“Our purpose in producing these consumer guides is not to pick winners and losers, but to give the public, the profession and policymakers solid information on which to make judgments about how best to improve schools,” says CSRQ Center director Steve Fleischman, who formerly worked in the AFT’s educational issues department.

As with the report on elementary models, the survey of middle and high school models includes the most prominent national programs. Each model serves more than 40 schools in at least three states and is available for adoption in most states. Most have been adopted by low-performing schools with mostly high-poverty students. The four top-ranking models, which received a “moderate” rating for improving student achievement, were America’s Choice School Design, the School Development Program, Success for All—Middle Grades, and Talent Development High School. Six models received “limited” ratings, and the other eight received a “zero,” including well-known programs such as the Coalition of Essential Schools and High Schools That Work. (See the accompanying chart for the full list of models and ratings.) None of the models received a “very strong” or a “moderately strong” rating in the area of raising student achievement.

The report ranks the 18 models in four other areas in addition to raising achievement, including outcomes such as attendance, discipline and graduation rates; involving parents, families and the community; providing a link to the research base for the model’s design; and offering effective professional development and implementation support.

To gather evidence on the programs’ effectiveness, the authors screened nearly 1,500 documents and reviewed almost 200 studies. Ultimately, only 41 of the studies were judged to be rigorous enough to be used in compiling the ratings. In their study on elementary schools, the researchers found twice as many sufficiently rigorous studies. “Some models may be delivering results but may not yet have had sufficient time to demonstrate their effectiveness through evidence that meets the No Child Left Behind Act’s requirement for scientifically based research,” Fleischman notes.

The study’s authors offer a similar caveat to the one provided in the elementary school report: Implementation is crucial. This means that some schools might see significant results with a carefully implemented version of one of the lower-rated studies; conversely, a higher-ranked model might not boost student achievement if badly implemented. The authors also note that many of the models are constantly evolving and refining their designs, so the studies might not reflect the most recent variations of a program. Although more research is needed on the whole range of school reform models, this latest work from the CSRQ Center provides at least a user-friendly starting point in evaluating what really works.

The complete study, and the center’s other work on reform models, is available online at www.csrq.org.