside the school also play an important role. The state or district may question standardized test scores; taxpayers may ask if programs are working; parents may want to know if their children are learning enough to be competitive in college or in the job market. Such questions serve to enrich the perspective of school professionals.

Collect and manage data derived from the assessment of performance (e.g., test scores and writing samples). This activity gathers the needed evidence. Much of the data already exists (e.g., test scores and attendance records), either because of external mandates or because of routine recording. However, participants in the inquiry cycle must locate the data that exist and organize them for their own use. They must also ask themselves such questions as, What form will the evidence take - scores, reports, portfolios, stories? Where can we find it? How will we collect it?

Schools that wish to pursue accountability through inquiry recognize that any outcome or perception must be corroborated by other evidence. Therefore, they seek multiple measures, for example by supplementing data from multiple-choice standardized tests with data from practice. In general, schools want data that are generated at the school, and they want those data stored and accessible so that they can analyze them and compare and contrast various data sets.

Conduct mindful analyses of the data in light of the desired outcomes, and interpret information in light of the school's purposes. This activity transforms raw data into information that can be used to improve practice. Participants conduct close readings of all materials, including both numerical and narrative data. The data are grouped according to the school's questions, which guide the analyses and allow participants to note patterns and rules, to articulate relationships, and to further categorize the data. Participants consider unexpected outcomes as well as relevant models or theories that might clarify ambiguities. In sum, they interpret or assign meaning to the data. These interpretations then raise questions for action.

Take action based on knowledge. This activity requires the courage to face the results of the inquiry process and the willingness to make changes where needed. Participants ask a number of tough questions: What is actually happening? What practices should we continue, and how can we strengthen them? What practices do we need to change? What supports and resources do we need in order to adjust our practices? Based on their answers, they take action. Successes and failures have meaningful consequences, and participants may choose to change goals, instruction, curriculum, technical structures, or materials.

Assess the effects of actions. At this point, the inquiry cycle begins anew. Participants realize that any resulting actions or changes reflect their best guess - albeit a well-informed one- about what will bring about the desired improvements. Again, they analyze the effects of these changes by asking questions: What results will satisfy us that our changes have worked? Who should be at the table in this investigation? What will constitute evidence? How will we collect and analyze it?

The six activities that we characterize as the inquiry cycle attempt to capture a complicated, nonlinear, ongoing process that involves the interaction of behavior and thought. Each activity is essential, and none may be omitted. However, the steps may be reordered to suit the local context.

As illustrated by the scenarios at Monarch, Valley Middle School, and Uncas, participants may enter the process at any point, and they may find themselves revisiting a previous activity or jumping ahead from time to time. For example, analysis and collection can occur simultaneously, as can corrective action and data collection. The scenarios at Monarch and Uncas showed schools defining questions while analyzing data. In contrast, Valley Middle School developed its questions first. Local context determines the entry point and the order of inquiry.

The Challenges of Institutionalizing Inquiry

We are not proposing yet another educational fad. The inquiry cycle is an integral and natural process of professional practice for any good teacher or principal. Professional educators make decisions that are student-centered and knowledge-based. Reflective inquiry is essential to fostering effective learning. Through that process, educators are accountable. Expanding this facet of professional and individual practice into school culture, however, presents several challenges for teachers and principals.

1. Taking collaborative action as a school and community. Teachers and principals in the same school may engage in individual reflective inquiry and may share an accountability mindset, but these individual practices do not make a school an inquiry-- minded school. Institutions like schools - loosely coupled collections of individuals - may have difficulty practicing the inquiry cycle collectively. All individuals do not necessarily share the same purpose and vision. The first challenge in creating an inquiry-minded school is to harness individual energy to address whole-school improvement.

The first step requires articulating and defining individual beliefs and values so that the group can construct a purpose and vision that all can support. The next step is to generate questions that all can agree are important to answer collectively. Externally defined standards - from the state or elsewhere - can offer a concrete focus for these discussions.

A school community that has not previously engaged in this sometimes painfully democratic process may find the task difficult until it becomes embedded in school culture. Conceptualizing and articulating shared values and definitions is hard work for a group of people who are only loosely connected through the school. Yet it can become second nature if the school has the leadership, inquiry skills, collaborative habits, and adequate resources to generate questions about its goals and practices. Conducting the process publicly ensures that all responsible parties come to the table, making the community, not just the school, inquiry-minded.

2. Choosing the questions. Schools overflow with activity, so people have trouble knowing where to begin. Articulating the questions starts the process by establishing priority and focusing direction. But many schools report feeling bombarded by questions from the outside. The inquiry-minded school willingly expands its boundaries to include all questions that are relevant to teaching and learning. Questions that are essential to inquiry may arise from any part of the school community. Some originate outside the local context but offer an important perspective; others may need to be adapted to the specific setting. Teachers, principals, parents, and students may contribute to the original pool of questions to be considered.

The school community then sets priorities with regard to which questions must be addressed first. Prioritizing requires that inquiry-minded schools clarify their questions, establish perspective, and place their questions within the cycle. Schools do this by asking questions about the questions: What is this question about - instruction, resources, or student behavior? What do we already know about this issue? What assumptions do we bring with this question? What political, economic, or social influences may shape or underlie this question? What do we hypothesize might be an answer to the question? What theories and models might help explain the answers we get? Who cares? From a series of discussions, schools choose those questions that get at issues or concerns that are most important to them now.

3. Recognizing important data and managing and using information. Schools are deluged by data: demographics on students, courses in which they enroll, programs to which students are assigned, attendance figures, dropout rates, parent responses, standardized test scores, results of teacher-made tests, results of performance assessments, and so on. By themselves, these data say little or nothing that is useful in shaping school practices. They are simply words and numbers, waiting to be organized into meaningful patterns.

The challenge facing schools attempting to develop an inquiry-minded culture is how to recognize the important data, how to make sense of them, and how to take action to improve student learning based on that understanding of the data. Members of an inquiry-minded school community will determine what data are relevant, how to gather and record them, and how the various bits of data relate to one another. The key to this complex process is found in the guiding questions that serve to organize the data into meaningful patterns.

Schools then use the information the data have revealed to address questions about teaching and learning: Now that we know how our students are performing in a particular area, what are we going to do about it? What do we need to do to improve student learning? How are we using our resources? What activities or programs do we need to support? What activities or programs do we need to change or eliminate?

Challenges of the Inquiry Cycle for Policy Makers

Just as schools can manage external requirements better if they already have an internal process of continuous improvement and an ethic of accountability, we believe that policy makers can design better requirements if they tailor them to schools' inquiry processes.

Perhaps the greatest challenge policy makers face in doing so is their need to question the two prevailing assumptions regarding school accountability: one assumes that schools are solely responsible for student learning; the other assumes that schools have the capacity to meet high standards if only they have the will to do so. Once accountability is seen as shared, policy makers can act to deliver high-quality instruction by asking specifically what schools need and from whom. Once accountability is seen as an issue that involves a school's capacity as well as its will, policy makers can seek ways to develop the capacity for inquiry within schools. They can then build appropriate supports and resources into the accountability system.

Many state policy makers are already keenly aware of the need to change the conventional views of accountability. Some states - such as Vermont, Kentucky, Rhode Island, Maryland, and Massachusetts have already put systems in place that reflect a new way of thinking about accountability. Other states are in the process of modifying or extending existing systems. One of the challenges facing these states is to reeducate a public that has been barraged with oversimplified views of accountability.

"Sorry I'm late, but I had a wide-ranging discussion with the principal about disciplinary procedures."

Conclusion

We have seen and studied schools that are successfully using an ongoing inquiry cycle to improve student learning. Because these schools accept responsibility for demonstrating effectiveness, they are always asking questions about any newly adjusted practice. The questions may lead them to redefine their purposes and intended outcomes, or they may lead to further data collection. Wherever these questions lead, the inquiry process builds the capacity to improve as the school's knowledge base increases. Educators in these schools are active professionals who take charge of their own work and its impact. We hope that the look inside inquiry-minded schools that we have offered here will help policy makers and educators alike face what Laurie Olsen and her colleagues have described:

The challenge facing the restructuring and student advocacy movements is to work together to design a new kind of accountability. This new approach should foster the reflective climate within schools to work with the data that can tell them the most about student achievement . . . and should also provide a supportive external system of accountability. Weaving the two together is where the creative work must begin. As the restructuring movement progresses, . . . it is only through strong data systems, reflective processes, and external accountability that reform efforts will hold themselves to delivering schools that work for all children."