Scaling Up School Restructuring in Multicultural, Multilingual Contexts: Early Observations from Sunland County

Sam Stringfield, Johns Hopkins University Amanda Datnow, Johns Hopkins University Steven M. Ross, University of Memphis (1998) Introduction Two of the most rapidly developing fields in educational research are diversity education (e.g., Tharp & Gallimore, 1988) and school restructuring/school reform (e.g., Murphy & Hallinger, 1993; Newmann & Wehlage, 1995; Tyack & Cuban, 1995). Unfortunately, the intersection of the two fields is virtually uncultivated. None of the nationally disseminated school restructuring models was developed specifically for multilingual, multicultural contexts. With the exception of Éxito para Todos, the Spanish version of Success For All (Slavin, Madden, & Wasik, 1996), there is virtually no research on the effectiveness of the programs in achieving implementation, let alone improvements in student achievement, in multilingual, multicultural contexts. In the first study of its kind, three broad policy questions will be addressed through the study "Scaling Up School Restructuring in Multicultural, Multilingual Contexts." They include the following: How effective are current generation school restructuring proposals in improving the achievement of students in schools serving large numbers of language minority students in a multicultural context? Are some of the current school restructuring models better suited to multilingual, multicultural contexts than others? Can the various reforms be successfully modified to provide high quality educational services to all students in diverse multicultural, multilingual contexts? What actions at the federal, state, school district, subdistrict, and school levels increase (or decrease) the probability of obtaining full benefits from any or all of the restructuring models when the models are being implemented in multicultural, multilingual contexts? This paper presents early observations of 13 culturally and linguistically diverse schools in Sunland County.1 Each school is in the process of implementing an externally developed school restructuring design. Sunland County Public Schools provides education to students from a very richly diverse set of cultures and language groups. Sunland County has one of the largest second language populations of any district in the country, with Spanish and Haitian Creole being the most common of over 100 languages and dialects. In this paper, data from the 1st of 4 years of field work will be reported regarding early insights into the classroom, school, and district conditions and actions that are facilitating or hindering implementation of the various school reforms. Over the years 1998-2001, additional reports will be produced on issues related to institutionalization of the reforms at each school site and on the effects of the reforms on the academic progress of various groups of students at the schools. Background School Restructuring in Multilingual, Multicultural Contexts A major goal of multicultural education in the United States is school reform aimed at ensuring that students from diverse racial, ethnic, and socioeconomic backgrounds experience educational equality in our schools. Meeting the goals of multicultural education requires a substantial overhaul of the schooling process (Banks & Banks, 1995). While none of the nationally disseminated school restructuring models was developed specifically for multicultural contexts, most of the reforms emphasize equity, constructivist approaches to learning, team teaching, and cooperative learning, and all declare the need to respect diverse cultures. Can the current restructuring models work in multicultural contexts? If at some schools the answer to that question is "yes," when, where, and why do the reforms work? To what extent must the designs be modified to accommodate local, diverse communities? To what extent must these communities accommodate themselves to the designs? How can educators at the national, state, and local levels, as well as concerned community members, most effectively contribute to the successful implementation of restructuring efforts that have the potential to work for diverse populations? The District Context The major, longitudinal component of the study is taking place in 13 currently restructuring elementary schools in Sunland County. Located in a Sun Belt state, the Sunland County school system includes the city of Sunland and surrounding suburban and agricultural regions. Sunland County Public Schools (SCPS) is one of the largest public school districts in the United States, serving more than 300,000 students. The system includes over 250 schools spread over a county larger than the state of Rhode Island. The neighborhoods of Sunland include some of the wealthiest in the country and some characterized by extreme poverty. The district has schools that serve rural areas with primarily migrant farmworker populations and schools in the heart of the inner city that serve primarily families from diverse countries and backgrounds. For several decades, SCPS has experienced large influxes of immigrant students, many of whom have had no prior schooling. Given the coming demographic changes in America's school population (see Natriello, McDill, & Pallas, 1990), the challenges currently facing Sunland foreshadow the challenges other communities will face early in the next century. New American Schools In areas such as multicultural education, bilingual education, and magnet programs, SCPS is at the forefront of educational innovation. Several of the nation's most highly regarded school restructuring programs are represented in SCPS, and since the spring of 1995, the district has received special support from New American Schools (NAS) to implement several of the NAS designs at multiple sites. Created in 1991 as part of Goals 2000, New American Schools was charged with securing financial support from foundations and corporations to fund new designs for "break-the-mold" schools (Kearns & Anderson, 1996). Out of 686 proposals submitted for the competition, 11 received funding, and 9 of these were continued through 3 years of development. In 1995, after supporting the development and piloting of the designs (Phases 1 and 2), New American Schools formally launched its dissemination or "scale-up" operation (Phase 3) by inviting school districts and states to participate as "jurisdictions" that would implement the designs in individual schools over a 5-year period. SCPS was selected as one of these jurisdictions, and the school district chose to offer support for five of the NAS designs. NAS design teams provide yearly on-site training and implementation assistance to schools implementing NAS designs. Separately, the Rand Corporation has contracted with New American Schools to conduct an overall study of progress across the 10 national jurisdiction sites. The Rand evaluation provides an overview of implementation issues across jurisdictions (Bodilly, 1996). Methods This study is multimethod, examining diverse processes and outcomes through a combination of quantitative and qualitative methods. The longitudinal component of the study involves following 13 restructuring elementary schools over a 4-year period, with a single follow-up visit to each school during year 5. Quantitative data on inputs, processes, and outcomes will be melded with qualitative, longitudinal case studies of each site. The replication component will rely on less detailed qualitative data gathering and the integration of locally available quantitative evaluations, such as Title I evaluations. To the extent available, one third of the replicate sites will be in Sunland County, one third in Memphis (TN) City Schools, and one third drawn from a nationally nominated pool of schools providing exemplary, design-specific, restructuring services in diverse multicultural contexts. Study Sample The 13 longitudinal sites in the study were drawn from schools currently implementing nationally regarded school restructuring models. In collaboration with district administrators at SCPS, we identified six restructuring models for study in Sunland County. The list includes three New American Schools designs - Roots & Wings, Audrey Cohen, and Modern Red Schoolhouse (see Stringfield, Ross, & Smith, 1996) - and three externally developed designs: the Comer School Development Program (Comer, 1980; Comer & Haynes, 1996), the Core Knowledge Sequence (Core Knowledge Foundation, 1995; Hirsch, 1987, 1993), and the Coalition of Essential Schools (Sizer, 1984, 1992, 1996). (For brief descriptions of each design, see the Appendix.) We asked personnel from the district's Office of Instructional Leadership to nominate schools for inclusion in our study and to seek approval for participation in the study from the principals of each school. There were several criteria we asked them to use in making selections. First, we asked them to choose schools with potentially successful implementation efforts related to designs that had been in place for 2 or more years by the beginning of our study. Second, we asked for at least two elementary schools using each of the restructuring designs listed above. In some cases (e.g., Coalition of Essential Schools), there were only two elementary schools in the district using that design. With other designs, schools for our study were chosen from a pool of schools using those designs. Beyond CES, the number of elementary schools using a specific design ranged from 3 (Modern Red Schoolhouse) to more than 40 (Roots and Wings); however, in most cases there were fewer than 5 schools employing each design type. Finally, we asked district personnel to include schools that served a diversity of ethnic and linguistic minority groups, realizing, of course, that the composition of each school would vary according to the particular community served by the school. Although the various restructuring designs share certain characteristics, such as active learning, constructivist teaching, and a focus on reforming the whole school, the six designs we are studying in SCPS differ from one another in several key ways. One difference is the extent to which the design relies largely on local development of design details or is a more fully specified, detailed reform design. At one end of the continuum, the Coalition of Essential Schools (Sizer, 1984, 1996) and the Comer School Development Program (Comer, 1980; Comer & Haynes, 1996) provide frameworks for reform and leave particulars to each school. These designs emphasize the primacy of local development efforts, as long as the process is guided by a set of overarching principles (and structures, in the case of the Comer program). A bit further along the continuum, the Core Knowledge Sequence (Core Knowledge Foundation, 1995) provides detailed curricula for one half of each day for each elementary grade, while leaving issues of how to teach reading or history and how to organize the school to the judgment of the principal and faculty. Tending toward more specification are the Modern Red Schoolhouse and Audrey Cohen College designs. Modern Red Schoolhouse aims to extend and deepen the educational assets embodied in the classic "little red schoolhouse" (Heady & Kilgore, 1996) through an individually paced approach to learning established between a student and his or her parents and teacher, and specifies organizational and curricular elements (including the use of the Core Knowledge curriculum). The Audrey Cohen College System of Education redesigns the entire school setting, including the curriculum, to achieve meaningful purposes using an interdisciplinary approach (Cohen & Jordan, 1996). At the even more highly specified end of the continuum is Roots and Wings, a design that provides detailed descriptions of how to organize schools and classrooms and what and how to teach (Slavin et al., 1996). Roots and Wings extends and broadens a successful reading program developed at the Johns Hopkins University, Success for All, into a whole school restructuring program, complete with materials (Slavin et al., 1996). As for the effectiveness of these different approaches, the Special Strategies studies (Stringfield et al., 1997) found that within its largest sample of schools, students in schools using externally developed designs tended to achieve greater academic gains than did students in locally developed programs. Crandall et al. (1982) and Herman and Stringfield (1997) found replicating studies suggesting that several externally developed programs had potential to improve academic achievement in diverse contexts. For example, multiple studies found that Success for All, one of the more promising and frequently studied reforms, made dramatic differences in levels of student achievement gains at some schools and less substantial differences in others (Ross, Smith, & Casey, 1997; Slavin, Madden, Karweit, Dolan, & Wasik, 1992; Stringfield et al., 1997). Specifically, school and district-specific variables had large impacts on implementation levels, and implementation level was a major variable in eventual effects. All of the studies found that, where fully implemented, the program can have positive effects on student achievement, but that full local implementation required years of coordinated effort. All selected SCPS schools serve multicultural, bi- or multilingual communities. In almost all of the schools, the majority of students are from contexts traditionally regarded as placing them at risk of educational failure. For example, one of the schools in the study serves a population of 1,057 students, 53% of whom are Black, 22% Hispanic, 15% White, and 10% Asian. Sixteen percent of the students at this school speak English as a second language, and 72% receive free or reduced-price lunch. Another school has 885 students, 79% of whom are Hispanic, 18% White, 1% Black, and 1% Asian. Seventy-eight percent of the students speak English as a second language, and 46% receive free or reduced-price lunch. (See Table 1 for demographic information about each school in our study sample.) While these numbers represent ethnic and racial diversity, they do not express the wide diversity that exists within categories. Students classified as Hispanic in SCPS include young people of Cuban, Mexican, South American, or Latin American descent. Students classified as Black include Haitians, Caribbeans, Africans, and African-Americans. The school system serves students from every country in the Caribbean, Central America, and South America, as well as students from other nations around the world. Table 1. Longitudinal School Sample School Namea Design Total Number of Students Free and Reduced Lunch (%) Limited English Proficiency (%) Hispanic (%) Black (%) Asian, Indian, or Multiracial (%) White (%) Usec (%) Fourth Grade Reading Comprehension b (%) Fourth Grade Math Concepts b (%) Giant Mangrove R&W 798 88.5 2.5 6 93 0 1 107 26 49 Nautilus Beach R&W 1,126 88.3 19.8 15 75 1 8 178 22 36 Orchid Gardens R&W 1,411 85.9 18.9 18 72 3 8 155 28 45 Inland Tupelo MRSH 1,025 78.5 36.8 71 11 1 17 104 42 54 Longleaf Bay MRSH 609 28.9 17.7 71 2 1 26 103 52 69 Wild Cypress Core 1,188 72 33.3 91 1 1 7 123 37 66 Alligator Jetty Core 1,852 35 9.6 49 26 4 21 180 47 69 Endless Keys AC 885 46.3 21.8 79 1 1 18 129 52 69 Prairie Sawgrass AC 1,056 71.7 16 22 53 10 15 130 50 59 Flamingo Forest Comer 950 88.3 43.7 77 11 2 9 92 42 59 Hibiscus Cove Comer 1,225 74.9 51.9 92 1 1 5 129 42 66 Red Cedars CES 896 18 4.6 16 19 3 63 130 67 76 Limestone Way CES 946 30.1 14 52 7 4 37 106 ? ? District average ? 847 70.12 22.4 51 34 1 14 ? 35 51 NOTE: R&W = Roots and Wings; MRSH = Modern Red Schoolhouse; Core = Core Knowledge Sequence; AC = Audrey Cohen; Comer = Comer School Development Program; CES = Coalition of Essential Schools. a. For purposes of confidentiality, pseudonyms are used in place of actual school names. b. Percentile scores on 1995-1996 administration of Stanford Achievement Test. c. Percent utilization refers to the capacity of school space in relation to the number of students attending the school. Replication Sites Following Guba and Lincoln (1981), Miles and Huberman (1994), and Yin (1989), the research team has built a "stepwise replication" component into the study's design. During Years 2, 3, and 4 of the longitudinal study, researchers will visit schools implementing similar restructuring designs (e.g., additional Modern Red Schoolhouse or Coalition of Essential Schools sites) in Sunland and around the country. These stepwise replicates, each chosen to address emerging issues of replicability of findings, provide a relatively low-cost method of gaining deeper understanding of the contextual variables that may affect implementation. This additional component of the study began in fall 1997. Input Variables The study is gathering district information on students' ethnic background, language status, poverty status (free and reduced-price lunch participation), exceptional student classification (e.g., identified for special education or gifted programs), and performance in reading and math on the district-mandated Stanford Achievement Test. Process Variables The study is using a combination of low-inference classroom observations and more descriptive multiyear observations of students, classrooms, schools, and the district, measuring such discrete data as evidence of design implementation and mean student engagement rates. The study requires trained observer assessments of "authentic pedagogy" (Newmann & Wehlage, 1995).2 what degree teachers are accommodating and/or addressing diversity in their classrooms and to what degree there is evidence of implementation of the restructuring model. To gather this observational data, we are conducting 90-minute observations in classrooms across each grade level during subject matter instructional time (e.g., math, reading, science). In addition, during Year 1 of the study, researchers conducted whole school day observations of two first grade students in each school (both of average achievement level, including one limited English proficient [LEP] student per school). Where possible, we will continue to follow these students periodically throughout the duration of the study. Previous studies (e.g., Stringfield et al., 1997) have found such shadowing exercises to be enlightening in getting to the core of an innovation's effects on students. In order to garner perceptions of the impact of restructuring, we are conducting interviews and focus groups (Krueger, 1994) with principals, teachers, parents, and students. During Year 1 of the study, we interviewed the first grade teachers in each school, generally 4 to 6 teachers in each. We will interview each successive grade of teachers as the study continues. We are also interviewing numerous district office administrators and representatives of the various design teams regarding their perceptions of the conditions affecting successful implementation. Outcome Variables This initial report focuses exclusively on issues of implementation. However, as described below, outcome data are being gathered and will be presented in subsequent reports. The study is examining three outcome variables: (a) student attendance and other schooling-related measures, (b) student attitudes, and (c) student achievement. Attendance, rate of promotion, and referral to Title I or special education data will be gathered from archival records. Attitudes toward school will be measured both through standardized questionnaires administered during the 4th year of the study and through fourth grade student focus groups. Student achievement measures will assess writing, complex problem solving, reading comprehension, and abstract reasoning. The quantitative achievement measure is being provided by the district-mandated Stanford Achievement Test (SAT), administered in the spring of each year to all elementary school students. Scores for reading comprehension and mathematics will be examined. The State Writing Assessment, administered to all fourth graders in Sunland County, will be examined as well. Because both instruments are administered to all students in all schools, researchers will be able to compare gains not only among comparable groups from the diverse schools but also against well-matched control schools in the district. The Sunland County school district has agreed to provide quantitative demographic and achievement data for the 13 longitudinal schools, the replication sites, and the demographically matched control sites. In addition, the district has generously committed to facilitate the conduct of the study and to provide detailed feedback on all draft reports. Qualitative analyses, including those presented in this report, rely upon the triangulation of data from multiple sources: whole school day observations, classroom observations, focus groups, interviews, and documents. To identify patterns and key issues within and across sites, data are analyzed through the use of within- and cross-case displays (Miles & Huberman, 1994). Data from the 30 single-visit, stepwise replication sites will be critically important in attempting to gauge the generalizability of findings from the 13 longitudinal sites. In Years 2-5 of the study, quantitative analyses will examine the relationships between implementation level of the restructuring design (determined through 4-year case studies) and academic gains and experimental/control differences in achievement gains over three years. State-of-the-art statistical procedures, such as hierarchical linear modeling (Bryk & Raudenbush, 1992), will be used in analyses. Particular attention will be paid to interactions in outcomes over time between the restructuring model and specific cultural groups (Hispanic, African American, recent Asian immigrants, Whites, and other groups). First Year Observations This paper reports our observations at the end of the first year of case study data collection. These are preliminary findings, subject to modification over time. We discuss variability in fidelity of implementation across sites, citing possible explanations for this variance, and describe some of the factors we believe will affect the successful implementation of the six designs over time in these multilingual, multicultural contexts. Variability of Implementation Fidelity The schools in our study have all gone through the process of adopting nationally recognized restructuring designs in order to improve education for their students. They all have taken significant steps toward full implementation and have sustained 2 or more years of work toward implementation of the designs. Researchers from the "Eight Year Study" (Aikin, 1942), through the Rand Change Agent Study (Berman & McLaughlin, 1978; see also McLaughlin, 1990), to more recent assessments (Stringfield et al., 1997) have found some level of variation in implementation to be nearly ubiquitous. McLaughlin (1990) went so far as to suggest that unplanned variation should be embraced. Although all of the schools we studied were 2 to 4 years into the implementation process, fidelity of design implementation across and within schools varied. We encountered a few schools that appeared to have a fully implemented reform design, a staff that seemed universally committed to the design, and successful integration of the design with instruction for linguistically diverse students. We also observed schools in which perhaps half of the teachers were implementing virtually all the components of a particular reform, schools in which perhaps all of the teachers were implementing half or more of the components of a model, and schools that fell somewhere in between or below. At some schools, teachers described not having full understanding of a model and thus not implementing its elements. Conversely, some teachers understood their school's model but felt it was not consistent with what was best for children; therefore, they were not implementing it. At other schools, we discovered teachers who expressed enthusiasm for the model but suffered from lack of technical classroom skills for implementation or found their time divided by a host of competing programs that collectively hindered design implementation. Again, given the comprehensiveness and complexity of these designs, the difficulties experienced by some of the schools in achieving full implementation is not surprising and is consistent with outcomes in other school districts involved in restructuring (Stringfield & Ross, 1997). What accounts for this variability and the special challenges experienced by particular school staffs? We have identified several factors that may lead to the variable implementation observed between and within sites. Demographic and Numeric Shifts Some schools in our sample have experienced, and continue to experience, considerable demographic shifts. The changing nature of some schools' student populations has undoubtedly affected the capacity of those schools to implement a reform design. Schools often have to integrate new teaching staff to accommodate students who come to school speaking many languages other than English. In addition, substantial demographic shifts in some neighborhoods have required schools to start essentially from scratch in building new relationships with families. One principal told us that in the last 6 years, her school had changed from serving a predominantly white, U.S.-born, middle-income population with highly educated parents to serving a predominantly low-income Haitian immigrant population, many of whom had no prior formal schooling. Growth in student enrollment after the beginning of the school year has also made design implementation difficult for some schools. Principals related that they often receive many new students, enough to fill several classrooms, several months into the school year. Such occurances disrupt even the best-laid local scheduling plans, which are designed to facilitate organized changes necessary for reform, such as common planning time for teachers or team teaching. The Extent to Which Educators Chose Their Designs For several years, the Sunland County district held a fair exhibiting an array of school restructuring designs. Teams of educators from all district schools were invited to attend and choose, or reject, a particular design. Generally, these teams were comprised of a few teachers (sometimes those on the school site council) and the principal or assistant principal. Despite the district's efforts to inform teachers of the various designs through this method, teachers at most schools stated that they knew very little about any other designs that might have been available to them, and some stated that they had known little about the design they were adopting at the time it was chosen. It seemed that only a small number of schools had actually sent groups of teachers to see the designs exhibited at the fair, and very few schools had sent teachers to see the design in action at other schools prior to committing to specific reforms. In addition, in schools where only a small team of teachers (two to three people) attended the design exhibition and brought back information about the one design that was most appealing to them, there was less information about varied designs among the rest of the teachers. While in some cases school staffs found out about designs through the fair, in a number of cases designs were introduced to the teachers by school or district administrators. For example, several teachers reported their school's rationale in choosing a design: "We weren't doing too well in reading on the SAT scores, and the principal thought we needed a new program. We didn't have a choice. She kind of motivated us to vote for it." In their view, the principal brought the design to them and attempted to attain consensus for implementation through a faculty vote. Regretting the principal's choice of a particular design, one teacher remarked, "There are so many programs that I would prefer doing." Site-based management at many schools, therefore, resulted in decision making at a more proximal level of the bureaucracy, not democracy among teachers and administrators. There were some exceptions. A teacher at a school that adopted the Comer model explained, "We were actually part of the decision to become a Comer school. A group of teachers went with the principal to visit another school, to see it in practice, to talk to people, and then we brought it back to the rest of the staff." Still, the teachers at this school soon realized that buy-in to the program was "not going to be 100%." The principal of a Roots and Wings school also reported that the teachers "had an opportunity to look at the different programs that were in Sunland County Schools," and 84% of the teachers voted to accept Roots and Wings after visiting schools that were already implementing it. However, there was some disagreement among the teachers about how much choice teachers actually had, given the administration's strong interest in Roots and Wings. Teachers generally did not report having knowledge about diverse restructuring designs, and in some cases, they felt that the design at their school was implemented through a top-down approach. Nevertheless, teachers in a number of schools saw benefits in the designs their school was implementing and also in the district's pro-restructuring stance. For example, describing the benefits of the Comer model, a teacher remarked, "We've become more aware that we're here for the child." She added, "The things that have improved the most are parent involvement and community involvement." Another teacher described the benefits in terms of student attendance, "More of my kids are coming to school." Describing the benefits of another design, a teacher emphatically stated, "We used to be more didactic. We would give [the students] the information. Now they have to do more thinking on their own, and we don't all have to reach consensus." Another teacher added, "They've been more accepting of each other's answers." Describing the district's general position on reform, a teacher commented, "The district promotes all sorts of programs, which is good." Among many teachers, the perceived lack of choice appears to have led to some of the variation that we witnessed among and within sites. If teachers are not committed to a program, they are unlikely to implement it effectively. For example, one teacher, describing her frustrations with a more highly specified design, remarked, "I don't like the fact that we have a script, and we have to do exactly what the script says." Another added, "It doesn't allow any kind of creativity or individuality." Teachers at another school using this model had very similar complaints about the program, and while the principal has attempted to bring them on board, there is considerable resistance to the program. Teachers at a school using a much more loosely specified design, explained that they find the design principles to be "confusing" and that they simply "go along with it." However, the chosen restructuring design had only a marginal impact on their classroom practice. While almost all principals were content with the design they had chosen and the benefits it had brought to their schools, several principals apparently were not knowledgeable about a broad array of designs. This was because some principals relied on district administrators to provide information on restructuring designs. For example, one principal reported, "Every region has a curriculum coordinator, and that person makes us aware of what's out there, what they have to assist us with, what is available for teacher training. And that's how we got the program." When a school in their region seemed in need of a boost or some rejuvenation, the district coordinators apparently visited the school and made a presentation to the staff about a particular reform design, generally the one about which they knew the most. There were also principals who researched several designs on their own. A principal who proposed and opened a new school that uses the Coalition of Essential Schools design explained, "I just went to the library, pulled all these periodicals, and just sat there and read until we hit on something that would generate some ideas." Discussing the choice of the Coalition, she reported, "We wanted something that had name recognition. I had read some of Sizer's articles, and the more we read, we just said this is sort of us. . . . It just fit who we were, so we figured we'll put it in [the proposal]." Multiple Programs and "Opportunistic Additions" The low levels of design implementation we witnessed in two sites appeared strongly related to the possibility that these schools were engaged in doing too much at once. Echoing the sentiments of her colleagues, one teacher explained, "For every new program that came our way we said, 'yes, yes, yes'. . . so the first reaction is, 'How long is this program going to last?'" As an example, one school was at least theoretically implementing the Coalition of Essential Schools, the Core Knowledge Sequence, and the Paideia program. Each of these programs has separate goals and suggested means for achieving them, and the school saw themselves as using Core Knowledge and Paideia to operationalize the principles of the Coalition. However, no professional development had been offered on how to integrate the three successfully. Some teachers struggled with this, and others did not attempt the struggle. These factors led to sparse implementation of all of the programs at this school. Schools in which we saw conflicting programs were often those in which principals, in their effort to gain resources, connections, and positive public relations for their school, engaged in "opportunistic adding" of programs. The principals selected those elements of programs that they liked, or felt best suited their schools' needs, in lieu of implementing any one design fully. However, while this led to lower levels of implementation of the design, a positive feature was that this allowed educators to participate in the professional networks provided by association with each reform design, as well as to call positive attention to their school as a place that provided multiple programs to meet students' needs. The question over the next 3 years will be, "To what extent does the presence of multiple programs at one school affect successful implementation of any one design?" Past research seems to indicate that, unless the programs are carefully integrated, each may negatively affect implementation of the others (Stringfield et al., 1997). Factors Related to Design Implementation One characteristic of schools expressing difficulty implementing restructuring designs was the perceived disconnection between design implementation and accountability measures. The schools where we observed the greatest benefits of the design for educators and students and the highest levels of implementation were those that tended to have the following elements in place: A site administration that was supportive of the design and that gave teachers what they needed to implement it successfully, be it professional development, classroom resources, or instructional support. Strong support from someone in the district office who was knowledgeable about their design and their particular school context. Frequent contact with and recognition from a representative from the design team. A full-time facilitator on-site whose sole responsibility was to support the implementation of the design. In the best case scenario, this person had previously been a teacher at the school. School organization that allowed all students (including LEP and low-achieving students) to receive benefits from the design. Each of these five elements is discussed in detail below. Site Administrators There were some schools in which principals worked very hard to make sure that the staff bought into the design and that teachers had the support they needed for successful implementation; not surprisingly, these schools generally had stronger levels of implementation. However, while all of the principals expressed support for their particular design, most principals engaged in a relatively low level of implementation monitoring. Teachers who were not supportive of their school's reform design shared the belief that, due to the principal's lack of monitoring, there would not be negative consequences for failing to implement specific design elements. Some principals were very open about the fact that some staff did not implement the design. As one principal stated, "You have teachers who love it, and you have teachers who will have nothing to do with it." This lack of accountability for elements of the reform designs did not cross over into general lack of teacher accountability, however. Teachers in Sunland, and throughout that state, are held accountable for their students' achievement through results on the annually administered Stanford Achievement Tests. A number of teachers felt that their energies were better spent on preparing students for this standardized test than in absorbing the reform elements, for which they clearly were not held accountable, into their daily classroom activities. Teachers at some schools were concerned about the feasibility of implementing the components of their design. Some of the restructuring efforts required the infusion of resources into the schools, and often teachers expressed the opinion that those resources were not available to them. A teacher at a school using the Modern Red Schoolhouse design stated, "If we had the money to buy all the computers that we need and all the teachers and aides, the whole bit . . . [then it might work, but] it is all a pipe dream. It's not realistic at all." Many teachers were also concerned about the unpaid time that they would be required to invest if the restructuring design were fully implemented. District Administrators Most principals reported that district administrators made frequent visits to the schools in their region, providing implementation support for schools by offering and sponsoring workshops and some funding for teachers to attend conferences. For example, one principal explained that the district coordinators "assist in the scheduling of design team site visits. They communicate and meet with us in terms of implementation and the training. They worked with us in terms of implementation with our ESOL (English for speakers of other languages) students." Some principals also credited district administrators for bringing specific restructuring designs to their district. "If it wasn't for [a district administrator], the Coalition of Essential Schools wouldn't be in the county. . . . She promoted it and talked to her supervisor and said, 'give these folks a chance.'" However, although the principals reported receiving support for implementation, they did not seem to feel the pressure of accountability to the district for the implementation of a design. A general perception among principals was that the district administration viewed its main responsibility as enabling schools to adopt designs and improving student achievement on standardized tests. For example, when asked about the level of district pressure in terms of test scores, a principal, echoing the sentiments of many, answered, "What? Don't you see the bazooka pointed toward my head?" An assistant principal at another school stated, "Our school will be judged on test scores. Nothing else. We could have the most wonderful programs. We could have the best of the best, but if our test scores are low, the school is not doing well [according to central administration]." An additional factor affecting district policies and activities appears to be the extensive administrative turnover that has taken place in the district over the last year. The district recently shifted to a regionally elected school board structure, adding several new members. In 1996, the newly elected school board appointed a new superintendent, who in turn reorganized the district Office of Instructional Improvement. Some school administrators expressed uncertainty regarding possible district changes vis-à-vis reform. Varying Levels of Support from Design Teams; On-Site Facilitators Another reason for the variability of implementation across schools and across designs was the extent to which national design teams supported the local school implementation efforts. Schools had various levels of training for their reform design, ranging from none to ongoing professional development. Several schools had no training from design teams; instead, they were offered extensive and ongoing training from district liaisons, which they viewed as useful. Schools in which the support from the design team resulted in the training of an on-site coordinator, who then supported implementation efforts, tended to have more positive early implementation experiences, although the level of engagement of the on-site facilitators also varied. Levels of and enthusiasm for implementation tended to be higher in schools where teachers had positive interactions with design team members, had a full-time design facilitator on site, attended national conferences focused on their designs, and received materials for implementation in a timely fashion. In some schools, enthusiasm for implementation was low because these elements were not in place. For example, a principal perceived the growth of the design on a national level and within the district as having an impact on his school's implementation efforts. He stated, "The whole program grew so big so fast that they physically can't get those materials out." Similarly, teachers at another school were upset that the design team had canceled a visit to the school. A teacher explained, "Last year it was all gung-ho, and they were coming to different workshops and all. This year, we haven't seen any of that. . . .They canceled out." The Impact of School Organization: LEP and Ability Grouping An additional correlate of 1996-97 implementation levels was that schools that separated students according to perceived ability or special needs tended to have higher degrees of implementation variability across classrooms. Specifically, students who were placed in separate classes because they were perceived to be of low ability or were classified as limited English proficient (LEP) tended to receive fewer experiences with their school's particular reform than their higher achieving, fully English-literate peers. For example, in one Audrey Cohen school, we observed that the teachers were not using Audrey Cohen materials with the low-achieving classes. By contrast, in the gifted classes, evidence of implementation of this "purpose-centered" model of education was high. This was compounded by the fact that, as in many schools, the stronger, highly experienced teachers were often assigned to the gifted classes, and weaker or less experienced teachers were assigned to the low-achieving classes. Similarly, at a Roots and Wings school, students who were placed in ESOL classes received 60 minutes of reading instruction, compared to the 90 minutes of reading that were provided to native English speakers. Ninety minutes of reading are specified by the design. Furthermore, teachers in ESOL classes reported that the pace of reading instruction (and their general movement through the program) was much slower, even in the class where students were classified as ESOL levels 3 and 4 (relatively high levels of English competency). It appeared to observers that in several schools, not all students benefited from the implementation of a reform design, particularly those students who might have benefited most from school improvement. This is an issue we will watch closely over the next 3 years. Adaptations of the Designs to Multilingual, Multicultural Contexts Related to the issue of school organization, a key element having a strong impact on the implementation and success of designs in the Sunland County schools was the degree to which schools were able to make the designs fit in the multilingual, multicultural school contexts. A major goal of this study is to identify school reform models, components of models, and implementation strategies that raise the achievement levels of culturally and linguistically diverse student groups. In doing so, we hope to better understand how the various reform designs can be successfully modified to provide high quality educational services to all students in diverse multicultural, multilingual contexts. Since none of the designs was developed specifically for use in such contexts, we believe that we may see schools making sensible adaptations to the designs to better serve their student populations. Some of the schools we observed have made thoughtful adaptations to their designs in order to suit their particular student populations. In some cases, educators have found ways in which their design helps them educate a diverse student population more effectively. For example, an educator at a school using the Comer model, which is intended to address the developmental needs of children, explained, We are [very] sensitized to the families, and we are learning more about their culture and understanding more and more why the parents and the children act the way they do. We have a significant Haitian population, and their views on education are very different from our views or the Hispanic views. Another educator at the school reiterated, "We are more aware of the family because of Comer." A teacher in a Roots and Wings school thought that the program was helpful for LEP students, because it gives students many opportunities to communicate with each other and with the teacher. Another teacher added, "In this program it is okay to have a classmate help you, and I like that. ESOL kids need that." However, teachers in another Roots and Wings school complained that the specificity of the program prohibited them from adapting the design to suit the diverse student groups in their classroom. Echoing the sentiments of a number of teachers, one first grade teacher explained, "You want to know if we have modified it to suit the kids in our class? Well, we're not allowed to do that." Her colleague added, "It's not in the script." These teachers felt that if they made even sensible adaptations to the program, they would receive low ratings when the design team conducted an implementation check. Still, teachers said that they had made some changes to the program, such as rewriting lessons or extending the time allotted, in order to shelter instruction for LEP students. In addition, Sunland County administrators decided against the use of Éxito para Todos, the Spanish Bilingual version of the Success for All/Roots and Wings reading program, because it requires that reading and language arts instruction be provided half in English, half in Spanish. However, individual schools are making modifications, and the district is working with the Roots and Wings design team to develop customized materials that comply with district rules and still meet the needs of the diverse linguistic groups that their schools serve. For example, a principal explained, Our teachers are writing their own Treasure Hunts. But they also modify it because ESOL is an oral language development program, and Roots and Wings is a reading program. Certain parts of it you can do, and other parts you have to say, "Okay, I have to put this down and go with the ESOL strategies." When a child moves in and out of the classroom to the resource room we want there to be some continuity. He explained that an adaptation his school made to suit Roots and Wings to the needs of LEP students was in terms of pacing, "When you're supposed to do a lesson in 3 days, it could take up to 5." As stated above, the Roots and Wings design team is working with the district to develop an adapted program that will help schools better meet the needs of their LEP students and the requirements of state legislation. In addition, educators at an Audrey Cohen school also applauded the design team for allowing them flexibility in the design implementation for their LEP students. A school-level facilitator called an Audrey Cohen design representative to ask whether students could do the writing exercises in their home language. The College representative said yes, emphatically stating that they wanted children to be as comfortable as possible with the Audrey Cohen College System. Some of the adaptations schools had made to suit their diverse student populations were not adaptations per se but were instead features that one might see with or without a restructuring design in place. For example, a teacher reported, "During the holidays we do a lot of things. We choose a country. Each classroom has a country, and we learn about the foods, music, and traditions." This was not coordinated with, nor seen as related to, the "tone of decency" that the school was working on as part of its Coalition of Essential Schools implementation. A Match That Is Succeeding: Wild Cypress Elementary Bilingual School Next, we present a short description of a school where the implementation of the Core Knowledge Sequence and the instruction for limited English proficient (LEP) students seemed to be very successfully coordinated. The question of whether a fully bilingual education program can produce measured academic excellence has been answered in the affirmative at Wild Cypress Elementary Bilingual School. The school also appears to be a solid example of the advisability of carefully matching an externally developed reform with local practical strengths and needs before moving to reform implementation. Background Built in the 1930s as a Work Projects Administration project, Wild Cypress's physical facility has managed to age without losing some semblance of grace. The original building is substantially overcrowded. The district estimates that, even with 10 portable classrooms, Wild Cypress is operating at more than 20% above capacity. Yet, the feel of the school is warm and welcoming. Wild Cypress Elementary Bilingual School is a neighborhood school that serves almost 1,200 students pre-kindergarten through Grade 5. Seventy-two percent of the students receive free or reduced-price lunches. This poverty rate is more than double the state and national averages. Ninety percent of the students are Hispanic, 7% are White, 1.5% are Black (including several Haitian immigrants), and 1% are members of other groups, largely of Asian origin. Over one third of the students are classified as limited English proficient. While most of the LEP students speak Spanish as their native language, a small number of students speak other languages, including Haitian Creole, French, or one of several Asian languages. Responding to a large influx of Cuban refugees in the 1960s, Wild Cypress became a fully bilingual program over 3 decades ago. The phrase, "We're bilingual and bi-literate," permeates the school's promotional materials. First Year Observations Wild Cypress Elementary Bilingual School operates as a two-way bilingual school. A clear strength of the school is its insistence on providing a top-notch English- and Spanish-language education to all students, regardless of their home language. Each academic quarter, students rotate courses that they will take in English or Spanish. For example, students who take mathematics in English and science in Spanish during the first quarter will continue their mathematics lessons in Spanish and their science units in English during the second quarter. All students, regardless of home language, spend 60% of each school day in English instruction and 40% in Spanish-language instruction. For decades, the school has worked hard and successfully to provide high-quality basic skills instruction to all students. In the 1990s, the faculty and administration became increasingly concerned with the limits of the school district's competency-based curriculum and sought additional enrichment for all students. This search led to the schoolwide adoption of the Core Knowledge Sequence (Core Knowledge Foundation, 1995). Core Knowledge is not designed to replace all of a school's previous curriculum but to replace parts of it and to enhance the overall curriculum. In this regard, the choice was a particularly astute one for Wild Cypress and the multiethnic, multinational community that it serves. Core Knowledge allowed the school to build on its traditional bilingual academic strengths, while adding literature, history, and arts that reflected an appreciation of the history and cultures of the world in a clearly spiraling structure. For example, fourth graders read Ashanti Tu Zulu, Robin Hood, and, in Spanish, Cuentos para Chicos y Grandes. The school did not stop with a general philosophy and a set of specific curricular offerings. Rather, the faculty has worked together within grades to detail exactly which components would be presented in what order and in which language. For example, in fourth grade, the human circulatory system is studied in English, followed by a unit on the human skeletal system conducted in Spanish. During the same semester, students study world geography, map skills, Don Quixote in Spanish, and The Legend of King Arthur and Robin Hood in English. The striking level of integration of content across languages was confirmed during classroom observations. All students interviewed, regardless of background, appeared capable of answering questions in English or Spanish, and some were also capable of answering questions in a third language, such as Cantonese or Haitian Creole. Significantly, the Core Knowledge curriculum was carefully chosen by the faculty to build on the school's strengths and to help the faculty and students focus on the substantial contributions of all nations to the world's rich history and diverse cultural fabric. School staff sought a reform that could address specific concerns. They examined the options, and they chose based on their perceptions of their strengths and current limitations. The teachers gradually began implementing Core Knowledge 3 years ago, one unit at a time. The principal explained, "The first year we said, 'let's emphasize social studies.' So at each grade level we picked a [Core Knowledge] theme to do the first semester. In third grade, they did the explorers. In the fifth grade, they did Ancient Rome." Teachers added on units one by one, purchasing materials for each as they went along. In Year 3 of implementation, teachers reported teaching almost all of the units in the Core Sequence, and their support for the program seems very strong. In keeping with Senge's (1990) model of a learning organization, the faculty of Wild Cypress began with a shared vision of exemplary bilingual education for all students, engaged in daily and weekly team learning to find and implement a reform, sought and largely achieved personal mastery of new content and skills, and consistently engaged in systemic thinking regarding how to best blend and integrate the various content and instructional approaches. A carefully chosen reform plus a thoughtful, persistent, thorough implementation has resulted in consistently high achievement on standardized tests for a large number of potentially at-risk, multilingual students. That all of this has been achieved in a warm and welcoming atmosphere simply adds to the sense of remarkable success of Core Knowledge at Wild Cypress. The Coordination of ESOL Programs and Reform Designs Wild Cypress Elementary School represents an apparently successful attempt to integrate a two-way bilingual program with a reform design, in this case, Core Knowledge. However, that school was the exception in that it had received a waiver from the state's mandated program for LEP students. More typically, the interface between programs for LEP students and the reform designs was not as smooth. One school reduced the time required for design element implementation in ESOL classrooms. Other schools had ESOL classrooms that reflected design implementation for only part of the day. For example, in one Modern Red Schoolhouse program, teachers told us that because they were pulled out for home language instruction for up to 150 minutes per week, the ESOL level I and II students missed the content that reflected the curriculum espoused by the design. Teachers at this school also talked about feeling constrained by local regulations regarding ESOL instruction, arguing that these laws inhibited the implementation of their restructuring design: "The whole idea of the Modern Red Schoolhouse is that you can play around with your classes. And Sunland County said, 'No, you can't do this.' It was kind of squelched." The district regulations mandate that ESOL level I and II students receive instruction in English, except for 150 minutes per week when they receive basic subject area instruction in their home languages (if those languages are Spanish or Haitian Creole). In languages other than Haitian Creole and Spanish, the instructional time in the basic subject areas is provided once or twice a week in an individual or small group tutorial context.3 Not all schools appeared to be able to comply with this home language instruction rule, depending on the availability of instructors in a student's home language and the number of students in a given school who needed to be served. For example, at one school, there were only a few students who spoke Urdu in a class of native Spanish speakers. We observed them learning in Spanish during their home language instruction time instead of Urdu. The district faces a perennial shortage of certified Urdu instructors, and visits to the school by these instructors generally occur once every 2 weeks. In addition to these classroom activities, LEP students receive 1 hour each day of ESOL instruction, which is typically oral language skill instruction. Typically, students who are classified as ESOL levels I and II (non-independent or intermediate A) are placed in separate ESOL classrooms, if there are sufficient numbers of LEP students in a school. Conversely, if there are not sufficient numbers, LEP students may receive pull-out ESOL and home language instruction. Students are eligible to test out of ESOL classes when they reach ESOL level V. The district office reported that the average length of time in the ESOL program in the district was 2.9 years. All of the 13 schools were making significant efforts of one sort or another to improve schooling for culturally and linguistically diverse students. Our first year data-gathering efforts were focused on principals' and teachers' experiences with implementation and on classroom events. In Years 2-4, we will invest significant energy in gathering data on school contexts and on the schools' efforts to adapt designs to the needs of their specific populations. We expect that a full understanding of this critical area of our study will take several more years of data collection and analysis. Conclusions and Implications Are there clear, readily apparent ways in which the multicultural, multilingual texture of a community affects successful implementation of diverse reforms? The fact that several fully multicultural schools have been able to achieve at least some level of implementation of each of the designs does indicate that the use of diverse strategies is possible in multilingual, multicultural contexts. More finely tuned analyses of specific subcomponents of the various reforms interacting with the various communities await CREDE's longitudinal data gathering and analysis. Are there readily apparent ways in which the multicultural, multilingual texture of a school community facilitates successful implementation of the diverse reforms? Again, a definitive answer awaits much more data collection; however, two interesting, somewhat unanticipated points appear to be emerging. First, in an era of public accountability and test scores published school-by-school in newspapers, schools that serve large numbers of newly immigrated students and other second language learners typically begin the process with low mean scores. Low scores bring calls for change, and these calls may make educators' movement to reform somewhat easier. In contrast to schools in upper- and middle-income communities, inner-city schools may be more willing to seek alternative school structuring strategies. Several of our schools took on a restructuring design specifically out of concern that the traditional systems were not addressing the needs of their current populations. Second, in at least one case, and in future years we will be looking more closely for others, the multilingual/multinational/multicultural nature of the community itself made the adoption of a reform that has rich roots in diverse nations' and peoples' histories (e.g., Core Knowledge) particularly attractive within the school and the community. At Wild Cypress, having young children study specific aspects of African, South American, Asian, and European histories and cultures has become a method for explicitly valuing all children's (and their parents') heritages, and by extension, all children. Theoretically, other school restructuring designs have similar potential, and in Years 2-4 of this study, we will explicitly examine the multicultural, multilingual components of each reform effort. Based on our preliminary findings, we have distilled the following implications of what might make for successful restructuring in multilingual, multicultural contexts. We consider these implications to be hypotheses that we will test over the next 3 years. Refinements in the Choice Process If a model of districtwide reform, where schools are provided with an array of choices is going to work, the process needs to be better refined so that schools can make educated choices. Having a more coordinated and broader-based program for disseminating information about design options might alleviate the implementation problems we noted. Perhaps visits to other schools and meetings with staff already using the model should be more strongly encouraged. In addition, perhaps design teams need to be clearer about the elements that will be implemented. School personnel need to be fully aware of what is involved in design implementation. Multidimensional Support and Leadership For school restructuring to be successful in any context, and particularly in schools that serve diverse populations, multidimensional support and leadership are required from design teams, district personnel, and school site educators. For example, design teams must ensure that schools have the training and materials they require for implementation. Moreover, the district needs to provide schools with ongoing monitoring of implementation progress and other implementation support and resources as needed. This may mean funding a multiyear, whole school professional development program for schools engaged in restructuring. In addition, state departments of education need to consider how their policies regarding student assessment and the instruction of linguistically diverse students can be aligned to support school restructuring. Finally, school site educators must exhibit the leadership and commitment to implement the design, making thoughtful adaptations to suit their local contexts. High Standards for All Students Equity is an important goal of all of the school restructuring designs being implemented in Sunland County. In order to meet this goal, educators must ensure that all students receive the benefits of the reform designs and that high academic standards are universally maintained. The highest achieving students are often likely to get more exposure to the reform than the low-achieving (often at-risk and LEP) students. This requires more ongoing support from the design team and district. Forethought and Sensitivity The vast majority of the school restructuring designs we are studying in SCPS can be implemented, with adaptations, in urban schools serving diverse populations. However, successful implementation likely requires sensitivity and adaptability (without academic compromise) on the part of the design developers, local policymakers, and educators in schools. Adaptations to designs will be required. Moreover, some implementation efforts may require relief from state and local policies in order to be successful. In sum, since prior studies (Crandall & Loucks, 1983; Stringfield et al., 1997) have indicated substantial advantages to the use of externally developed designs, every effort should be made by districts and design teams to assist school educators in choosing the right design for their school, in adapting the design to their local context, and in implementing that design. After all, it is in America's multicultural, multilingual districts that students are in the greatest need of quality educational programs that offer them a rich set of learning experiences. This report is the first in a series of reports on scaling up school restructuring designs in Sunland County. Data gathering in this study is ongoing. Over the next several years, we will continue to follow the 13 schools in Sunland County and gather more school context data. In addition, we will continue to gather data on the central administration through detailed interviews and archival data sources. We will also conduct detailed interviews with representatives of each of the design teams, asking specific questions about their perceptions regarding local adaptations. Finally, we will analyze multiyear outcome data and report on the effects of the various reforms. Notes 1The names of the county, city, school district, and individual schools discussed in this paper are pseudonyms. 2Newmann and Wehlage (1995) characterize authentic pedagogy as involving authentic instruction and authentic assessment. Authentic instruction (a) involves students in higher order thinking; (b) addresses a topic with enough thoroughness to produce complex understandings; (c) engages students in extended and substantive conversational exchange; and (d) helps students connect substantive knowledge with public problems or personal experiences. Instruction that meets these criteria can be either teacher centered or student centered. Following along these lines, authentic assessment tasks involve the construction of knowledge, disciplined inquiry, and value beyond school. 3We are currently researching the specifics of this legislation and will report on its practical implications in future reports. References   Aikin, W. M. (1942). The story of The Eight Year Study. New York:  Harper.   Banks, J., & Banks, C. (Eds.). (1995). Handbook of research on multicultural education. New York:  Simon & Schuster Macmillan.   Berman, P., & McLaughlin, M. (1978). Federal programs supporting educational change:  Vol. 8. Implementing and sustaining innovations. Santa Monica, CA: Rand.   Bodilly, S. (1996). Lessons for New American Schools Development Corporation's demonstration phase. Santa Monica, CA:  Rand.   Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models. Thousand Oaks, CA: Sage.   Cohen, A., & Jordan, J. (1996). The Audrey Cohen College System of Education: Purpose-centered education. In S. Stringfield, S. M. Ross, & L. Smith (Eds.), Bold plans for school restructuring:  The New American Schools designs. Mahwah, NJ:  Lawrence Erlbaum.   Comer, J. (1980). School power:  Implications of an intervention project. New York:  Free Press.   Comer, J., & Haynes, N. (1996). Rallying the whole village:  The Comer process. New York:  Columbia University, Teachers College.   Core Knowledge Foundation. (1995). Core Knowledge sequence. Charlottesville, VA:  Author.   Crandall, D. P., & Loucks, S. F. (1983). Executive summary of the study of dissemination efforts supporting school improvement. Andover, MA:  The NETWORK.   Guba, E., & Lincoln, Y. (1981). Effective evaluation. San Francisco:  Jossey-Bass.   Heady, R., & Kilgore, S. (1996). The Modern Red Schoolhouse. In S. Stringfield, S.M. Ross, & L. Smith (Eds.), Bold plans for school restructuring:  The New American Schools designs. Mahwah, NJ:  Lawrence Erlbaum.   Herman, R., & Stringfield, S. (1997). Ten promising programs for educating disadvantaged students:  Evidence of impact. Arlington, VA:  Educational Research Service.   Hirsch, E. D., Jr. (1987). Cultural literacy:  What every American needs to know. Boston:  Houghton Mifflin.   Hirsch, E. D., Jr. (1993). What your 5th grader needs to know:  Fundamentals of a good fifth grade education. New York:  Dell.   Kearns, D. T., & Anderson, J. L. (1996). Sharing the vision:  Creating new American schools. In S. Stringfield, S. M. Ross, & L. Smith (Eds.), Bold plans for school restructuring:  The New American Schools designs. Mahwah, NJ:  Lawrence Erlbaum.   King, J. A. (1994). Meeting the educational needs of at-risk students:  A cost analysis of three models. Educational Evaluation and Policy Analysis, 16(1), 1-19.   Krueger, R. A. (1994). Focus groups. Thousand Oaks, CA:  Sage.   McLaughlin, M. (1990). The Rand change agent study revisited:  Macro perspectives and micro realities. Educational Researcher, 19(9), 11-16.   Miles, M., & Huberman, M. (1994). Qualitative data analysis (2nd ed.). Thousand Oaks, CA:  Sage.   Murphy, J., & Hallinger, P. (1993). Restructuring schools. Newbury Park, CA:  Sage.   Natriello, G., McDill, E., & Pallas, A. (1990). Schooling disadvantaged children:  Racing against catastrophe. New York:  Columbia University, Teachers College.   Newmann, F., & Wehlage, G. (1995). Successful school restructuring. A report to the public and educators by the Center on Organization and Restructuring of Schools. Madison, WI:  Wisconsin Center for Education Research.   Ross, S. M., Smith, L. J., & Casey, J. P. (1997). Preventing early school failure:  Impacts of Success for All on standardized test outcomes, minority group performance, and school effectiveness. Journal of Education for Students Placed At Risk, 2 (1), 29-54.   Senge, P. (1990). The fifth discipline. New York:  Doubleday.   Sizer, T. R. (1984). Horace's compromise. Boston:  Houghton Mifflin.   Sizer, T. R. (1992). Horace's school:  Redesigning the American high school. New York:  Houghton Mifflin.   Sizer, T. R. (1996). Horace's hope. New York:  Houghton Mifflin.   Slavin, R., Madden, N., Karweit, N., Dolan, L., & Wasik, B. (1992). Success for All:  A relentless approach to prevention and early intervention in elementary schools. Arlington, VA:  Educational Research Service.   Slavin, R.E., Madden, N.A., & Wasik, B.A. (1996). Roots and Wings:  Universal excellence in elementary education. In S. Stringfield, S. M. Ross, & L. Smith (Eds.), Bold plans for school restructuring:  The New American Schools designs. Mahwah, NJ:  Lawrence Erlbaum.   Stringfield, S., Millsap, M., Yoder, N., Schaffer, E., Nesselrodt, P., Gamse, B., Brigham, N., Moss, M., Herman, R., & Bedinger, S. (1997). Special Strategies Studies final report. Washington, DC:  U.S. Department of Education.   Stringfield, S., & Ross, S. M. (1997). A "reflection" at mile three of a marathon:  The Memphis Restructuring Initiative in mid-stride. School Effectiveness and School Improvement, 8(1), 151-161.   Stringfield, S., Ross, S., & Smith, L. (Eds). (1996). Bold plans for school restructuring:  The New American Schools Development Corporation designs. Mahwah, NJ:  Lawrence Erlbaum.   Tharp, R., & Gallimore, R. (1988). Rousing minds to life. New York:  Cambridge University Press.   Tyack, D., & Cuban, L. (1995). Tinkering toward Utopia:  A century of public school reform. Cambridge, MA:  Harvard University Press.   Yin, R. (1989). Case study methodology. Beverly Hills, CA:  Sage. Appendix School Restructuring Designs Audrey Cohen College System Of Education Founded in 1964, Audrey Cohen College was invited in 1983 to spread its "purpose-centered system of education" to the New York City public school system at both the elementary and secondary levels (Cohen & Jordan, 1996). The Audrey Cohen System of Education was selected as a New American Schools (NAS) design; in late 1995, 21 schools were using the system to educate some 20,000 children. The Audrey Cohen curriculum is not organized around traditional academic disciplines but around five "dimensions of effective learning and action": purpose, values and ethics, self and others, systems, and skills. Focusing on their purpose, Audrey Cohen students are encouraged to work with community members, in and out of the school, to plan, prepare, carry out, and evaluate a project. Younger children work together, older students work on individual or group projects, and teachers facilitate their students' work. Each semester, students work on a defined goal, showing their accomplishment of a purpose by performing a "constructive action," such as the creation of a videotape on their community's environmental situation. The values and ethics dimension explores the costs, benefits, and conflicts associated with their area of interest, and the systems dimension explores the systems of which they are a part and which they hope to affect. Simultaneously, the self and others dimension explores how students relate to others, and the skills dimension builds their mathematical, technical, linguistic, and physical abilities. Audrey Cohen schools call for teachers to develop certain skills and qualities: teamwork, comfort with the switch from disciplines to dimensions, a whole new web of contacts with community members, and a willingness to shift their role from directing to facilitating learning. Newer Audrey Cohen schools are able to tap into the community networks already established and have less initial work than the first Audrey Cohen school in any given community. Audrey Cohen College requires that each elementary school hire a school resource specialist for the first 1 to 3 years of operation. The tasks of these resource specialists include collecting materials, coaching teachers, and maintaining contact with community members and the design team. School districts pay licensing and participation fees for the program. In return, the school receives staff development, as well as guidance on standards and assessment. Audrey Cohen College 345 Hudson Street New York, NY 10014 (212) 343-1234 Modern Red Schoolhouse The NAS design called Modern Red Schoolhouse aims to extend and deepen the educational assets embodied in the classic "little red schoolhouse" (Heady & Kilgore, 1996). Six tenets embody the goals of this educational design: All students can reach high educational standards, in different time frames and ways. The school's work should introduce students to principles of democratic government and pluralism. The curriculum covers the basic disciplines as well as certain work-related skills. Although principals may be appointed centrally, all decisions about instruction and the use of resources should be based at the school. Schools should combine flexibility and accountability. Advanced technology is crucial to a modern education. Each teacher in the school should have a computer, and a Modern Red Schoolhouse should maintain a target ratio of one computer for every six students, although their concentration may vary with student age level. Students and staff should choose their schools. After considering any applicable court orders, geographical proximity, and sibling preference, selection for a Modern Red Schoolhouse should be random. Education in a Modern Red Schoolhouse is framed in an Individual Educational Compact between a student and his or her parents and teacher. The compact spells out the student's goals, the responsibilities of all parties, and any special services that the student may need. The Modern Red Schoolhouse groups students into primary, intermediate, and upper divisions, ending at the traditional Grades 4, 8, and 12, respectively. Assessments at these three stages, based on the College Board Advanced Placement exams, test students in mathematics, science, English, history, and geography, using traditional tests as well as oral reports and projects. Schedules for students and teachers vary, and curricula are divided into "Hudson units," reflecting academic accomplishment rather than seat time. During the elementary grades, students in Grades K-6 in Modern Red Schoolhouse schools are taught the Core Knowledge Sequence for one half of each day. Developed by the Core Knowledge Foundation, the Core Knowledge Sequence provides a planned progression of specific knowledge in language arts, history, geography, math, science, and fine arts, designed so that students build on knowledge from year to year in Grades K-6. Core Knowledge is based on the premise that "only by specifying the knowledge that all children should share can we guarantee equal access to that knowledge" (Core Knowledge Foundation, 1995, p. 3). Modern Red Schoolhouse Hudson Institute 5395 Emerson Way Indianapolis, IN 46226 (317) 545-1000 Roots and Wings An NAS design, Roots and Wings extends and broadens a successful reading program, Success for All, developed at the Johns Hopkins University, (Slavin, Madden, & Wasik, 1996). Roots and Wings has two goals for all students: (1) achievement of world-class standards in reading, writing, mathematics, science, history, and geography; and (2) development of skills such as problem solving and self-reflection. Roots and Wings works to keep all children achieving without tracking and is committed to the research-supported concept that learning happens best with one-to-one tutoring. The Roots and Wings curriculum has several parts. The reading component is based on that of Success for All; for younger children it is called Reading Roots and for older children, Reading Wings. The writing program is called Writing from the Heart and, later, Writing Wings. The mathematics program, Math Wings, is based on work done by the National Council of Teachers of Mathematics. Finally, WorldLab consists of units based on themes or topics that integrate science, history, geography, and writing. Roots and Wings requires a commitment from schools and districts that are beginning the program. Support from district administrators and the principal and a secret vote of support from at least 80% of teachers are required. The main program elements, without which Roots and Wings cannot be implemented, are the following: a ratio of 1 certified one-to-one tutor for every 50 Title I-eligible students a full-time facilitator a family support team, and an attendance team if attendance is low full-day kindergarten and/or half-day pre-kindergarten use of the main curricular components listed above commitment to group reading classes according to reading level, not grade level library improvement to support the reading program commitment to reduce use of special education classes and retention building advisory committee scheduling adjustments staff development grade-level team meetings adequate staff and funding from the district These program elements combine to form a whole school restructuring program. Roots and Wings CRESPAR 3505 N. Charles Street Baltimore, MD 21218 (410) 516-8896 Comer School Development Program Dr. James Comer and his staff at the Yale Child Study Center have developed and implemented a schoolwide restructuring process, the School Development Program (SDP) (Comer & Haynes, 1996), to address the needs of the whole child. Incorporated into the program are three mechanisms (the governance and management team, student support team, and parent program), three guiding principles (collaboration, consensus, and no-fault problem solving), and three operations (comprehensive school plan, staff development, and periodic assessment and modification). Central to this model is a school management and governance team composed of the principal, teachers, parents, a mental health specialist, and support staff. This team develops the school's "master plan." Although on-site professionals and community people determine the specific shape of the program, the instructional program typically includes a Focus Program, a small-group pull-out tutorial provided at least three times per week to students who are at least a year behind grade level, and a Discovery Room to entice or draw out troubled learners (King, 1994). The SDP is essentially content-free and, in principle, could be adapted to diverse local curricula. Specific additional reforms, such as cooperative learning, Reading Recovery, and Success for All, could all, in theory, be implemented at a school under the SDP umbrella. School Development Program Yale Child Study Center 230 South Frontage Road P. O. Box 7900 New Haven, CT 06520-7900 (203) 785-2548 Core Knowledge Sequence Developed by the Core Knowledge Foundation, the Core Knowledge Sequence provides a planned progression of specific knowledge, designed so that students build on knowledge from year to year in Grades K-6 (e.g., Hirsch, 1993). Following the premise that individuals must find common ground in order to interact, Core Knowledge's intent is to provide disadvantaged students equal access to knowledge and thus to a base of common experience enabling them to fit in and communicate with other members of society. The Core Knowledge Sequence represents the first major effort to specify a common core curriculum for children in American schools. The most distinguishing feature of the Core Knowledge Sequence is its content specificity. Intended as 50% of a school's curriculum, the Core Knowledge Sequence provides grade-level specific topics to be covered in history, geography, mathematics, science, language arts, and fine arts. Material in subsequent years relies on knowledge and skills learned in previous years; following the Core Knowledge Sequence guarantees that all students are equally prepared for increasingly complex material. In addition, the progressive nature of the sequence, as well as the degree of detail regarding content, helps prevent repetition of material and gaps in knowledge that sometimes occur between school years when teachers develop curricula autonomously. Elements from the Core Knowledge curriculum are carefully chosen to reflect an "inclusive multiculturalism" (Core Knowledge Foundation, 1995), carefully balancing lessons that increase knowledge of mainstream culture (such as American History or science) with others that encourage students to respect and celebrate cultural diversity (such as units on modern-day Egypt or ancient China). While the Core Knowledge Sequence specifies content, it does not specify classroom process or implementation procedures. It does not provide guidelines for how to teach the material, and it provides only general guidelines about how a school might implement the sequence (Core Knowledge Foundation, 1995). Core Knowledge Foundation 2012-B Morton Drive Charlottesville, VA 22903 (804) 977-0021 Coalition of Essential Schools Dr. Theodore Sizer's (1984) Horace's Compromise, a fictional account of one high school's efforts to restructure, led to the creation of the Coalition of Essential Schools (CES). CES is a widely implemented schoolwide restructuring project. As of 1996, CES listed over 100 schools as either "Exploring," "Planning," or "Member" schools. At the heart of this design is the belief that a school should be a place where decency prevails, and where social and professional relationships are characterized by tolerance, generosity, and fairness. The CES philosophy proposes an ideology about schooling and learning that places "personalization" high on the list of imperatives. CES advocates a total restructuring of traditional school organizations, practices, and beliefs. CES suggests Nine Common Principles that should be used as a framework for schools to provide personalized education to all students: Schools should have an intellectual focus. Goals should be simple and universal. Teaching and learning should be personalized for each individual student. The guiding metaphor should be "student as worker, teacher as coach." Diplomas should be awarded for demonstration of mastery. Adults should unanxiously express high expectations of students. Administrators and teachers are generalists first, then specialists. The maximum student/teacher ratio should be 80:1 per day. CES should be implemented with a per-pupil cost increase of no more than 10%. The principles are deliberately ambiguous to allow program adaptation to local requirements. Practical applications of the principles recommended by CES include reduced class sizes; curriculum based on interdisciplinary questions; school policies based on a system of trust and shared values, as well as belief and expectation that students can succeed; teachers functioning in several roles (such as counselor, advisor, and manager) and shifting from didactic methods to the facilitation of independent learning; and modified scheduling, including block scheduling and double periods, to spend more time on fewer subjects and extended time in school. (Sizer recommends an extra hour per day and an extra 6 weeks per year.) An additional program, RE:Learning (developed in 1988 between CES and the Education Commission of the States) adds six additional goals supporting the CES principles on the state level. However, the state in which our longitudinal sites are located is not a RE:Learning state. Coalition of Essential Schools Brown University Box 1969 Providence, RI 02912 (401) 863-3384 The work reported herein and the editing and production of this report were supported under the Educational Research and Development Centers Program, Cooperative Agreement Number R306A60001-96, as administered by the Office of Educational Research and Improvement, U.S. Department of Education. However, the contents do not necessarily represent the positions or policies of the National Institute on the Education of At-Risk Students, the Office of Educational Research and Improvement, or the U.S. Department of Education, and you should not assume endorsement by the Federal Government. Scaling Up School Restructuring in Multicultural, Multilingual Contexts: Early Observations from Sunland County, Research Report 2. © 1998 by the Center for Research on Education, Diversity & Excellence, Santa Cruz, CA and Washington, DC.