The Comprehensive School Reform (CSR) Program was established as a demonstration program in 1998 and authorized as a full program in 2002 as part of the No Child Left Behind Act (NCLB). It is one approach to help low-performing K–12 public schools meet state performance standards. CSR emphasizes two major concepts. First, the approach mandates that school reform should be comprehensive in nature, strengthening all aspects of school operations—curriculum, instruction, professional development, parental involvement, and school organization. Second, CSR should involve the use of scientifically based research models—that is, models with evidence of effectiveness in multiple settings.
In 2002, the U.S. Department of Education contracted with WestEd and COSMOS Corp. to conduct a five-year study of the CSR program, the Evaluation of the Comprehensive School Reform Program Implementation and Outcomes (ECSRIO), which involved a survey of 500 CSR schools and 500 comparison schools, case studies of 30 sites, and analyses of student achievement in all schools receiving CSR funding in 2002.
This report presents the third-year findings of this study and focuses on the relationship between CSR award receipt and growth in student achievement and whether aspects of program implementation are associated with higher CSR achievement outcomes. Overall, the findings indicate that schools receiving CSR awards did no better in mathematics and reading achievement than comparable schools not receiving CSR funding.
The report’s major findings, organized by research question, are:
Was receipt of a CSR award associated with improvements in school-level mathematics and reading achievement?
- Receipt of a CSR award was not associated with achievement gains in mathematics or reading achievement through the first three years of award.2
Were schools that received CSR awards more likely to implement the legislatively specified components of CSR than other schools?
- No, both CSR and non-CSR schools implemented an average of fewer than four components in 2003 and fewer than five in 2005 at both the elementary and middle school levels.
Was fidelity of CSR implementation associated with gains in school-level mathematics and reading achievement?
Two analyses address this question:
Comprehensiveness of implementation
- The comprehensiveness of implementation, as measured by the number of CSR components implemented, was not related to mathematics and reading achievement gains in CSR schools.
Adoption of models with a recognized scientific research base
- Only one-third of 2002 CSR awardees chose reform approaches with recognized scientific research bases.
- Low-performing elementary schools that adopted models with stronger evidence of effectiveness had gains in mathematics achievement that were not found in higher-performing schools.
- Adoption of a CSR model independently determined to have had limited scientific evidence of effectiveness was associated with higher gains in middle school mathematics achievement in all CSR schools, whether they were low-performing or not. There is also weaker evidence that CSR middle schools that adopted models with limited scientific evidence may have experienced gains in middle school reading achievement relative to schools that adopted other models.
- There was weaker evidence that low-performing middle schools that adopted models with moderate or higher bases of evidence showed improvement in mathematics compared with schools using other models.
- In no other instances was adoption of models with a scientific research base related to achievement gains.