Houston Story: 1 of 3 (The Setup)

This year I took on a consulting project, with the Houston school district. Was it the right move? Not sure. Let me explain.

The Houston Independent School District is 4 times larger than Boston. It is embarking on a massive school turnaround efforts. Like many urban districts. Federal pressure to do so.

A turnaround is tougher than starting a charter from scratch. Much.

Successful turnarounds? They exist mostly in lore. More hopefully, a little in reality. Mastery in Philly. Green Dot in LA. AUSL in Chicago. All have pulled off legit turnarounds, my peeps tell me.

(Locally, we get to see one effort up close starting this week. Orchard Gardens School has been persistently failing. They have a new leadership team and a bunch of new teachers. We began tutoring kids there last year, so we'll see the change in action).

The Houston Chronicle ran an August 16th story: "HISD launches an academic moon shot with Apollo 20."

Sandy Rodriguez, a soft-spoken 16-year-old with her lip pierced, got kicked out of Sharpstown High School last spring for getting into a fight with a group of girls.

On Monday, she returned to the freshly painted campus for the first day of a new school year pleased to be back, even if it was a week earlier than the start date for most other HISD campuses.

"It's better," Rodriguez said, "so you can learn more stuff."

More than 8,500 students at nine HISD campuses, including Sharpstown, will be in school for an extra five days this year — on top of an extra hour every day - as part of Superintendent Terry Grier's reform plan to boost academic achievement. The so-called Apollo 20 initiative, slated to expand by 11 elementary schools next year, includes daily math tutoring for all sixth- and ninth-graders and an extra math or reading class for students below grade level as well.

That daily math tutoring? That's a MATCH story.

The normal turnaround is this:

Human capital (better teachers and principals) More time Data driven instruction Culture and High Expectations

I believe Houston's Apollo 20 is the first-ever school district to try the MATCH Corps model. The high-dosage tutoring idea has spread a little over the 6 years since we started MATCH Corps, but not far.

A couple of other Boston charters now use it. A Princeton sociologist named Nick Ehrmann is trying our model in NYC. Achievement First, YES Prep have a modified version. A charter network called National Heritage Academies in Michigan and Ohio are trying it out this year.

We think that pretty much every traditional school in America has tutoring backwards.

Typically, it's "Let's scrounge together some extra help, and throw it at the struggling kids; let's use our weakest leaders to hire, fire, train, and deploy the tutors. The main course is regular classes. Tutoring is a little gravy."

We think of tutoring as a co-main-course in terms of importance: it has great power to raise achievement, if you solve logistics and cost issues.

Once you elevate the importance of tutoring, you need to zoom in on tutoring dosage. Again, schools are usually indifferent to dosage. They take whatever dosage they happen to have around, and just spread it around.

Let's say your medicine dosage needs to be 100 mg per day. But there's a shortage. So you are given 20 mgs one day, 0 the next, 4 mgs the next, etc.

Of course that doesn't work. It's probably better just to skip the medicine entirely. Indeed, study after study shows sloppy, low-dosage tutoring doesn't help.

What's the right way? Estimate what kids need, provide that actual dosage, measure the change.

High-dosage tutoring is surprisingly cost-effective if you measure it this way: Student Achievement Gain Divided By Cost. It's all about management mindset.

Our other key discovery is some top notch recent college grads would prefer to earn less than teachers in exchange for working with 2 kids at a time. They think the work can be much more satisfying. And they can keep the commitment to a single year.

MATCH Corps frequently reject $43,000 per year full-time teaching jobs to take our work for $7,500 per year plus housing.

Combine high tutoring dosage for every kid, and lower cost for recent college grads, you've got our model.

Harvard EdLabs founder Roland Fryer is a cautious believer. In other words, he thinks our idea is plausible. He's going to ruthlessly measure the effect of high-dosage tutoring this year.

Roland's the driving force behind Apollo 20. And he wanted to rope us in to help.

It started with a simple one hour meeting in Harvard Square, back in May....