Model for Redesigning Middle Schools Found To Boost Scores
The directors of a schoolwide improvement model targeting high-poverty, urban middle schools are pointing to a new analysis of test scores in Philadelphia as evidence that they've got a good thing going.
It hasn't been an easy school year for the Talent Development Middle Schools model, which was created by researchers at Johns Hopkins University in Baltimore and most widely implemented in the City of Brotherly Love.
While the model was being used in 11 Philadelphia middle schools at this time last year, that number has since dropped to seven because of governance changes triggered by the state's 2001 takeover of the 208,000-student school system. Of the remaining schools, four have been turned over to education management organizations, become charter schools, or both.
Still, the program's directors are pointing to scores on Pennsylvania state tests as a sign that their model deserves a second look.
"Despite a lot of turmoil, we've been able to make progress," said Robert Balfanz, a co-director of the Talent Development Middle Schools model, who conducted the study.
The Talent Development model is among a small but growing number of designs for whole- school reform that are aimed at disadvantaged youngsters in the middle grades. The model's middle school version is being used in 18 schools in six states. It stresses schoolwide use of the same curriculum, intensive in-classroom coaching of teachers, the creation of small learning communities, and extra classes in core subjects during the regular school day.
The National Forum to Accelerate Middle Grades Reform, which started in 1997, now includes representatives of seven whole-school models, Talent Development among them, said Deborah Kasak, the forum's executive director. Some of those designs were made possible by grants from the U.S. Department of Education to allow models focused on elementary or high schools to expand to middle schools, she said.
"Often, middle-grades schools aren't given the attention they need," said Ms. Kasak, based in Champaign, Ill. "That some of the models are choosing to go into the middle grades, we think is really encouraging."
In the Philadelphia study, results varied among the 11 Talent Development schools, which were divided for analysis into two groups: four schools in their first year of implementation in the 2001-02 school year, and seven others that had been using the model for three to six years.
For the second group, researchers combined averages of 8th graders' mathematics and English scores on the Pennsylvania System of School Assessment in the two years before the schools began using the model, and then compared them with scores for the two most recent school years.
On average, the seven schools raised their scores from their premodel baseline by an amount that was nearly double that of district 8th graders as a whole, and about 20 percent higher than comparison schools selected on the basis of their demographics and prior achievement, according to the study. The absolute scores of the seven schools were below the districtwide average, however.
The greatest gains came in the two schools considered the model's flagships: Central East Middle School, which began using the model in 1995-96, and a school that started the following year, Jay Cooke Middle School. Those schools benefited from core groups of faculty members committed to the model—a rarity in many high-poverty urban middle schools—who kept the program on track despite teacher turnover, Mr. Balfanz said.
Yet one school that has also achieved a relatively stable teaching force, the 900-student Roosevelt Middle School, saw the lowest gains of all seven of the multiyear Talent Development schools. ("Mission: Stability," Jan. 9, 2003, from Quality Counts 2003: "If I Can't Learn From You.")
Roosevelt is among 25 schools that have been "restructured" by the district itself, and as a result, it has dropped the Talent Development model.
James J. Kemple, the principal investigator for a third-party evaluation of Talent Development being conducted by the New York City-based Manpower Demonstration Research Corp., said last week that he viewed the methodology of Mr.Balfanz's analysis as sound. MDRC is planning a March release for its first report on both Talent Development's middle and high school models.
Vol. 22, Issue 19, Page 6