Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13

Converging Reform "Theories" In Urban Middle Schools: District-Guided Instructional Improvement In Small Schools Of Choice

by Chrysan Gallucci , Michael S. Knapp, Anneke Markholt & Suzanne Ort - 2007

Purpose/Objective/Research Question/Focus of Study: The purpose of the paper is to explore an instructive case in which two potentially opposed reform theories converge on schools, in order to understand how productively or unproductively the two theories coexist. We attempt to answer the following questions: Can these two reform theories coexist, or do they get in each other’s way? In what ways, if any, do the two complement each other? Separately, or together, how do they affect instructional practice and the school-level conditions that support teaching and learning? Finally, can policies and leadership, in this case reflecting two such different reform theories, provide mutually supporting conditions for teaching and learning in urban schools?

Setting: The theories are analyzed as they occurred over the 1990s and into this century in four middle schools in one New York City Community School District.

Research Design: The study utilized a qualitative multiple case study design. Four small middle schools within one district were sampled using criteria related to student population, school structure, and configuration of resources such as time and staff. Data were collected over three years during three site visits per year and included extensive interviews, observations, and analysis of pertinent documents.

Conclusions/Recommendations: We argue that the reform theories primarily complemented one another in this case. In the schools, differences in steps taken by school leaders and staff to realize the first reform theory enabled schools to respond productively to, rather than resist, district initiatives based on the second theory. The sophistication of the district’s vision for teaching and learning (both for students and adults), combined with a flexible approach to implementing the second reform theory, reduced the likelihood of conflicting reform messages.

Urban school reform strategies diverge sharply over the role that school and district leaders or central offices can play in stimulating, guiding, and supporting substantive changes in teaching and learning. In fashioning these strategies, leaders at both levels struggle with what high-quality teaching in urban settings looks like. Questions center on the locus of authority and the role of innovation (how much to encourage and nurture school-level discretion and innovation), variable capacity (how to accommodate the substantial differences across schools in capacity to realize reform, while pushing for improvement in the schools that struggle the most), and equity and accountability (how to ensure all children get served well and that all get the same access to learning opportunities). Overseeing schools with substantial inequities, highly varied capacity, and endemic low performance, while functioning within state reform contexts often preoccupied with testing and accountability, it is no wonder that urban district leaders lean toward a reform strategy and underlying “theory of action” that centralizes control and initiative and asserts a compelling notion of how instruction may be improved. To find a recent example of such a strategy and theory at work—in which change is driven by strong leadership from the "top" and guided by the district leaders’ explicit ideas about learning, instructional improvement, and system change—one need look no further than the dramatic attempts to renew teaching and learning in San Diego (Darling-Hammond et al., 2005; Hightower, 2002).

Interestingly, the first four years of San Diego’s district-driven reform was led, in part, by Chancellor of Instruction Anthony Alvarado, who earlier in his career had embraced a reform theory that emphasized differentiation at the school level. As Superintendent of District 4 in New York City during the late 1970s, Alvarado promoted school-level innovation and the creation of “small school” communities, a theory of reform strikingly different from that mentioned above and one thought to be especially good at engaging young children and developing strong teacher and parent commitment to the renewal of teaching and learning. During his tenure in District 4, examples of powerful urban school communities emerged (Fliegel, 1993; Meier, 1995). But later Alvarado rejected this strategy of “letting a thousand flowers bloom,” on the grounds that not enough of them did bloom. In well publicized work within New York City’s Community District 2, and in what would be a precursor to his San Diego strategy, Alvarado switched gears and opted for an approach designed to promise greater equity, guided by a strong, district-level vision of learning and professional work (Elmore & Burney, 1997; Stein & D’Amico, 2002).1

While Alvarado moved on from a strategy that relied on school-level innovation and differentiation, others have not. Throughout the 1990s, experiments have continued with various reform strategies that encompass differentiation among schools, rather than the standardization of effort and result that is implied by strong, centralized approaches to instructional improvement. Some experiments have featured decentralization of authority and initiative to the school site (e.g., Bryk, Sebring, Kerbow, Rollow, & Easton, 1998; Cross-city Campaign for Urban School Reform, 1995; Henig, 1994), others have championed choice among schools by the “consumers” (parents, students; Hill, 2002; Kirp, 1992), and still others the restructuring of schools into smaller more “personalized” units (Ancess, 2000; Darling-Hammond, 1997; Raywid, 1996). Though there are important differences among them, these approaches to reform place their faith in the devolution of decision-making authority to the school site, sharing decision making among school-level stakeholders, and school personnel’s responsiveness to a local-level “market” for good schooling. Central to all these efforts is the assumption that, given discretion and strong incentives, educators can develop innovative, responsive school programs that serve the needs of urban children well.

The two tendencies—to seek reform through centralized guidance for instructional improvement or through differentiation of effort in smaller, more personalized school sites—are not mutually exclusive, even though they proceed from very different premises about the reform of urban education. If so, it is a distinct possibility that the two strategies might be pursued simultaneously, and as such might represent a potentially productive merger of “top-down” support and “bottom-up” change for which some observers have called (Fullan, 1994, 2001). It is also possible that the simultaneous appearance of the two would signal nothing more than a state of “policy churn” that has come to characterize much of urban school reform (Hess, 1999). From the vantage point of the teacher or school, such a state of affairs would not be new, offering yet another instance of converging reforms with which they must cope (Knapp, Bamburg, Ferguson, & Hill, 1997). As such the result could be interference between multiple improvement programs or competing reform assumptions that researchers have identified in “busy” reform environments (Hatch, 1998; Kimbrough & Hill, 1981).

Our purpose is to explore an instructive case in which two potentially opposed reform theories converge on schools, in order to understand better how productively or unproductively the two theories coexist. The case resides in another of the New York’s Community School Districts, which we will refer to as District M (pseudonym), during the latter 1990s and first two years of the new century. District M pursued a district-guided instructional improvement strategy with many resemblances to the San Diego and District 2 approaches, but also maintained a policy promoting the development and support of small alternative schools of choice at the middle school level, which were reminiscent of the thousand flowers that Alvarado once championed. This paper considers what the convergence of the two policies meant for three middle schools within District M, and with what implications for instructional practice.

The juxtaposition of potentially opposite theories of reform in this district raises questions that we attempt to answer in this paper: Can these two reform theories coexist, or do they get in each other's way? In what ways, if any, do the two complement each other? Separately or together, how do they affect instructional practice and the school-level conditions that support teaching and learning? Finally, can policies and leadership, in this case reflecting two such different reform theories, provide mutually supporting conditions for teaching and learning in urban schools?

In using the term reform theory, we intend to describe theories of action, as do other scholars who find the concept a useful way of capturing reformers’ assumptions about the focus and effects of their actions (e.g., Hatch, 1998; Malen, Croninger, Muncey, & Redmond-Jones, 2002; Weiss, 1995). We suggest that District M operated with two broad sets of principles and propositions, orientations, and assumptions that are embodied in two theories of action that guided the district’s educational renewal efforts. The concept of theory of action is often associated with Argyris and Schon (1982) who argue that while these “theories” may be incomplete, untested, or even contradictory, they constitute a framework that individuals use to guide, interpret, or justify their actions. The term has been extended by others for both program and policy evaluation studies (see Malen et al., 2002; Smylie et al., 2003; Weiss, 1995). In adopting this approach for our analysis, we describe District M’s theories of action based on the rationales articulated by district and school informants and on the internal logics and premises of the policy that we, as analysts, observed and identified. This approach allows us to then discuss how productively the district’s reform theories converged on schools and instructional practice.

Several other observations about the theory of action construct help to frame our analysis. First, these are not “theories” in the usual sense of that term, nor are they necessarily consistent, coherent, or easily testable. Second, theories of action are not always explicit or well articulated—as they are often implicit and knowable largely by inference from action. For this reason, Argyris and Schön (1982) have coined the term “theory-in-use” to highlight the way assumptions about reform might manifest themselves in practice. Third, theories of action generally embrace more than one set of assumptions, notably what Fullan (2001) refers to as a “theory of education” (assumptions about teaching and learning itself) and a “theory of change” (assumptions about the way organizational actions and conditions can change teaching practice and ultimately learning outcomes).

Our goal is not to evaluate these theories in the District M case, either separately or together, but to take advantage of an unusual opportunity to examine how they interacted with one another at the school level. Our thesis is that the two reform theories—district-guided instructional improvement and the development of innovative, small schools of choice—not only coexisted in this instance but, to some extent, complemented each other’s strengths and, likewise, compensated for weaknesses. Even so, it is clear that, when juxtaposed, these two theories of reform can and do create tensions in the system and, thus, potential obstacles for educators’ work. On the whole, however, we argue that this was a constructive tension in this instance, and can be so elsewhere.

We unfold our argument in several stages. To begin with, we examine in more detail what the two reform theories presume about instructional improvement and systemic change, and discuss how each might complement or interfere with one another. Next, we look at how the two reform theories were enacted in the district in question, first, by offering brief portraits of the three schools we studied and showing how they embraced the tenets of the first theory (educational improvement through small schools of choice). We describe how the second theory, embodied in a district-guided instructional improvement strategy, manifested itself, and how instruction reflected the district’s insistence on certain forms of standards-based practice. We explore what enabled the two reform theories to coexist in this case, while noting the difficult-to-resolve tensions that their interaction creates. We conclude by offering observations about the meaning of reform convergence in this case and suggest what may be implied for other attempts to combine strong district leadership with school-level discretion and innovation.


Our analysis concentrates on two sets of ideas about urban school reform that have at their hearts the same or similar goals—that is, to provide equitable and high-quality learning experiences to traditionally underserved urban students—but they go about the task in very different ways. The first theory of reform, emphasizing small schools of choice, seeks to capitalize on energies of teachers, students, and parents who form distinctive learning communities through voluntary association and collaborative work. The second reform theory, featuring centrally-defined learning standards, extensive professional development, and more explicit accountability requirements, places its bets on clear guidance from authoritative sources, along with supports and incentives for improving learning and teaching performance.


The idea of small schools of choice has deep roots in New York City and in the district we studied. With the encouragement of district leaders such as Alvarado (when he was in District 4), innovative school leaders across the city started small schools in the 1970s as alternatives to what they saw as overcrowded, bureaucratized, and failing inner-city schools (Fliegel, 1993; Meier, 1995; Raywid, 1996). As a response to the challenges of urban education, these attempts rested on four interlinked premises, which are most easy to realize in schools operating on a relatively small scale (e.g., fewer than 400 students):

Distinctive school character and mission. Through the teachers’ involvement in creating the school curriculum and in order to attract students who want to be there, the school would develop a distinctive “character” (Hill, Foster, & Gendler, 1990) that reflects a clear and compelling mission. Students who enrolled would be in the school knowing what kind of a school community they were part of.

Personalized relationship between teacher and learner. To engage fully in learning and to be helped to overcome the obstacles to their learning, students must be known and appreciated by adults. This is especially likely to happen in settings that ensure extensive contact between learners and teachers, with whom the learners could form lasting relationships. Small school size enables this personalization, as does a variety of structural changes in the school’s program, schedule, grouping, and staffing (Darling-Hammond, 1997).

Teacher collaboration in, and ownership of, school design and curriculum. Teachers would engage learners more deeply, it was assumed, if they participated in creating the curriculum—and indeed, the design of the school as a whole—“through continuing dialogue, face to face, over and over”(Meier, 1995, p. 109). This way, this approach to urban schooling sought to tap into teachers’ creative energies and their sense of ownership over the work of educating an often-challenging school clientele.

Parental engagement in public school choice. By reaching out to parents, who often feel alienated from city schools, and attempting to draw them into the school community in various ways, educators would be most likely to engage their support for their children’s education. In addition, as alternatives to conventionally organized schools, the school would seek to attract a voluntary clientele, and would only be able to do so if the school offered what parents and students valued; conversely, if they failed to attract “customers,” the school would be, appropriately, “out of business” (Lieberman & Sostre, 1998).

These ideas, by themselves, do not necessarily constitute a theory of action for educators working at the district level, whose reform energies are directed at an entire system of schools. Rather, they were conceived largely by visionary individuals working within particular schools, who sought—or fought for—permission and support from the relevant district authorities in the public school system (e.g., Meier, 1995). Nonetheless, certain district-level leaders, Alvarado among them, did more than give permission—they actively promoted the development of such schools, in part, because they believed these schools would serve urban school children better than their traditional counterparts. In this sense, they made and enacted assumptions about instruction itself and how it might be improved, and their assumptions fall within our boundary definition of “theory of action.” To be sure, district leaders probably had other motivations for pushing this solution to urban education, among them, the likelihood that small, distinctive alternatives to traditional public schooling would be more likely to retain middle-class children of all races within the public system, thereby stemming a pattern of middle-class flight. Such motivations signal a larger political theory of action at work that goes beyond immediate questions of instructional improvement.

The district strategy emphasizing small schools of choice says little about teaching and learning per se other than to assume that engaged teachers who collaborate in the development of curriculum in relation to a distinctive school mission will be likely to teach well.2 Similarly, students who aren’t allowed to “slip through the cracks” and who develop relationships with the adults in the building are likely to learn more than they otherwise might. The research linking small school size and improvements in student performance is mixed; however, some studies suggest that small school programs produce equal or superior achievement for students in general, with the most positive gains for ethnic minority students and students of low socioeconomic status (e.g., Cotton, 1996; Darling-Hammond, 1997; Raywid, 1996).

Alongside the promotion of small schools of choice, a second and very different theory of reform evolved in this district and elsewhere during the 1990s, which addressed questions of learning and teaching from the district level more explicitly. At the heart of the second reform theory is active instructional leadership by district administrators and staff (Murphy & Hallinger, 1986, 1988; Peterson, 1999; Resnick & Glennan, 2002), wedded to notions of “standards-based reform” (Fuhrman, 2001). Though initially championed in state-level reform, standards-based reform strategies have been enacted in urban districts across the United States (e.g., Christman, 2002; David & Shields, 2001; Massell & Goertz, 2002; Snipes, Doolittle, & Herlihy, 2002; Togneri & Anderson, 2003). While there are many variants, the following features characterize currently popular versions of this strategy:

System-wide commitment to high learning standards for all students. The linchpin of this “theory” is consensus across the district on the importance of bringing all children to a set of demanding learning standards (e.g., David & Shields). These standards make clear the goals for instruction and (through assessments linked to the standards) the desired levels of learning in relation to each.

Priority on foundational, “gateway” subject areas. By concentrating attention on subjects on which future learning in many subjects depends—especially language arts and, to a lesser extent, mathematics—educators across the district, it is assumed, will be able to maximize their effort at improving essential learning experiences for students.

Investment in teachers’ professional development. To get teachers up to speed in the priority subject areas, the “theory” holds, extensive and varied opportunities for professional learning are necessary.

Attention to the professional learning of school leaders. The “theory” further asserts that strong, school-level instructional leadership is essential to realizing and sustaining instructional improvements (Fink & Resnick, 1998). The district is assumed to be in a good position to structure and provide opportunities for school leaders to sharpen their instructional leadership skills.

An emphasis on professional accountability. Finally, the emphasis on standards, professional development, and instructional leadership communicates a sense of responsibility for improving professional work. District leaders and staff are seen as an essential part of the accountability structure, and all members of the system (school and district) commit to holding each other accountable for their respective roles in helping students learn.

This reform theory rests on a logic that posits a greater role for the district in directing the activities of school-based educators and providing them with support for their work. In short, it presumes that school staffs want and need a clear sense of direction and incentives (other than those that originate within the school or in its relationship to a client population).

Potential Interactions between the Two “Theories”

The two “theories” differ in a number of respects. A glance at the following dimensions related to instruction and its improvement reveal the potential for conflict, as well as possible areas of complementarity.

Focus on what to teach and how. In the first “theory” the district leaves choices about the content of the curriculum and the means for engaging learners in that content up to the schools, while in the second, the district actively promotes particular content and pedagogical principles.

Standardization of curriculum. Related to the preceding dimension, the first “theory” does not place high value on the standardization of curriculum, while the second does, by pushing for performance to a predefined common (and generally high) standard, while attempting to equalize the opportunity for all students to reach that standard.3

Focus on the quality of relationship between teachers and students and among teaching staff. The first “theory” concentrates on the quality of the relationships between teacher and student, on the assumption that sustained, personalized contact between them will provide the support students need to succeed; the second “theory” is relatively silent about this, focusing instead on teachers’ content knowledge, approach to pedagogy, etc. Strong collegial relationships are also a primary target of effort under the first “theory,” but more of a by-product under the second (e.g., by virtue of participation in professional development that emphasizes collaborative learning).

Locus of initiative for improving instruction. The first “theory” locates primary initiative for undertaking improvement efforts within the school, by contrast with the second, which assigns a central role to district-level actors, who motivate, guide, and “teach,” in pursuit of instructional improvement goals.

Locus and form of accountability for instructional improvement. Under the first reform “theory”, educators within the school are responsible to each other, as co-developers of the school’s program, and ultimately to parents, who choose to place students in the school or elsewhere; by contrast, under the second reform theory, the district (and state) assume a greater role in holding school staff accountable for their performance.

Differences on these dimensions are likely to show up both in district actions and in the schools’ responses to those actions. As an earlier line of research has established, multiple reform programs and policies from the federal level can interfere with one another at the local level (Kimbrough & Hill, 1981). We also know from studies of ambitious state and local reforms implemented in the past decade that teachers can find the multiple agendas frustrating and confusing (Hatch, 1998, 2002). Recent work on the “multiple accountabilities” facing local educators reinforces the point (Firestone & Shipps, 2003), as does research documenting the likely failure of multiple concurrent reforms, including high-stakes testing, comprehensive school designs, and other curricular initiatives in a high-poverty district (Berends, Chun, Schuyler, Stockly, & Briggs, 2002). But research also hints at the possibility that, as enacted in and around schools, the two sets of reform “theories” simply co-exist or even complement each other. Synergistic effects of multiple government programs have been documented in districts and schools (Knapp, Stearns, Turnbull, David, & Peterson, 1990), and, in principle, the school-level activities set in motion by the two reform “theories” could be compatible with one another. Seen this way, the first set of reform ideas might create a “shell” in which standards-based practices promoted by the second reform theory might develop.

Whatever the effects, it is clear that teachers and others in schools are faced with the prospect of sorting out “converging reforms” (Knapp, Bamburg, Ferguson, & Hill, 1998). How the two reforms come together in the minds and practice of school staff and leaders are likely to depend on the school’s staff, leadership capacity, and leadership (Berends, Bodilly, & Kirby, 2002); the school’s “institutional readiness for renewal” (Portin, Beck, Knapp, & Murphy, 2003); and the school’s organization for teacher learning and other aspects of the joint daily work of the school (Darling-Hammond & Snyder, 2003; Little, 1999). Individual teachers and other school staff are also in a position to integrate, accept, reinterpret, or reject reform messages from the district and elsewhere (Coburn, 2005; Jennings, 1996). The net result is the working theories-in-action embodied within the efforts of individuals and groups within the school.

The role assumed by the district in this set of transactions is also likely to be crucial. Indeed, in the overall balance of top-down and bottom-up supports for instructional improvement—the manner in which district players engage the school in matters of teaching and learning will have much to do with how any reform initiatives are received (McLaughlin & Talbert, 2002; Resnick & Glennan, 2002). Here, the balance struck by the district between flexibility and directive guidance is likely to influence the enactment of multiple reform “theories,” and their compatibility, once enacted. Inflexibly or mechanically applied, a district-guided instructional improvement strategy pressing for adherence to high learning standards can surely negate the benefits of differentiated small schools of choice. But carried out with close attention to the strengths and weaknesses of each school, and to the possibility that the schools themselves may be able to realize the district's goals in heretofore unconsidered ways, such an improvement strategy may bear fruit.


The study reported on here was part of the broader Study of Policy Environments and the Quality of Teaching (referred to as the “Core Study”), an ongoing research project conducted by the Center for the Study of Teaching and Policy (CTP) between 1998 and 2002.4 The questions explored in CTP’s Core Study focused on (1) the nature of the “teaching policy” strategy in the Core Study states and districts; (2) the formation and maintenance of that teaching policy strategy; (3) the interaction of the teaching policy strategy with other policies and contextual conditions; and (4) the nature and role of resources in the teacher policy strategy.

This paper draws together data collected in a New York City community school district (called District M) and in three of its middle schools. In order to study the complexities of the district and school teaching policies within the broader context of New York State and New York City reform activities, we utilized a nested, multiple-case design (Yin, 2003). Case study methods are especially appropriate for intensive, in-depth examination of complex phenomena, such as the connections between educational policies enacted across multiple arenas and the instructional practices of individual schools and teachers (LeCompte & Preissle, 1993; Merriam, 2001). Below we describe the rationale that we used to sample the middle schools, our data collection activities, and the methods that we employed for data analysis.

At the time of our study, District M served approximately 15,000 students in 36 school programs covering kindergarten through eighth grade. The student body in District M was diverse, both ethnically and economically. Approximately 80% of the children were students of color, mainly African American and Latino, and more than 70% of the students were eligible for Title 1 services. Leadership in the district had been largely stable over the 1990s. The Superintendent at the time of our study had been the Director of Staff Development from the late 1980s until assuming the top leadership position in 1997.

Following the sampling procedures used across all Core Study districts, we selected schools in District M that (a) were viewed by district personnel as schools that were serving their students well, and (b) were serving similar, high-poverty student populations with comparable expenditures per pupil. Anticipating that school structures and the configuration of resources (e.g., time and staff) might affect the quality of teaching and learning, we included some schools that were significantly restructured (e.g., schools in which time and personnel resources were used flexibly to maximize collaborative time between adults as well as time between students and teachers) and others that had relatively traditional structural arrangements (e.g., schools with 25–30 students per classroom and teachers who taught 5–6 periods per day with little scheduled time for adult collaboration).The three middle schools included in the study served primarily African American and Latino student populations. They represented a range of structural characteristics and they were recommended by district leadership after we described our selection criteria. We interviewed principals and several teachers at approximately six middle schools about their willingness to participate in the study before making final selection decisions. Over the three years of this study, we visited each school three times per year for three to four school days each visit. During the site visits we interviewed the principal (two to three interviews per year), a sub-sample of classroom teachers (once or twice a year), and other specialized school personnel (for a total of between 14 and 23 interviews per school). We collected field notes in classrooms (between 14 and 33 classroom observations per school) and other school settings, and we collected relevant documents including items such as Comprehensive School Plans, test score data, master schedules, and staff rosters, as well as curriculum artifacts and student work. Simultaneously, time was spent during each of our field visits interviewing district staff, observing their work, and examining a range of documentary evidence assembled by the central office.

This data collection focused on a number of topics—at the school level, predominant patterns of instructional practice, students’ access to content and responses to instruction, as well as teachers’ engagement in professional development and instructional planning, the allocation and use of resources in the school, and school leadership and organizational issues. Throughout our school-level data collection, we watched and listened for evidence of the school’s response to anything in the “teaching policy” environment, especially, district-level actions and conditions that might affect teaching and learning. We also probed the thinking of school administrators and teachers to get at their understandings of the policies and conditions that guided their practice. To explore patterns of instruction and student response to it, we visited a variety of classrooms and returned repeatedly to a small number so that we could observe how instructional practice evolved over time. At the district level, we concentrated on learning about the reform strategy, the roles and activities of central office personnel (especially those involved in direct work with the schools), the allocation of resources to support reform, and other salient leadership issues confronting the district. Here, we watched and listened carefully for evidence of an active connection with people in the schools—through communication activities, visits to the schools, and various forms of professional development—such that reform messages and intentions might be communicated.

Data analysis was ongoing over the four years that the study was conducted. Raw data were reviewed following each site visit and site teams made adjustments in interview and observation protocols based upon data gaps in the developing school and district cases (Miles & Huberman, 1994). Individual case summaries were developed at the end of the first year of data collection for each school case and for the district as a whole, and elaborated at the end of each subsequent years. The summaries were descriptive in nature and were developed using a common framework based upon constructs that were drawn from the overall CTP Core Study research questions.5 This paper is based on a cross-case analyses of data collected in the three middle schools, supplemented by district-level data.6

In order to check our understandings of the teaching policy environment in District M and the middle schools with the experiences of our interviewees and others who worked in the district and the schools, we presented our research findings to them periodically during the three years of data collection. They provided us with feedback regarding our findings and that feedback was incorporated into our final analyses.


The two reform “theories” discussed above were evident, first of all, in the thinking and actions of district and school leaders in District M. By the time our study began, the two “theories” had been enacted sequentially and gradually in such a way that each one was realized, to varying degrees, in the structures and daily practice within the schools. The net effect was, at least, a relative absence of resistance to the enactment of either theory. At most, as we demonstrate later, the combination of the two may have been partially responsible for the considerable increase in student performance in this district over time. The brief chronology of the implementation of the two theories, detailed below, serves to introduce the three schools and offer a first layer of evidence that, in each one, the two were at least coexisting without substantial interference, if not more actively complementing one another.

The First Reform Theory in Action: Establishing and Sustaining Small Schools of Choice

In the District M case, district reform efforts based on the first reform theory preceded initiatives reflecting the second. Since the early 1990s, District M had encouraged and supported the development of small, alternative schools within its boundaries, in part to draw middle-class and white families back into their public schools. Since 1991, small schools had been sanctioned system-wide through an official choice policy that operated across all middle schools in the district. In the 5th grade, all district students applied to middle school indicating ranked preference for up to four schools. During the second year of our study, district leaders reported that approximately 80 percent of the students were placed in either their first or second choice school. By 1992, 24 schools of choice had been established (Fliegel, 1993), and by 1998 when we began our study, all middle schools in District M were choice schools.

It would be a mistake to attribute the existence and nature of these schools to district policies and leadership alone. Each middle school we studied had been guided by structures and traditions that have become well established within New York's small schools movement (Fliegel, 1993; Raywid, 1996), and in one instance, an early version of the school’s structure was in place prior the enactment of district policy in 1991. That said, there was clearly a strong belief among the District M’s top leaders that these kinds of schools work well for a wide range of urban children and should be encouraged. We heard repeatedly about the benefits of the schooling such schools offered, and during the three years of our study, the District took steps to create an additional small school of choice, located in a part of the district that held few such schools. The District leaders praised the manner in which these schools engaged parents, opened up options for the full range of students, especially those in the more impoverished part of the district, and provided a means for keeping students from “falling through the cracks.”

Under their watch, district leaders thus endorsed, promoted, and supported a distinctive set of small middle schools that were enacting, at the school level, many if not all of the basic tenets of the first reform theory. The three schools we studied reflect a range of such schools in the district.7

The James B. Conant Education Complex (pseudonym) was simultaneously a single school and collection of four semi-autonomous mini-schools under the same roof. Each mini-school acted as a “school of choice,” with its own name, thematically organized academic program, and distinctive student population. The building’s mini-school organization, well-established over a decade and a half, was responsible for preserving the separate identity and largely separate operation of the small school programs. The four mini-schools were housed in a large and significantly overcrowded city school building that was also shared by two other, completely autonomous schools. At the time of our study, there were a total of 1,300 students in the building. In keeping with District M’s choice policy, the students at James B. Conant came from all over the district. The programs drew differentially from more and less affluent sections of the district. As a whole, the students were relatively poor—about three-quarters of the students received free- or reduced-price lunch—and were predominately students of color (about 86% African American and Latino; 10% white).

The Conant Complex could also be understood as a sharply tracked group of mini-schools, with programs and curricula representing differentiated and stratified tracks. To best reflect the range of what Conant offered, we focused our study on its two largest mini-schools—the Alvin Ailey School and the Discovery Institute (pseudonyms), each of which served about 300 students. Alvin Ailey’s students were mostly African American and/or Latino students who came disproportionately from the most impoverished end of the district. In contrast, Discovery Institute drew the majority of its students from the portions of the district that were home to a somewhat more affluent population with a larger proportion of White students. Alvin Ailey offered performing arts courses and other arts programs that drew some talented students, but, overall, it was a rather typical example of a low-performing urban middle school (as test scores suggest) especially in the seventh and eighth grade where literacy and mathematics scores were generally two to three times lower than the corresponding scores in the other mini-schools (see table in Appendix A). The Discovery Institute, on the other hand, had a long tradition as a small school of choice that offered strong science education. Our interviews with school and district informants suggested that Discovery was generally considered a good option for students who showed strong abilities in math and science during their elementary school years.

Cisneros Middle School (CMS) (pseudonym), the second school in our study, contrasted sharply with both Alvin Ailey and Discovery Institute (and indeed the whole of the Conant Complex). Created 10 years earlier under the direction of a strong school leader with a vision of providing academic rigor for Latino students, the school featured an atmosphere that respected the students’ home culture and offered a nurturing relationship to the students, combining instruction in both Spanish and English in a “dual language” model. Housed on the third floor of a large school building located in a desirable section of the district, space was a perennial problem for CMS—teachers were often without their own classrooms and the small office served as office, faculty room, planning room, copy room, and reception area. The school served approximately 200 students, nearly all Latino children who typically heard Spanish spoken at home, although the majority of students were English language dominant. All of the students qualified for free-or-reduced-price lunch.

At Cisneros Middle School an emphasis on personal relationships was readily apparent in the students’ casual and friendly demeanor and teachers reported that they valued close relationships with their students. Special organizational attention was paid to classes and activities that foster the family-like tone of the school (such as advisories, Friday Clubs, and whole-school community meetings held weekly). Simultaneously, the teachers held high expectations for their students’ academic success (especially in the content areas of reading, writing, and mathematics) and according to students they were strict taskmasters. Overall, Cisneros was considered one of the better choices in the district—particularly for students who did not expect to enter the accelerated programs that some schools in the district offered.

Parkside Middle School (pseudonym) displayed yet another character, emphasizing the restructuring of time and other resources to support both smaller class sizes and increased teacher collaboration. While turnover among the young staff created some gaps in expertise, the school was characterized by a strong professional culture and high standards for practice. Like Cisneros, Parkside was nestled on the third floor of a building that also housed an elementary school. It was located at a midpoint in the district between the low-income neighborhoods to the north and the increasingly gentrified sections to the south. Parkside served about 200 students, half Latino and two-fifths African American students; approximately 80% of the students received free- or reduced-price lunch. Informants reported that the students came from primarily stable housing projects and working-class neighborhoods.

The principal was brought into the school in 1995 to revitalize a “chaotic” program that was then a middle school extension of the elementary school. She and a core group of four teachers spent the next three to four years restructuring the program with a mission of serving students of color in a “progressive” humanities-based curriculum.8 In a manner similar to Cisneros, the staff at Parkside placed priority on close relationships with their students. This small school was considered generally successful with a typically underserved population of students.

Enacting the first reform “theory” in the schools. In short, the staffs of these three small

schools had created programs that displayed the central features of the first reform “theory.” They took seriously and had contributed to the development of schools with distinctive character and missions. In most cases, they sought to develop strong personal relations with each other and with their students; and to varying degrees they had worked together as staffs to develop school-specific curriculum or adaptations of curricula from elsewhere. And, they were attracting “customers.” Though some of these schools were more sought-after than others, they were all populated to a great extent with students whose parents had selected the school as a first or second choice out of four ranked preferences. While not all teachers in these schools would have described what they were doing in the terms we used earlier to characterize this reform theory, they were clearly enacting the theory in their daily work. The depth and fidelity of that enactment, however, varied across the schools, a matter which we explore further later in the paper.

Layering on a Second Reform “Theory”: District-guided Instructional Improvement

Gradually over time, alongside its continued commitment to small schools of choice, the leaders of District M evolved an approach to instructional improvement that featured a prominent role for central office staff and a set of reform ideas that they championed across all schools. This evolution is hard to date, and although there were elements of it emerging in the early half of the 1990s, the approach took a more formal and articulated form with the installation a new Superintendent in 1997 (who had been an Assistant Superintendent in District M for the preceding six years). The district elaborated its standards-based, instructional improvement strategy by drawing on ideas about learning, instructional improvement, and system change from various sources, including NYC Community School District 2 and the Institute for Learning at the University of Pittsburgh (Elmore & Burney, 1999; Resnick & Glennan, 2002; Stein & D’Amico, 2002). Drawing on these sources and their own collective experience, the district fashioned an approach to the challenges of urban education that emphasized system-wide commitment to high learning standards for all students, especially in literacy and mathematics.

Referencing New York City content standards primarily, the district placed emphasis on a set of learning principles that assert all learners’ capacities to meet these standards, and on practices that would lead students to produce “standard-bearing” work. (The meaning of “standard-bearing” work took time to evolve, but eventually the district published a compendium that offered teachers extensive examples from the district’s schools of what it considered to be “up to standard”.) The district’s leaders had a particular form of literacy and mathematics curriculum in mind: it promoted “balanced literacy” and a conceptually-oriented, experiential form of mathematics teaching (exemplified by the text series Connected Math favored by the district at the middle-school level). The priority between the two, however, was clear: literacy came first.

To help establish standards-based approaches to literacy and mathematics teaching, the district invested heavily in professional development of several kinds. It provided resources to the schools (e.g., school-based coaches), accompanied by periodic district-wide in-service activities and support for participation in teacher education courses in nearby training institutions. At the same time, the district went to considerable lengths to work with principals (and often assistant principals) through monthly all-day meetings devoted largely to instructional issues. The Assistant Superintendent for Curriculum, Instruction, & Professional Development also convened an additional study-group for leaders of lower performing schools. District leaders explicitly tied the professional development they supported to an accountability system that emphasized professional responsibility for student learning aligned with city and state standards. To set the tone, district leaders conducted highly visible “walk-through” visits to each school and its classrooms once a year, followed by extensive and detailed written feedback and often accompanied by other forms of coaching. Generally struggling schools were visited more frequently by the Superintendent or Assistant Superintendent. A large cadre of district professional developers (approximately 18 for literacy and mathematics alone in the second year of our study) worked in schools and classrooms to reinforce the messages that district leaders were trying to transmit concerning the improvement of instruction.

Both in conception and in practice, the district leadership pursued a consistent path in trying to stimulate and guide instructional improvements in the schools. The main elements of the second theory of action were not only an explicit topic of conversation among district leaders, but they also dominated their interactions with people in the schools. In this sense, the reform “theory” was fully enacted by the district.

As with the first reform “theory,” there was considerable evidence that the second reform theory was extensively implemented in the three schools. Put another way, the district’s persistent messages regarding teaching and learning, in relation to the district’s conception of “standard-bearing” work, played out in instructional practice and in the daily affairs of the schools. It could have been otherwise: we might have found considerable grumbling about the district’s efforts to promote balanced literacy, along with a refrain of “this too shall pass.” Yet, we heard a whole lot less than we expected, and we were around long enough and in enough situations to have heard such talk if it existed. Instead, here is what we found.

Response in literacy instruction. Across all the language arts and literacy classes that we visited in the schools, we observed instruction that contained elements of balanced literacy (such as students reading independently, participating in guided reading sessions, and responding to literature through group discussions and writing assignments). We saw evidence of classroom “libraries” in most of the rooms we visited, and schools had generally organized their schedules to include up to two hours of literacy-related instruction per day. School leaders and teachers reported that these practices were a response to the district’s explicit and oft-stated desire to see evidence of these, and other, elements of balanced literacy in all schools and classrooms. Even at Alvin Ailey, where the tendency toward isolated practice and idiosyncratic attention to reform attempts was prevalent, messages from district professional developers were getting through to teachers, even if they were received in more categorical terms than intended. For example, according to one English teacher, choral and round robin reading were not favored, class sets of books were “verboten,” and classroom libraries were “in.”

Response in mathematics instruction. Across the schools, there was also evidence of attempts to respond to the district’s push for improved mathematics instruction in a more constructivist vein, though the response was more tentative and mixed, perhaps reflecting the district’s own allocation of resources (e.g., fewer mathematics coaches) and slower timetable for addressing mathematics (which was not a priority until midway through our study).9 Furthermore, in almost all cases, changing approaches to mathematics represented a more fundamental shift than what balanced literacy implied, and there was initially much reluctance to make the change. Yet Cisneros staff embraced the school-based mathematics staff developer assigned to them immediately and eagerly worked with her in groups and individually to build their capacity to undertake the new approach to mathematics. At Parkside, the school initially paid less attention to mathematics teaching until mediocre results (made more visible by the stricter accountability requirements) prompted a rethinking of this omission; soon thereafter the school started using the coaching resources provided by the district for this purpose. Only Conant dragged its feet on accepting the changes, as a matter of school-wide (or mini-school) policy. Nonetheless, some individuals in the two mini-schools worked with the district professional development staff and began experimenting with the Connected Math curricular materials. The net effect was a mixed response, by the time our data collection closed: while teachers in various classrooms were using constructivist techniques, we also observed test-oriented and traditional mathematics instruction using curricula such as Sequential One, viewed as a curriculum that more directly prepared students, especially eighth graders, for the upcoming New York State High School Regents’ exams.


Our data permits no direct way of assessing the effects of investment in these reform theories on student learning. However, some indirect evidence is suggestive. Across the decade of the 1990s, and continuing during the period of three years in which we collected data, District M’s performance relative to other community school districts improved dramatically. Across these ten years, the district moved from a ranking of 31st out of 32 community school districts in reading achievement to 16th place by 1997 and 15th place in 1998. In mathematics achievement, the district moved from a ranking of 29th (of 32 districts) to 15th place by 1998. How much of these results can be attributed to the combination of small schools promotion and district-guided instructional improvement is hard to say, independent of other forces like demographic shifts, which may have accounted for some of the changes. District leaders were convinced that their efforts were part of the explanation and their account is plausible. Their consistent and strong centralized messages about instructional practices, backed up with specific ideas about improvement and lots of supporting resources, apparently found willing ears in a series of schools that had invested heavily in their own distinctive missions, as well as consistent attention to school-level professional accountability. At a minimum, there was no sign of widespread resistance to the district’s instructional improvement efforts, and there was good evidence, to which we now turn, that embracing the first reform vision enabled the schools to respond productively to the district’s efforts that reflected the second reform “theory.”


The apparent coexistence of the two reform theories in the three middle schools we studied begs a deeper question: what enabled this to be the case? Why didn’t the reform theories set up competing pressures, such that efforts to enact one got in the way of the other? Our analysis suggests two basic answers:


In the schools, steps taken by school leaders and staff to realize the first reform theory enabled schools to respond productively to, rather than resist, district initiatives based on the second theory. In particular, the schools’ responsiveness to the district could be traced to the extent to which they had made operational the small school goal of support for professional learning and focus on knowing students well.


On the district’s part, a sophisticated vision of teaching and learning (for both students and adults), combined with a flexible approach to implementing the second reform theory, reduced the likelihood of conflicting reform messages. At the same time, the district’s posture enhanced the possibilities for synergies across the two sets of reform ideas.

We elaborate below, with examples from each school.

How the First Reform Theory Enabled the Schools’ Response to the Second

The fact that the three middle schools differed in how fully they implemented either reform theory made it possible for us to explore how the degree of implementation might be related to the coexistence and compatibility of the two sets of reform ideas. With respect to the first reform theory, in particular, the three schools can be arrayed along a continuum from partial realization of the small schools reform theory (by the two Conant mini-schools) to fuller realization (by Cisneros and Parkside).

At first glance, all of the schools reflected in some degree the logic of the small schools of choice reform theory. All were deliberately organized to be relatively small learning communities aimed at fostering relationships among students and teachers, and all advertised themselves as such to “consumers” (parents), who responded to this feature. Reflecting the research on small schools (Darling-Hammond, Ancess, & Falk, 1995; Lee, Bryk, & Smith, 1993; Lee & Smith, 1995; McDonald, 1996; Newmann & Associates, 1996) and the middle school movement (Atwell, 1998), these schools took as a pre-condition to successful operation that students have the potential to be known by their teachers and that in small schools a clear sense of purpose can develop. In other words, the schools operated from the assumption that size matters. In addition, all of the middle schools had made some structural changes (beyond the given of small size) aimed at building stronger student-teacher relationships, influencing patterns of instruction, and improving student learning outcomes.

However, the schools differed in the degree to which these structural changes were made with coherence and clarity of purpose—presumed attributes of small schools. For example, at both Cisneros and Parkside, decisions about scheduling, grouping, class size, and teacher assignment were made with the schools’ overriding sense of mission in mind, one that prominently featured the desire to personalize instruction.

At Cisneros, the organizational groupings of students were constantly evolving—sometimes students were grouped by grade and/or by ability, sometimes completely heterogeneously—but decisions were guided by the drive to build strong relationships among teachers and students. Classes tended to be small for math and literature (about 14-18 students), and teachers were also responsible for a 45-minute weekly advisory class for 12 to 15 students. The focus of the advisory class, in keeping with the school mission, was on the affective needs of students, parent contact, and socialization and academic support. Daily homeroom periods supplemented this structure. Classes at Cisneros were generally 45 minutes long, although literature and math classes lasted 50–60 minutes. There were weekly whole-school assemblies, which were likewise designed to reinforce and build connections among students and teachers.

At Parkside, students were organized along grade-level teacher teams for most subjects, and class periods extended to 90 minutes although there was flexibility based on curricular priorities. By comparison with a traditional middle school structure, teachers saw fewer students per day (70 to 75 on average, in class configurations with 15 to 20 students per teacher) and they changed classes fewer times during the day. Within teams, teachers were given wide latitude to develop grade-level schedules. Team-taught classes gave students a home base, which functioned like an advisory (though they did change from year to year). Some classes were heterogeneously grouped (such as social studies, arts, and science) and some were ability-based (there were three levels of math and literacy classes). Support was provided to students informally through close relationships with teachers and through “extra” classes for strugglers and for students who were accelerated, offered both during and after school (lower ability classes were generally smaller than were those for middle and high ability achievers).


In contrast to the purposeful changes made at these two schools, the corresponding decisions about grouping, scheduling, class size, and teacher assignment at Conant lacked consistency and often appeared to be management-oriented. Across the mini-schools at Conant, students were organized by grade levels and cycled through teachers and subjects in traditional 42-minute periods. In a typical day, teachers taught five periods of 30 students each, for a daily total of approximately 150 students. There were some exceptions, notably at Alvin Ailey, where there was some experimentation with self-contained 6th grade classes. This arrangement, however, seemed motivated as much by the desire to better manage rowdy incoming sixth graders as to establish a more personalized relationship between teachers and learners. Other attempts at this school to modify the traditional schedule (e.g., a double-period humanities block) were truncated by the tendency to use time during the extended class periods as if the original schedule were still in place, for example, by changing activities every 42 minutes, rather than taking full advantage of a double-period block of time.

A parallel set of differences in the schools could be seen in the way staff time was organized, and in the extent to which the staff collaborated with one another in developing the school program—another key feature of the small schools reform theory. Once again, Parkside and Cisneros were notable for the amount and quality of time the staff spent together and the results of their collective effort in creating curriculum. In Discovery and Ailey, though weekly times were set aside for staff planning, we observed time being used to troubleshoot problem students or manage joint logistics (e.g., where to take the students for a fieldtrip) rather than to engage in substantive issues related to what was being taught and what was being learned.

These differences in the realization of the first reform theory were related to the way the school staffs responded to district actions guided by the second reform theory. In short, the more fully the first reform theory was realized, the more the school took advantage of district resources, requirements, accountability system, and interventions reflecting the second reform theory.

Parkside and Cisneros illustrate how this pattern appeared within schools that more fully embodied the principles of small schools of choice. In the case of Parkside—a school which was built around a homegrown social studies course called Connections designed to integrate the humanities (history and nonfiction literature) with political and social sciences—staff might have found the district’s insistence on a balanced literacy approach to be intrusive and distracting. Instead, in part as a response to the district’s literacy initiative, the school put high priority on developing a Readers and Writers course, taught in small, extended period grade-level classes (about 15 students to one teacher). The curriculum was geared especially for the middle school learner and was developed through the staff’s shared understanding of In the Middle by Nancy Atwell (1998). The emphasis in these classes tended to be on student interpretations of literature and on the development of writing skills. For example, in one Readers and Writers class of eighth graders, the students read a memoir chosen from several selections previewed by the teacher. Over a few weeks, students developed a list of “criteria” for writing memoirs based on their reading and then wrote a series of their own memoirs. Selections were published in a class book made up of the students’ best work. In general, the pedagogies of choice at Parkside tended to favor student responsibility for knowledge-building or problem-solving activities. This work went well beyond what the district was asking for (e.g., shared and guided reading, writing), but met the spirit of the external push for improved literacy while providing a compatible support for the Connections program.

Cisneros School’s staff responded differently to the district’s initiatives, but also found ways to make the district’s literacy and mathematics improvement initiative “work” for a school that was committed to celebrating the lived experience of the Latino student body and pride in the Latino culture. A teacher at Cisneros said, “What is a focus (here) are issues involving Latinos. You can see that more within the curriculum: books written by Latinos, the Latino experience, being American and Latino at the same time; bilingualism.” The development of students’ Spanish language abilities was therefore an explicit goal, alongside instruction for literacy and other subjects in English. Much of the instruction at CMS was focused on “rigorous academic training” and preparation for the state and city tests that were the markers for academic success—and given great weight by the district’s instructional improvement efforts. To that end, the staff prioritized access to and success in traditional academic language and texts. In this environment, the district’s new push for a more constructivist approach to mathematics instruction met with enthusiastic attempts by school staff to make sure students “got the new math,” even though most of the teachers at this school had little or no background in anything but traditional arithmetic teaching. A district math staff developer described her reception at this school:

You go to [Cisneros], you hear very little conversation about the kids being educable, or what the system does…I go [there], and I am pursued from the moment I walk in the door to the moment I leave, “Have you got ten minutes? Can you look at this?” Or: “ I am working on this—can you look at it? Can you get into my class today?” Or: “what do you think of this? What do you think of that?” Or: Can you watch this kid? Can you talk to this kid? I had a conversation with the parent…”


On the other hand, in the absence of a strong sense of purpose and curricular coherence, the teachers in the Conant mini-schools tended to fall back on their individual interpretations of the materials they had available. Although there were discernable differences at the mini-school level between Discovery Institute and Alvin Ailey, overall teaching practice across the two programs tended to be teacher-driven and text-reliant, and highly individualized by a particular teacher’s strengths or goals.10 Perhaps because several teachers in these two mini-schools struggled to “manage” their students, teacher-directed instruction was perceived as a sign of a successful teacher. One teacher did say that she thought good instruction occurred when students were “engaged”—that is, focused, participating, and interested—during a lesson. However, the same teacher commented, “There is no school-wide vision for pedagogy in the building.” There were clear steps in that direction such as attempts by the Principal, district staff developer in math, and mini-school leaders in the second year of our study, to meet weekly for coordination of their professional development efforts for the first part of the year, but beyond that the effort failed to gel as a community of educators interested in instilling a richer repertoire of approaches to mathematics teaching.

The differing realizations of the first reform theory in these schools, and hence their capacity to take advantage of district’s directed instructional improvement efforts, are traceable to the nature of school organization and leadership, and to the professional culture that came about as a result. For example, at Conant Educational Complex, leadership was fractured and responsibilities were dispersed among a mix of personnel, in part a function of being four mini-schools under one roof. An assistant principal who reported to the building principal led each Conant mini-school. Beyond the general thematic focus of each school (performing arts or science education), at neither of the two mini-schools did teachers report a vision for their collective teaching practice, an image of a school-wide philosophy, or adherence to a type of pedagogy. At Alvin Ailey, assistant principals (more than one over the three years of our study) had friendly and supportive relationships with their staff, but they played little role in conceptualizing or developing program-wide curricula or pedagogy. Discovery Institute had a more focused and purposeful program that had achieved significantly more success on standardized measures of achievement (in part attributable to a more academically selective student population). Nonetheless, as in Alvin Ailey, the assistant principals’ leadership was easily preoccupied with management and administrative concerns, rather than on setting school-wide goals or engaging staff in collaborative work.

Given the fractured leadership structure and the lack of school-wide mission and focus, the conception of professional learning at the Conant Complex was more limited than it needed to be: teachers were considered either experienced and capable, or not—and there were few opportunities for veteran teachers to collaborate or continue developing their skills. This view was in direct contrast to the district’s view of the teacher—indeed, of all educators—as evolving and continual learners. For example, the teachers at Conant generally taught five periods a day; and had two periods of preparation time. But, none of the teachers interviewed reported that they used common prep time (or common lunch) to plan collaboratively. The individualistic culture at Conant (especially in evidence at Alvin Ailey) and the lack of support structures to nurture collaboration resulted in a relatively stagnant curriculum and pedagogy; this fact offered one plausible explanation for the generally lower level of performance in the Conant small school programs (here, Discovery Institute was somewhat of an exception, but its higher performance levels probably reflected differences in its student population as much as the power of its instruction).

Contrast that scenario with the strong and stable leadership story at Parkside. There the faculty, an in-house professional developer, and the principal were focused on a strongly articulated and widely held mission: to provide a progressive, humanities-based, project-oriented education that met the needs of their students. To ensure that professionalism and strong norms for collaborative work and professional learning were realized, the principal organized the school such that the faculty had time each day to meet in grade-level teams and/or as subject teams. Teams of three to four teachers per grade level met during their prep periods, during their common lunch periods, and often after school hours. The teams were responsible for developing curriculum and instructional strategies for the core humanities course as well as for managing the overall educational program for their grade level.

Staff meetings at Parkside were largely devoted to sharing instructional strategies (teachers shared their work, for example, on the development and use of rubrics or a literacy unit on memoirs). The principal assigned common reading related to middle school education during the summer months; teachers discussed and planned curriculum based on these commonly developed understandings. At Parkside, faculty scrutinized and supported each other’s work—both newcomers and old timers—through the course of the school day (as well as during extensive team meeting time and faculty meeting time). The team structure allowed the school, in the view of the school principal, to successfully socialize the new teachers, even those who were not traditionally prepared teachers. The principal also relied on staff developers, both in-house and district “coaches,” as well as on her own energies to bring her new teachers up to speed.

How the District Enabled the Reform Theories to Coexist Productively

The manner in which the district pursued its reform goals, especially its more recent emphasis on standards-based instructional improvement, had a good deal to do with the coexistence of the two sets of reform ideas at the school level. Here, the district appeared to anticipate and forestall what might otherwise have been a collision. Consider the potential collision between a notion of balanced literacy teaching, generated initially for the elementary school level, and the prevailing practices and norms in the middle schools’ literacy programs.

Inflexibly applied, the balanced literacy model could have generated heated resistance, and initially it did elicit some grumbling from middle school teachers who felt it was not appropriate for their level. In response, the district listened and learned. It set about adapting its ideas about literacy teaching to the unique circumstances of the middle school grades, and at the same time respecting what was already being done effectively at this level. Here is how the Assistant Superintendent, who had formerly been an elementary school principal, described her own adjustment early in the second year of her tenure in District M:

Middle school classrooms are a different issue and I am learning while I am in this district that middle school teachers don’t often see themselves as teachers of reading, even though our students are students that need to learn to read.

Prompted by this insight and by continuing questions from leaders of the lower performing schools about how to help their struggling readers, she gathered a group of school people together to help her “invent a middle school reading piece”:

I did shared reading lessons [in the middle school]. Now a shared reading lesson is seen as large text, “Big Books,” in grades K-1-2. You read aloud and the kids join you and you may teach a strategy. But here I am using this in a 6th and 7th grade class. So I got the materials ahead of time and thought about what I would want to make explicit in the text to support readers at the 6th and 7th grade level. How would you use it in large text to do that? It was such fun to figure it out together—all of us. The teacher wrote me a thank-you note; she had never thought of it; the principal is all excited and she is making a literacy block in the morning…We used an overhead projector; you aren’t going to use a big book, because it’s insulting…. We used an overhead projector of text they were already familiar with so we could talk about strategies young adults use, all the while knowing that they were struggling readers…

This vignette illustrated a pattern that was pervasive in the district: by taking the time to figure out what their audiences knew, needed, and valued, the district was able to stay true to its goal of standard-based instructional improvement, while respecting the position of each individual school. In this sense, the district didn’t deviate from its long-held assumption that small schools, each with their own character, would provide the student population with greater opportunities than they would receive from traditional, large urban middle schools serving a neighborhood attendance area. At the same time, the superintendent looked at the district’s push for standards-based instructional improvement as a means to enhance the value of small schools of choice. As she put it: “School cultures are important, but you also have to embrace a district-wide effort, and develop a community of learners—or, pride in your community.”

The district’s responsiveness to the schools was met with a corresponding spirit at the schools. Parkside’s response to the district’s push for a portfolio-based assessment system offers an illustration. On their own initiative, after attending several summer institutes at Harvard University, the Parkside staff had developed and implemented a process-based portfolio system for the school. When the district chose to participate in a city-sponsored pilot of a standards-based portfolio system, a move that district leaders saw as supporting its instructional improvement agenda, the staff at Parkside strongly resisted the change. They felt that the district’s presentation of the new system was a poor fit with their well-developed understanding of performance-based assessment systems. However, the principal at Parkside saw an opportunity to think deeply about the school’s portfolio process. Under her leadership, they reconsidered the district’s standards-based system and worked hard over a two-year period to integrate it into their own process. As a result, students in all grades were guided in presenting two portfolios at the end of the school year. They were aware of the difference between the two portfolios and able to articulate that one system was about their “best (standard-bearing) work” and one highlighted their learning in several thematically based areas of growth.

The district, to their credit, exhibited respect for Parkside’s efforts. They let Parkside’s principal take the lead with her staff in developing a response to the district policy. In the end, the district invited the school to present their process to school leaders and staff developers across the district as a model for thinking deeply about assessment practices.

Tensions Between the Reform Theories

The example of Parkside’s response to the district’s attempts to promote portfolio-based assessment underscores the fact that converging courses of action guided by the two reform theories can generate significant friction. The first response of Parkside staff to the district’s initiative in this instance was to be annoyed and resentful. That is entirely understandable: the school’s collective action to define its approach to assessment (thereby realizing a basic tenet of first reform theory) ran up against a different set of ideas and demands from the district about assessment and instructional improvement (a manifestation of the second reform theory at work). However, the tension was managed productively, given some flexibility on both sides, and the ability of leaders to visualize a better outcome through compromise than through rigid insistence or continued resistance.

There were many instances of friction in this case. Though derived in part from the potential incompatibilities between the two theories, the frictions were made more acute by certain conditions in the context of the schools, in particular, intensifying accountability pressures. For example, new state and city assessments implemented during the time of our study, and the accompanying accountability requirements, created mixed messages regarding the importance of complex instruction in the face of the press for test preparation. The district was caught in the middle, mediating between the state and city messages and its own. And as scrutiny of student performance and accountability demands increased, the district had more incentive to assert centralized control, consistent with the second reform theory, and abandon the more decentralized stance implied by the first reform theory. As scholars have noted in studying decentralization and centralization, the messiness and inconsistencies of the former by their very nature encourage the system to revert to centralization as concerns for evaluation increase (Weiler, 1990).

These pressures caused the district to change its focus (e.g., on newly tested subjects like science and social studies) and to seek to assert more control. The Superintendent and her assistants changed the way they spoke about “success” over the course of our study. One assistant said early on, “What I look at is school environments, teaching practices, student engagement. I look at practices of assessment. Ongoing. I don’t mean standardized tests, but assessment that is embedded in instruction.” A year or so later, the superintendent noted, “The [state] test scores came back about a week and a half ago and they did not yield the results that I expected and that I demanded. And I went crazy…we did increase by 11% last year in 4th grade, but just 1% this year. It’s in the right direction, but I’m devastated by it.” She communicated her concerns to the schools accordingly.

This pressure was felt in the schools that we studied, even if the school was producing relatively good test scores, with the result that the school staffs put their locally developed curricula aside somewhat in order to prepare students for the tests. At Alvin Ailey, the science teacher spent nearly a month preparing students for the new state test in science, which covered a broader set of science knowledge than is typically taught at the school. At Parkside, the grade-level schedules were manipulated to intersperse short (generally, 35 minute) periods for math test preparation across the grade levels, and teachers began to rethink their core humanities course to make sure that it prepared their students for the new social studies test. A teacher at Cisneros said, “Unfortunately, one of the things that helps organize the priority is all this testing that the kids get. It’s always in the back of my mind. I want to do all these great things in writing workshop, but they have to pass that math and literature exam.” While generally feeling supported by district leaders, the Parkside principal expressed concern, when a test preparation “test” arrived from the district unannounced one morning requiring immediate attention. She put into words what a number of her colleagues were feeling: “This year the system is just squashing schools.”


In this paper, we have presented evidence that two different and potentially conflicting theories of educational reform can be enacted in an urban school district in such a way that they complement each other, each compensating in particular ways for the other’s weaknesses. This phenomenon took place in a mid-sized city school district that struggled with the challenges of many contemporary urban settings (such as teacher turnover, shortages of well-prepared school leaders, the “flight” of middle class families, and changing governance structures), but here the purposeful combination of reform ideas coincided, over the decade of the 1990s, with substantial district-wide gains in student learning outcomes. While our analysis cannot demonstrate a conclusive causal connection, it is plausible that the joint effect of these reform activities has contributed much to the improvement trend.

The way that the theories were implemented was critical in realizing whatever successes can be reasonably attributed to them. The introduction of the reform theories in this instance was accomplished gradually, sequentially, and with some sophistication and flexibility. In the three middle schools on which we have concentrated, this purposeful combination of reform theories in large part worked, with each theory setting the stage for the other, though more so in some schools than others.

These cases make tangible an emerging argument concerning the power and limits of the theories of action underlying urban educational reform strategies. This line of analysis asserts that, by highlighting certain actions and conditions as essential to the renewal of teaching and learning, these theories simultaneously assume that other conditions (not targeted or touchable by the reform theory) are supportive of the targeted changes (Hill & Celio, 1998). In effect, the strategy in question—whether it emphasizes comprehensive school reforms, vouchers, or intensive investment in professional development—presumes a “zone of wishful thinking” (p. 17), in which events that the policymaker does not control can work for or against the theory’s premises. A voucher strategy, for example, shifts the allocation of resources from institutions to “consumers” (parents), empowering them to select and support schools that are responsive to their preferences, and thereby creating a compelling set of incentives for schools to perform well. But it presumes “an adequate supply of teachers willing to work in competitive environments, parental diligence in choosing schools, and mechanisms to guarantee a supply of good schools in areas serving less-demanding parents” (Hill & Celio, p. 22). These conditions, the zone of wishful thinking for this strategy, are not included in its theory of action, yet without them, a voucher strategy is very likely to fail. A similar analysis can be done with virtually any reform theory.

This line of thinking leads naturally to a corollary: that reform proposals may be “specifically designed to cause the events that are found in other proposals’ zones of wishful thinking” (Hill & Celio, 1998, p. 23). Thus, by combining reform theories, leaders may be able to create mutually reinforcing conditions for improvement. Though there is no guarantee that this complementarity will pertain in all instances of converging reform theories, it appears to be the case in the instance we have been examining.

In each of the four small-school programs that we observed over this three-year period, we saw evidence of the combined effects of reform policies based on the two theories in question. The district had pushed a set of standards-based ideas vigorously, making strategic use of a cadre of professional developers and insisting on professional accountability on the part of all personnel in the district. In response to district-level leadership, educators in these schools directed a great deal of energy toward teaching a form of "balanced literacy" they deemed appropriate to middle school students, and more powerful forms of mathematics teaching that fit their views of what diverse, middle school students needed. Where the small schools’ efforts had been most successfully implemented, school leaders and staff made the greatest use of the requirements and resources afforded by district efforts guided by the second reform theory. The efforts of school staff resulted, at minimum, in a high degree of content alignment across the schools (such as the balanced literacy practices) and, at the very least, surface-level attempts to implement the district’s curricular messages. And, in some cases, the combination of small school reform energies and district-initiated mandates and instructional support resulted in productive, synergistic improvements in instructional practice, as in the way Parkside staff integrated the district’s standards-based portfolio system into their already well-developed use of performance assessment systems.

Constructive co-existence of the two reform theories rested as much on flexible enactment by school and district leaders, as on the fact that the causal assumptions of each compensated for the other’s weaknesses. Neither the small school choice strategy nor the district-driven instructional improvement initiative was implemented mechanically in this case; rather, they were adapted according to the particularities of each school place (Kirp, 1992). The success of the combination of reform theories, then, reflected a balance between district and school initiative; the way leaders at both levels enacted the theory enabled the inherent strengths of each theory of action to offset the weaknesses of the other.

Had no balance been struck between the district-driven instructional improvement efforts and the small schools’ development of a distinctive mission and character, this might have been just another instance of ongoing “policy churn” or the “spinning of wheels,” in which urban districts initiate reform proposal after unsuccessful reform proposal (Hess, 1999). Rather, the district we studied offers an instructive counter case, in which schools were encouraged by the district over a sustained period of time to make local sense of complex instructional ideas pushed by the district and to take advantage of the district’s support in that process.11

Our data indicate, however, that the two reform theories were not implemented as fully insome schools as others, nor were the corresponding synergies among the theories achieved equally. In contrast to Parkside and Cisneros, the two mini-schools we observed at Conant made less use of their small size or the possibilities for reconfiguring time than they could have, and these schools engaged in less overall professional capacity-building. Not surprisingly, classroom teachers’ practice was more individualized and variable and there was considerable variation among student outcomes. District professional developers found it difficult to penetrate the individualized professional norms that reigned in this school’s hybrid organizational structure—a large, urban school redesigned to accommodate four mini-school programs. This is just the kind of restructure that Raywid (1996) suggested might pose challenges for newly formed mini-schools. In such settings, small programs can face difficulty in challenging the strongly established norms and interaction patterns of the original large school, especially if a clear sense of purpose and a strong professional community are not present in the mini-schools. Given this difficulty, it is possible that the district could have done more to support the mini-schools, for example, by pressing harder for school-level structural changes, facilitating the liaison-like role of the professional developer, and insisting upon school conditions that encourage ongoing conversations among teachers, school leaders, professional developers, and district leaders. But doing so would have meant a further adaptation of the first reform theory—which assumes that failing schools will do just that: fail to attract “customers” and close their doors.

By contrast, both Parkside and Cisneros Middle Schools made considerable effort to shape their school practices in ways that supported professional growth and development, knowing their students well, and taking advantage of district supports and initiatives. These school staffs balanced their own interests, skills, and preferences, on the one hand, against the demanding expectations of the district, on the other. This required of school staffs the willingness to treat demands from the outside as opportunities and a potential resource, as much as an intrusion. And it required the district to maintain a flexible and respectful stance, that invited these schools to interact with, and adapt, the reform messages coming from the central office, shaping them to work most effectively for their particular conditions and clientele. In order for the balancing act to result in high-quality outcomes, however, it was incumbent upon strong leadership exercised by school administrators (as at Parkside) or teacher-leaders (as at Cisneros) to engage with the complex ideas espoused by the district. That required staffs at both schools to examine materials, communicate with staff developers, and produce compromises that maintained the integrity of local programs while working toward the kinds of student outcomes desired by the district.

The longer-term implications of this merger of reform theories will never be known. In the year following our study’s conclusion, the New York City Schools were reorganized once again, with all community school boards dissolved and districts such as District M subsumed into larger regional groupings. Yet the District M case stands as a provocative demonstration that widely differing reform theories may coexist, and even productively complement one another, in urban settings. Whether such a merger is possible on the much larger scale of the New York City regional “districts” or in unitary systems as large as Chicago is hard to say. But, in principle, a productive merger is possible that avoids sole or heavy reliance on a centralized strategy of reform, on the one hand, or that presses for a “market” among largely autonomous schools on the other. In such a merger, however, tension is inevitable between a district-driven instructional improvement agenda and one that promotes small schools of choice. Maintaining sufficient flexibility in meeting the needs of the schools and keeping the pressure on for high-quality standard-bearing work is a tricky balance for district leaders to strike, one that is only complicated by the often turbulent context of urban districts. And for educators working in small schools of choice, the work of negotiating district demands must be accomplished simultaneously with local creative endeavors and the day-to-day urgencies of classroom teaching. To simplify their lives, if nothing else, educators might tend to reject one theory in favor of the other. The challenge is to engage both, accept the tension, and to struggle productively within it.


Relative Performance on City and State Tests in Case Study Schools

City and State Test Scores, 2000

Four Middle School Programs in District M

(Percent of Students at Level 3 and Level 4)

Language Arts

Grade 6 (CTB-R)

Grade 7 (CTB-R)

Grade 8 (State ELA)









 Conant: Ailey




 Conant: Discovery






Grade 6 (CTBM)

Grade 7 (CTBM)

Grade 8 (State)









 Conant: Ailey




 Conant: Discovery






1. In neither District 4 nor District 2 was Alvarado’s reform strategy as uniform as the statements here imply—In District 4, some attempts were made at a more standardized approach, and District 2 encouraged alternative schools of choice, as well. Nonetheless, there was a clear shift in the central tenets of reform.

2. Proponents of small schools of choice who speak from the vantage point of individual schools, on the other hand, like Deborah Meier, are quite specific about teaching and learning. The district-level theory assumes that this will be the case.

3. The push for high standards does not necessarily entail standardization of curriculum, but such standardization is likely as leaders seek to equalize educational opportunities. In such instances, leaders may push, by design or default, for teachers to do the same things the same way, so that all students have “equal” access to learning opportunities.

4. The CTP Core Study was a five-year study of four state teaching policy environments that included the study of an urban district in each state and a set of schools in each district. The four states included New York, California, Washington, and North Carolina.

5. The reader is referred to www.ctpweb.org for a description of the broader CTP Core Study. For an example of the multi-level analyses emerging from the Core Study (in a different state) see Darling-Hammond, L. Hightower, A. M., Husbands, J. L., LaFors, J. R., Young, V. M., & Christopher, C. (2005). Instructional leadership for systemic change: The story of San Diego’s school reform. Lanham MD: Scarecrow Press.

6. The cross-case analyses included the development of themes that explained the data across the schools and within the categories suggested by the analytic framework and the Core Study research questions. For example, we read and reread all of the school cases and the district case and looked for instances that explained the similarities and differences between the schools and classrooms that we studied within the context of the district teaching policies. Using the content analysis methods described by Miles and Huberman (1994), members of the research team assigned codes to meaningful data segments, chunking them into categories based on the relationships between teaching policies and instructional practices. In this manner, we produced the set of themes and explanations that are put forth in this paper.

7. We refer throughout the paper to three middle schools that were included in the study. However, one of the schools was a large complex that housed four mini-schools that were not as independent in their organizational structure as the other schools that we studied. Although we selected two of the four mini-schools for inclusion in the study, we refer to our school sample as including three middle “schools” because that is the way the district referred to them.

8. The term “progressive” is used here to describe an educational tradition that emphasizes the discovery of knowledge, the development of habits of mind and problem-solving skills, and favors student-directed, project-or problem-based, hands-on learning experiences. The role of the teacher in this tradition tends to be that of a facilitator of a mix of individual or small group, student-directed learning experiences interspersed with teacher-led small or whole-group discussions of relevant materials.

9. In using the term “constructivist” in reference to mathematics instruction, we draw upon the

recommendations of the National Council of Teachers of Mathematics (NCTM) that call for teachers to, for example, build mathematics knowledge through the use of problem solving and to orchestrate classroom dialogue in ways that facilitate students’ construction of mathematical concepts.

10. Here, “teacher-directed” instruction refers to practice in which teachers provide whole groups of students with extensive explanations and factual information; knowledge is treated as fixed, to be transmitted from teacher to student. Generally, the teacher structures class time to include teacher-led whole-group discussions and/or individual completion of teacher-assigned lessons.

11. Hess (1999), by the way, notes that a few urban districts do produce sustained improvement. He argues that there is a need for research that examines in more detail the contextual, organizational, and leadership qualities that characterize such successful district efforts.


Ancess, J. (2000). The reciprocal influence of teacher learning, teaching practice, school restructuring, and student learning outcomes. Teachers College Record, 102, 590-619.

Argyris, C., & Schon, D. A. (1982). Theory in practice. San Francisco: Jossey-Bass.

Atwell, N. (1998). In the middle: New understanding about writing, reading, and learning. (2nd ed.), Portsmouth, NH:  Boynton/Cook Publishers.

Berends, M., Bodilly, S. J., & Kirby, S. N. (2002). Facing the challenges of whole-school reform: New American Schools after a decade. Santa Monica, CA: RAND Corporation.

Berends, M., Chun, J., Schuyler, G., Stockly, S., & Briggs, R. J. (2002). Challenges of conflicting school reforms. Santa Monica, CA: RAND Corporation.

Bryk, A., Sebring, P., Kerbow, D., Rollow, S., & Easton, J. (1998). Charting Chicago school reform. Boulder, CA: Westview Press.

Christman, J. (2001). Powerful ideas, modest gains: Five years of systemic reform in Philadelphia middle schools. Philadelphia: Consortium for Policy Research in Education.

Coburn, C.E. (2005). The role of non-system actors in the relationship between policy and practice: The case of reading instruction in California. Educational Evaluation and Policy Analysis, 27, 23-52.

Cotton, K. (1996). School size, school climate, and student performance. American Association of School Administrators (Online). Available: http://www.aasa.org/issues_and_insights/prof_dev/rural_cotton.htm

Cross-City Campaign for Urban School Reform (1995). Reinventing central office. Chicago: Author.

Darling-Hammond, L. (1997). The right to learn: A blueprint for creating schools that work. San Francisco: Jossey-Bass.

Darling-Hammond, L., Ancess, J., & Falk, B. (1995). Authentic assessment in action: Studies of schools and students at work. New York: Teachers College Press.

Darling-Hammond, L., Hightower, A. M., Husbands, J. L., LaFors, J. R., Young, V. M., & Christopher, C. (2005). Instructional leadership for systemic change: The story of San Diego’s school reform. Lanham, MD: Scarecrow Press.

Darling-Hammond, L., & Snyder, J. (2003). Organizing schools for student and teacher learning: An examination of resource allocation choices in reforming schools. In Plecki, M. L., & Monk, D. (Eds.), School finance and teacher quality: Exploring the connections. The 2003 Yearbook of the American Educational Finance Association. Larchmont, NJ: Eye on Education.

David, J. L., & Shields, P. M. (2001). When theory hits reality: Standards-based reform in urban districts. Menlo Park, CA: SRI International.

Elmore, R. F., & Burney, D., (1997). School variation and systemic instructional improvement in Community School District #2, New York City. Unpublished manuscript, University of Pittsburgh, Learning Research and Development Center.

Fink, E., & Resnick, L. (1998). Developing principals as instructional leaders. Phi Delta Kappan, 82, 598–606.

Firestone, W. A., & Shipps, D. (2003). How do school leaders create effective programs for school improvement out of conflicting accountabilities? Presented at the annual meeting of the American Educational Research Association, Chicago, IL, April 21-25.

Fliegel, S. (1993). Miracle in East Harlem: The fight for choice in public education. New York: Random House.

Fullan, M. (1994). Coordinating top-down and bottom-up strategies for educational reform. In R. F. Elmore, & S. H. Fuhrman (Eds.), The governance of curriculum. Alexandria, VA: Association for Supervision and Curriculum Development.

Fullan, M. (2001). Leading in a culture of change. San Francisco: Jossey-Bass.

Fuhrman, S. (Ed.) (2001). From the capitol to the classroom: Standards-based reform in the states—One hundredth yearbook of the National Society for the Study of Education(NSSE). Chicago: University of Chicago Press.

Hatch, T. (1998). Differences in theory that matter in the practice of school improvement. American Educational Research Journal, 35, 3-32.

Hatch, T. (2002). When improvement programs collide. Phi Delta Kappan, 93(8), 626-634.

Henig, J. (1994). Rethinking school choice: The limits of the market metaphor. Princeton, NJ: Princeton University Press.

Hess, F. M. (1999). Spinning wheels: The politics of urban school reform. Washington, DC: Brookings Institution Press.

Hightower, A. (2002). San Diego’s big boom: Systemic instructional change in the central office and schools. In A. Hightower, M. S. Knapp, J. A. Marsh, & M. W. McLaughlin (Eds.), School districts and instructional renewal (pp. 76-93). New York: Teachers College Press.

Hill, P. T. (Ed.). (2002).

Choice with equity—An assessment by the Koret Task Force on K-12 education. Stanford, CA: Hoover Institution Press.

Hill, P. T., & Celio, M. B. (1998). Fixing urban schools. Washington, DC: Brookings Institution Press.

Hill, P., Foster, G. E., & Gendler, T. (1990). High schools with character. Santa Monica, CA: The Rand Corporation.

Jennings, N. E. (1996). Interpreting policy in real classrooms: Case studies of state reform and teacher practice. New York: Teachers College Press.

Kimbrough, J., & Hill, P. T. (1981). The aggregate effects of federal education programs. Santa Monica, CA: The Rand Corporation.

Kirp, D. L. (November, 1992). What school choice really means. Atlantic Monthly Online. Available: http://www.theatlantic.com/politics/educatio/kirpf.htm

Knapp, M. S., Bamburg, J. D., Ferguson, M. C., Hill, P. T. (1998). Converging reforms and the working lives of frontline professionals in schools. Education policy, 12, 397-418.

Knapp, M. S., Stearns, M. S., Turnbull, B. J., David, J. L., Peterson, S. M. (1991). Cumulative effects of federal education policies at the local level. In Odden, A. (Ed.), Policy implementation (pp. 105-124). Albany: SUNY Press.

LeCompte, M.D., & Preissle, J. (1993). Ethnography and qualitative design in educational research (2nd Ed.). San Diego: Academic Press.

Lee, V., Bryk, A., & Smith, J. (1993). The organization of effective schools. Review of Research in Education, 19, 171-268.

Lee, V., & Smith, J. (1995). Effects of high school restructuring and size on early gains in achievement and engagement. Sociology of Education, 68, 241-270.

Lieberman, A. J., & Sostre, L. (1998). Implementing public school choice. New York: Center for Educational Innovation, Manhattan Center.

Little, J. W. (1999). Organizing schools for teacher learning. In Darling-Hammond, L., & Sykes, G. (Eds.), Teaching as the learning profession: Handbook of policy and practice (pp. 233-262). San Francisco: Jossey-Bass.

Malen, B., Croninger, R., Muncey, D., & Redmond-Jones, D. (2002). Reconstituting schools: “Testing” the “theory of action.” Educational Evaluation and Policy Analysis, 24, 113-132.

Massell, D., & Goertz, M. (2002). District strategies for building instructional capacity. In A. Hightower, M. S. Knapp, J.A. Marsh, & M. W. McLaughlin (Eds.), School districts and instructional renewal (pp. 43–60). New York: Teachers College Press.

McDonald, J. (1996). Redesigning school. San Francisco: Jossey-Bass.

McLaughlin, M. W., & Talbert, J. E. (2002). Reforming districts. In A. Hightower, M. S. Knapp, J. A. Marsh, & M. W. McLaughlin (Eds.), School districts and instructional renewal (pp. 173–192). New York: Teachers College Press.

Meier, D. (1995). The power of their ideas: lessons for America from a small school in Harlem. Boston, MA: Beacon Press.

Merriam, S. (1998). Qualitative research and case study applications in education. San Francisco: Jossey-Bass.

Miles, M. B. & Huberman, A. M. (1994).  Qualitative data analysis: 2nd edition.  Thousand Oaks:  Sage Publications.

Murphy, J., & Hallinger, P. (1988). The characteristics of instructionally effective school districts. Journal of Educational Research, 81(3), 175-181.

Murphy, J., & Hallinger, P. (1986). The superintendent as instructional leader: Findings from effective school districts. Journal of Educational Administration, 24(2), 213-236.

Newman, F., & Associates (1996). Authentic achievement: Why are some schools more successful than others? San Francisco: Jossey-Bass.

Peterson, G. (1999). Demonstrated actions of instructional leaders: An examination of five California superintendents. Education Policy Analysis Archives, 7(18); on the Web at: http://epaa.asu.edu/epaa/v7n18.html.

Portin, B. S., Beck, L., Knapp, M. S., & Murphy, J. (2003). The school and self-reflective renewal: Taking stock and moving on. In Authors (Eds.), Self-reflective renewal in schools: Local lessons from a national initiative (pp. 179-200). Westport, CT: Praeger.

Raywid, M.A. (April, 1996). Taking stock: The movement to create mini-schools, schools-within-schools, and separate small schools. New York: ERIC Clearinghouse on Urban Education. Available: http://eric-web.tc.columbia.edu/monographs/uds108/

Resnick, L. B., & Glennan, T. K. (2002). Leadership for learning: A theory of action for urban school districts. In A. M. Hightower, M. S. Knapp, J. A. Marsh, & M. W. McLaughlin (Eds.), School district and instructional renewal (pp. 160-172). New York: Teachers College Press.

Smylie, M. A., Wenzel,. S. A., Allensworth, E., Fendt, C., Hallman, S., Luppescu, S., & Nagaoka, J. (2003). The Chicago Annenberg Challenge: Successes, failures, and lessons for the future—Final technical report of the Chicago Annenberg Research Project. Chicago: Consortium on Chicago School Research.

Snipes, J., Doolittle, F., & Herlihy, C. (2002). Foundations for success: Case studies of how urban school systems improve student achievement. Washington, DC: Council of Great City Schools.

Stein, M. K., & D’Amico, L. (2002). Inquiry at the crossroads of policy and learning: A study of a district-wide literacy initiative. Teachers College Record, 104, 1313–1344.

Togneri, W., & Anderson, S. E. (2003). Beyond islands of excellence: What districts can do to improve instruction and achievement in all schools. Washington, DC: Learning First Alliance.

Weiler, H. N. (1990). Comparative perspectives on educational decentralization: An exercise in contradiction? Educational Evaluation and Policy Analysis, 12, 433-448.

Weiss, C. (1995). Nothing as practical as good theory: Exploring theory-based evaluations for comprehensive community initiatives for children and families. In J. Connell, A. Kubitsch,  L. Schorr, & C. Weiss, (Eds.), New approaches to evaluating community initiatives, concepts, methods, and contexts (pp. 65-92), Washington DC: The Aspen Institute.

Yin, R. K. (2003). Case study research. (4th ed.). Newbury Park: Sage Publications.

Cite This Article as: Teachers College Record Volume 109 Number 12, 2007, p. 2601-2641
https://www.tcrecord.org ID Number: 14487, Date Accessed: 1/25/2022 5:44:58 PM

Purchase Reprint Rights for this article or review
Article Tools
Related Articles

Related Discussion
Post a Comment | Read All

About the Author
  • Chrysan Gallucci
    University of Washington
    E-mail Author
    CHRYSAN GALLUCCI, Ph.D.,is research assistant professor in educational leadership and policy studies at the University of Washington and a research associate with the Center for the Study of Teaching and Policy (CTP). Her teaching and research focus on education policy and its connection to issues of professional learning, especially for education professionals who work in diverse systems and engage in problems of practice related to the improvement of teaching and learning. Dr. Gallucci specializes in qualitative methods of research. She is research director for the Center for Educational Leadership and co-director of the Masters in Instructional Leadership program at the University of Washington.
  • Michael Knapp
    University of Washington
    E-mail Author
    MICHAEL KNAPP, Ph.D., is a professor of educational leadership and policy studies and director of the Center for the Study of Teaching and Policy in the College of Education at the University of Washington engages in teaching and research focused on educational policymaking, school reform, leadership development, and policy research methods, with emphasis on how policy and leadership connect to classroom and school improvement. His studies have often concentrated on the education of disenfranchised populations, mathematics and science education, and the professional development of educators. Dr. Knapp has written extensively about his research, including eight books, among them, School Districts and Instructional Renewal (2002), Investigating the Influence of Standards (2002), Self-Reflective Renewal in Schools (2003), and Leading Learning in Schools and Districts (2006).
  • Anneke Markholt
    University of Washington
    E-mail Author
    ANNEKE MARKHOLT, Ph.D., began her career as an English as a second language specialist for the Tacoma Public Schools where she taught for 10 years. Currently, she is a project director for the Center for Educational Leadership at the University of Washington, where she designs and directs partnerships between CEL and school districts focused on the development of instructional leadership at all levels of school district systems. She is particularly interested in the intersection of teaching, learning, and the leadership capacity necessary for school systems to engage in instructional improvement. Prior to her work with CEL, she spent five years as an associate researcher for the Center for the Study of Teaching and Policy at the University of Washington.
  • Suzanne Ort

    E-mail Author
    SUZANNE ORT, Ed.D., began her career as a social studies teacher at an alternative high school in the Bronx and then as a teacher of English as a second language in the Czech Republic. She has worked as a research associate at the National Center for Restructuring Education, Schools, and Teaching (NCREST) at Teachers College, Columbia University on several research and development projects including revamping the Regents assessment system in New York State and large scale studies of policy implementation in New York City Community School Districts 2 and 3 (with the Center for the Study of Teaching and Policy). Currently she is a “teaching coach” at Park East High School in New York City through the Institute for Student Achievement.
Member Center
In Print
This Month's Issue