One of the most expensive elements of HISD's school reform initiative is daily math tutoring for sixth- and ninth-graders.

A nationally watched experiment to improve nine struggling Houston Independent School District campuses improved students’ math skills but had a minimal impact on reading achievement, a new study found.

HISD Superintendent Terry Grier launched the Apollo school reform effort last year and has drawn attention for turning to the district’s competitors — charter schools — for improvement ideas.

“As remarkable as the math scores were, the reading scores are almost as underwhelming,” said Roland Fryer, a Harvard University economist who helped design the Apollo program and released his first-year evaluation on Oct. 6.

Still, Fryer described the overall results as promising and similar to those seen by successful charter schools such as KIPP. Charter schools are taxpayer funded but are exempt from certain state laws.

Yet taking the Apollo school reform effort outside the nation’s seventh-largest school district — or even sustaining it in HISD — could be difficult, Fryer said, noting the cost, local politics, and a shortage of talented teachers and principals.

“Perhaps the most important open question is the extent to which these efforts are eventually scalable,” he wrote.

For more school reform news, see:

On ed tech, we’re asking the wrong question

Merit pay in the spotlight at Miami-Dade schools

Beyond ‘Superman’: Leading Responsible School Reform

Fryer has said his paid consulting work on the Apollo project would not sway his study, which he prepared for the National Bureau of Economic Research.

The Apollo school reform program, which cost about $19 million last year, began with a staffing overhaul. Grier replaced all the principals of participating schools, and about half the teachers were new.

Students are in school about an hour longer each day, and they started class a week early.

One of the most expensive elements is the daily math tutoring for sixth- and ninth-graders. Struggling upperclassmen take an extra computer-based math or reading class.

Fryer’s research found that the tutoring — pairing one tutor with two students — was extremely effective, equating to between five and ninth extra months in school.

But the double dose of reading and math courses was only effective in eighth-grade math.