Like many of the kids who wind up in failing public schools, Harvard economist Roland Fryer grew up amid violence and drugs, and with little hope for the future.

Now, Fryer is the youngest African-American tenured professor at Harvard and a leading expert on education reform. His latest project: figuring out the right formula to help kids excel in school.

Most education experiments today are being done in charter schools — some of which have flourished, while others have failed. But at the slow rate charter schools are growing, Fryer said he realized that it could take decades to close the achievement gap in education.

So a few years ago, Fryer and his team at Ed Labs, a Harvard education research “laboratory,” spent a year scrutinizing 40 charter schools in New York to find out what they did right — and wrong. They analyzed lesson plans and homework. They also videotaped classrooms and broke the footage down into two-minute chunks to see what was happening.

Of the 500 variables they studied, Fryer’s team found that five accounted for a full 40 percent of the schools’ success. “That was just astonishing to us,” Fryer said, especially since so many seemed, at least on paper, to be relatively common-sense notions:

• Qualified principals and teachers: Teachers were rewarded for performance, and held accountable if they didn’t bring value to the classroom.
• More time spent in school: The schools’ day, week, and year were all extended to give students more time “on task.”
• Tutoring: Students received group or individual tutoring in core subjects.
• Data-driven instruction: Students were regularly evaluated and tutored in areas where they were weakest.
• Culture and expectations: The schools were focused on the importance of education in preparing them for their future — college, a career or both.

If they worked for the charter schools, Fryer wondered whether they might work in public schools, too. “It was an educated guess, in the sense that I didn’t know,” he said. But he added, “But I thought it was worth taking a risk, because we don’t really know how to turn around our bottom schools.”

All he needed was a school system willing to take a chance on his theory.

Enter Terry Grier, the superintendent at Houston Independent School District. Grier had only been on the job for a few months in 2009 when he learned that four of Houston’s 44 high schools were failing.

Grier had a choice between closing the schools down, handing them over to charter schools or trying something new. He heard about Fryer’s work and got in touch.

After a long phone conversation, Grier gathered a team and headed to Boston to hammer out a plan, which they named Apollo because of Houston’s historic role in the race to the moon. They started with nine schools — the four failing high schools and five struggling middle schools — and last year expanded to 20 schools.

Apollo 20 cost $61 million — an estimated $2,000 per student to pay for the tutors and the costs of keeping the schools open longer. Most of the money came from state and federal funds, but the district also had to raise $18 million from private groups.

Does it work?

“We were absolutely elated,” when the schools’ first-year results came out, Grier said. In math, for example, sixth- and ninth-grade students who had received “high-dosage” tutoring excelled in the math section of Texas’ aptitude exam and a nationwide test.

The second year results were less impressive, and Grier said he believes it’s because they lowered the time students were required to be tutored in math and reading. (You can read Fryer’s full evaluation of Apollo’s first year here [pdf].)

Still, Grier said he was encouraged. “When you go back and look at these schools, all 20 at one time were considered the worst performing schools, academically, in [Houston Independent School District]. That can’t be said any longer,” he said. This year, he said they plan to work even harder to get the scores back up.

Another Experiment

Last year, officials in the Denver, Colo. public school system learned about what Houston was doing and decided to adapt it for their district.

Allen Smith, the superintendent, said he liked Fryer’s approach. “Those five tenets were things that, if you paid attention, they could be implemented fairly easily and systematically across the network,” Smith said.

The nonprofit group Blueprint Schools Network, founded by former EdLab researchers, has been visiting Denver every month to evaluate its progress.

This year, six of the nine schools met or exceeded expectations, Smith said. Students in the program showed particular improvement in math and reading, subjects for which tutors were provided. (See the full report on the data here.)

“Inside, I’m really happy,” Smith said. “We’re off to a start that I think is better than most people thought.” But, he added, “I would still like to make sure we’re doing what we need to do.”

The Next 100 Schools

Fryer says he’s now working to scale the plan so that it can be easily implemented nationwide. Some of the tenets, like hiring good teachers, are still instinctive at best, and can’t be easily duplicated.

“What we have is a soufflé,” he explained. “I can tell you the ingredients, but it’s very complicated to make. We need to move this from soufflé to the popcorn button on the microwave.”

That may happen soon. Fryer is in talks with several school districts to start similar programs in 100 schools around the country by next fall. “This is not (just) about Houston,” he said. “This is about being a lighthouse for other schools.”

---