Comprehensive School Reform for High-Poverty Schools

PROGRAMS DEVELOPED IN 1980S AND 1990S

The movement toward schoolwide projects in Title I schools encouraged a flowering of whole-school innovations in high-poverty schools. Much of this innovation has taken place in individual schools and districts, without any intention to make changes on a broader scale, but there arose in the late 1980s and early 1990s a set of programs, explicitly designed by university researchers and non-profit organizations not only to improve whole schools but also to be replicable, to eventually offer large numbers of schools well-developed alternatives to current practices. These programs came to be known as Comprehensive School Reform (CSR). CSR sees the school as the primary unit of change in education. It seeks to implant effective practices in all of the central areas of school functioning most likely to affect student achievement: Curriculum, instruction, assessment, grouping, accommodations for struggling students, parent and community involvement, school organization, and professional development (see Stringfield, Ross, & Smith, 1996; ERS, 1998; CSRQ, 2006a, 2006b; Borman, Hewes, Overman, & Brown, 2003). In 1998 the U.S. Department of Education defined comprehensive school reform as innovative programs that include all of the following elements:

Table 1ILLUSTRATION BY GGS INFORMATION SERVICES. CENGAGE LEARNING, GALE.

1. Coordination of resources: The program identifies how all resources (federal/state/local/private) available to the school will be utilized to coordinate services to support and sustain the school reform effort;
2. Effective, research-based methods and strategies: A comprehensive school reform program employs innovative strategies and proven methods for student learning, teaching, and school management that are based on reliable research and effective practices, and have been replicated successfully in schools with diverse characteristics;
3. Comprehensive design with aligned components: The program has a comprehensive design for effective school functioning, including instruction, assessment, classroom management, professional development, parental involvement, and school management, that aligns the school's curriculum, technology, and professional development into a schoolwide reform plan designed to enable all students to meet challenging state content and performance standards and addresses needs identified through a school needs assessment;
4. Professional development: The program provides high-quality and continuous teacher and staff professional development and training;
5. Measurable goals and benchmarks: A comprehensive school reform program has measurable goals for student performance tied to the state's challenging content and student performance standards, as those standards are implemented, and benchmarks for meeting those goals;
6. Support within the school: The program is supported by school faculty, administrators, and staff;
7. Parental and community involvement: The program provides for the meaningful involvement of parents and the local community in planning and implementing school improvement activities;
8. External technical support and assistance: A comprehensive reform program utilizes high-quality external technical support and assistance from a comprehensive school reform entity (which may be a university) with experience or expertise in school-wide reform and improvement;
9. Evaluation strategies: The program includes a plan for the evaluation of the implementation of school reforms and the student results achieved.

Ideally, a comprehensive school reform model is one in which each of the elements is carefully integrated around a shared conception of how students will learn and develop. Most CSR models require that staff members vote to adopt the model, and most require a super-majority in favor (say, 80%). The idea is to engage the energies and enthusiasm of a given school staff around a common vision and a common set of strategies, but not to ask the staff to completely design its own reform model. Comprehensive school reform designs are provided by organizations (mostly nonprofits) that provide professional development, teacher and student materials, and perhaps most importantly a network of like-minded schools around the country that share similar visions and support one another's efforts.

In schools implementing CSR designs, teachers have colleagues who are working toward similar objectives, sharing a vision and a language to describe that vision and sharing practical strategies for achieving the vision. Almost all CSR models include a facilitator or coach within the school who visits teachers' classes, organizes opportunities for teachers to work with one another, facilitates discussions about data, student work, classroom teaching practices, and other elements, ensures coordination among program elements, and acts as a communication link between the principal and the teachers. Comprehensive school reform takes the view that genuine, lasting change takes place in supportive groups of like-minded professionals, and that schools are capable of establishing norms of practice and expectations for continuous improvement that would be difficult to establish on a teacher by teacher basis.

Comprehensive school reform grew explosively in the 1990s, helped first by the 1991 appearance of the New American Schools Development Corporation (NASDC), a coalition of large corporations that funded the development and scale-up of CSR models (Kearns & Anderson, 1996). A second major boost came from the 1997 Obey-Porter Comprehensive School Reform legislation, which made grants available to high-poverty schools to adopt “proven, comprehensive” CSR models. By 2001 there were an estimated 6,000 schools, mostly high-poverty elementary schools, using CSR models, with or without Obey-Porter funding. A lack of support for CSR in the second administration of President George W. Bush ended Obey-Porter and slowed the growth of CSR, but there are still thousands of schools implementing CSR programs. Further, perhaps no education

Table 2ILLUSTRATION BY GGS INFORMATION SERVICES. CENGAGE LEARNING, GALE.

reform initiative in history has been as thoroughly researched as CSR, and this research continues.

RESEARCH ON COMPREHENSIVE SCHOOL REFORM PROGRAMS

Comprehensive school reform models have been extensively evaluated in large-scale quantitative as well as qualitative studies. A review of experimental research on comprehensive school reform models was published by Borman, Hewes, Overman, and Brown (2003), who categorized programs according to the numbers of well-designed experiments on each and the consistency of positive achievement effects. A simplified adaptation of their main results appears in Table 1. Reviews using somewhat different procedures were carried out by the Comprehensive School Reform Quality Center (CSRQ, 2006a, 2006b) at the American Institutes for Research. The CSRQ reviews, summarized in Tables 2 and 3, emphasized the number of “conclusive” studies done on each program and the proportion of significantly positive findings. The following sections discuss research on some of the most prominent of the CSR models.

Success for All. Success for All (Slavin & Madden, 2001) is the most widely used and extensively evaluated of the CSR models. It provides schools with specific curriculum materials and extensive professional development in reading, writing, and language arts, along with detailed assessment, cross-grade grouping strategies, within-school facilitators, and other school organization elements. The program gives one-to-one tutoring to primary-grades children who are struggling in reading, and extensive outreach to parents. It provides detailed teacher's manuals and about 26 person-days of on-site professional development to enable schools to engage in a substantial retooling process. Originally focused on elementary school, prekindergarten to grade 6, Success for All now has a middle school (grades 68) program as well (Chamberlain et al., 2007). Programs in mathematics, science, and social studies were also developed, and the term Roots & Wings was used todescribe schools using all of these elements (Slavin, Madden, Dolan, & Wasik, 1994). However, most schools, including many of those categorized as Roots & Wings in the Borman et al. (2003) review, use only the reading program, and the Roots & Wings term is no longer used. Research on Success for All and Roots & Wings are combined for discussion in this chapter.

Borman and colleagues (2003) identified a total of 46 experimental-control comparisons evaluating Success for All, 31 of which were carried out by third-party investigators. A mean effect size of +0.20 (combining Success for All and Roots & Wings) was obtained across all studies and measures. A longitudinal study by Borman & Hewes (2003) found that students who had been in Success for All elementary schools were, by eighth grade, still reading significantly better than former control group students and were about half as likely to have been retained or assigned to special education.

Comprehensive School Reform Quality Center (CSRQ) (2006a) rated the strength of evidence for the Success for All elementary program as “moderately strong,” the highest rating given to any program (one other, Direct Instruction, also received this rating). A total of 34 studies were rated as “conclusive.” CSRQ (2006b) rated the evidence for the Success for All middle school as “moderate,” with two conclusive studies.

Since the review by Borman and colleagues, a number of additional studies of Success for All have been carried out. Most importantly, a national randomized evaluation of Success for All was reported by Borman and colleagues (2007). A total of 35 schools were randomly assigned to use Success for All either in grades K-2 or in grades 3-5. The primary grades in 3-5 schools were used as controls, as were the intermediate grades in K-2 schools. By the end of the

Table 3ILLUSTRATION BY GGS INFORMATION SERVICES. CENGAGE LEARNING, GALE.

study, Success for All second graders were scoring significantly better than controls onall reading measures (Borman, Slavin, Cheung, Chamberlain, Madden, & Chambers, 2007). Taken together, there are now more than 50 experimental-control studies of Success for All involving more than 200 schools throughout the United States. Since 1998 Success for All has been developed and disseminated by the non-profit Success for All Foundation, and is currently working in about 1,200 schools in 48 states in the United States, and 100 schools in England.

Direct Instruction. Direct Instruction (DI) (Adams & Engelmann, 1996), once known as DISTAR, is an elementary school program originally designed to extend an effective early childhood curriculum into the early elementary grades, in a federal program called Follow Through. Like Success for All, DI is primarily intended to help high-poverty schools succeed with all students, and the program is even more systematically specified for teachers.

The DI reading and math programs have long been marketed by SRA, a division of the McGraw Hill publishing company, under the titles “Reading Mastery” and “Connecting Math Concepts.” The publisher provides limited professional development with the program, but schools can contract with providers of professional development, primarily the National Institute for Direct Instruction (NIFDI) at the University of Oregon. Such schools receive approximately 32 person-days of professional development in their first year, similar to the services provided in the Follow Through studies. Research on DI has overwhelmingly focused on the model with extensive professional development, not on use of the books alone, and research findings for DI should therefore be assumed to apply only to the program with professional development. Certainly only this form could be considered a comprehensive reform model.

Borman and colleagues (2003) identified 40 experimental-control studies of DI, of which 38 were third party. The mean effect size was +0.15. CSRQ (2006a) rated DI's evidence of positive effects as “moderately strong,” with 11 “conclusive” studies.

School Development Program. James Comer developed one of the earliest of the comprehensive reform models, the School Development Program (SDP) (Comer, Haynes, Joyner, & Ben-Avie, 1996). The focus of SDP is on the whole child. Rather than focusing on specified curricula and instructional methods, SDP concentrates on building a sense of common purpose among school staff, parents, and community, working through a set of teams in each school that develop, carry out, and monitor reforms tailored to the needs of each school. A school planning and management team develops an overall plan, and mental health and parent teams focus on issues beyond the classroom.

Borman and colleagues (2003, p. 155) listed SDP as one of three CSR programs with “strongest evidence of effectiveness.” A set of three high-quality third-party evaluations described mixed evidence of the program's impact. One, a randomized evaluation in Prince George's County, Maryland, found poor implementation and no achievement effects (Cook et al., 1999), but a partially randomized study in Chicago (Cook, Murphy, & Hunt, 2000) and a matched study in Detroit (Millsap, Chase, Obeidallah, Perez-Smith, & Brigham, 2000) found small but positive impacts on achievement. CSRQ (2006a, 2006b) rated the evidence for SDP as “moderate,” with three “conclusive” studies at the elementary level and two at the secondary level.

America's Choice. America's Choice (AC) (NCEE, 2003) is a comprehensive reform model that focuses on standards and assessments, instruction aligned with standards, extensive professional development, and parent involvement. In particular, the program mandates a core curriculum in literacy and mathematics, tutoring for struggling students, and a school leadership team to coordinate implementation.

Borman and colleagues (2003) included only one study of the America's Choice design, but more recently researchers at the Center for Policy Research in Education at the University of Pennsylvania have carried out several evaluations. A longitudinal matched study in Rochester, New York, found that America's Choice students made greater gains than other students from 1998 to 2003 in reading and math (May, Supovitz, & Perda, 2004). A matched study in Duval Co., Florida (Supovitz, Taylor, & May, 2002) compared America's Choice and other schools on state tests, and results favored the AC schools in writing and, to a small degree, in math (but not reading). A 1-year matched study (Supovitz, Poglinco, & Snyder, 2001) also compared matched AC and control schools in Plainfield, New Jersey, and found greater gains for the AC students on the state English Language Arts test. CSRQ (2006a) rated the evidence of positive effects for America's Choice as “moderate” at the elementary level, with six “conclusive” studies, and also “moderate” at the secondary level, with five “conclusive” studies (CSRQ, 2006b).

Modern Red Schoolhouse. Modern Red Schoolhouse (Heady & Kilgore, 1996) is a program that emphasizes standards-based teaching, appropriate uses of technology, and frequent assessment. It provides customized professional development to help schools build coherent curriculum aligned with state standards and then implement aligned practices. In recent years, Modern Red Schoolhouse has begun to focus more on district reform and leadership.

Borman and colleagues (2003) identified four experimental-control studies of Modern Red School-house, with an average effect size of +0.17. CSRQ (2006a) rated the evidence for Modern Red School-house as “limited” at the elementary level.

Accelerated Schools. Accelerated Schools (Hopfenberg, Levin, & Chase, 1993; Levin, 1987) is a process-oriented school reform model that emphasizes high expectations for children and giving students complex and engaging instruction. Each school staff designs its own means of putting into practice the basic principles: High expectations, powerful learning based on constructivist principles, and avoidance of remediation.

Borman and colleagues (2003) identified three studies of Accelerated Schools with a mean effect size of +0.21. CSRQ (2006a) rated Accelerated Schools as “moderate” in research evidence, with three studies rated “conclusive.”

Expeditionary Learning/Outward Bound. Expeditionary Learning (Campbell et al., 1996, p. 109) is a design built around “learning expeditions,” which are “explorations within and beyond school walls.” The program is affiliated with Outward Bound and incorporates its principles of active learning, challenge, and teamwork. It makes extensive use of project-based learning, cooperative learning, and performance assessments.

Borman and colleagues (2003) identified four experimental-control evaluations of Expeditionary Learning, which had positive effects. However, CSRQ (2006a, 2006b) did not rate any studies of Expeditionary Learning as “conclusive.”

The experience of comprehensive school reform shows the great potential of whole-school reform for high-poverty schools. Research on CSR has clearly established that fundamental reforms can be introduced, implemented with quality, and maintained over many years. The longstanding belief dating back to the Rand Change Agent study of the 1970s (McLaughlin, 1990) that every school has to create its own approach to reform was conclusively disproved. Not all CSR approaches have been adequately researched, but in particular those with well-specified designs, clear expectations for what teachers and students will do, and extensive teacher and student materials, have been repeatedly found to be effective, scalable, and sustainable in a broad range of circumstances. Quality of implementation matters, of course (Aladjem & Borman, 2006), but it has been demonstrated that high-quality implementations of CSR can be achieved and that in such schools, children benefit.