Tutoring Doesn't Work?

In the world of bad headlines.... From EdWeek:

Duncan to Florida: Tutoring Doesn't Work

U.S. Secretary of Education Arne Duncan said today he doesn't understand why Florida passed a law requiring districts to continue offering free tutoring to students in struggling schools.....

He pointed to a recent U.S. Department of Education study, which examined so-called "oversubscribed" school districts in three states, Connecticut, Ohio, and Florida. Oversubscribed districts are rare and basically are those that have schools where more students wanted tutoring services than the district could afford to help. In those cases, schools prioritize the most academically needy students. The study looked at kids who got the services, and those who just barely missed out. And it found that there wasn't a substantial difference in achievement between the students who got the tutoring and those that didn't.

Hmm.

I think it would be fair to say:

1. The federal law which required failing schools to let parents take $1500 or so per year of funding away from the district, and to direct that money towards tutoring of their child, did not work.

I agree with that study.

2. We could debate cause there. A typical version of what happened in real life was:

a. Most parents did not know they had this right.

b. The district recaptured almost all of the money by offering "tutoring" that was described by my friends in the district as often being "lightly staffed after school study halls."

c. A few for-profit providers that provide SAT prep and such also entered this market.

My sense from a wildly unscientific sampling of conversations: the quality of this tutoring was not impressive.

d. Match actually created our own little program. We'd hire our own juniors and seniors who had ACED the state's math MCAS exam. They tutored, 1 on 1, Boston 6th graders who were failing math.

I loved that little program. My colleague Alan Safran set it up for us. We actually subsidized it a little. The effort just lasted one year, though. Squashed by the bureaucracy.

(I still believe an obvious, large untapped resource pool are the black and Hispanic inner-city high school kids who are acing state exams; pay them to tutor younger kids from the same neighborhoods. Believe me, they want jobs! If anyone wants to start a non-profit to do that, let me know, I'll help in any way I can).

3. Whatever the failures of that specific federally-imposed tutoring effort, I doubt that Secretary Duncan was arguing that tutoring "generally" doesn't work.

High-dosage tutoring by full-time tutors used efficiently DOES work.

Here's our evidence:

a. Our high-dosage tutoring program at Match led to big gains.

b. It was then copied by other local charter schools, which led to giant gains.

c. We then brought it to Houston Public Schools as part of Roland Fryer's turnaround effort there. It led to spectacular gains. Here's an analyst from the teachers union describing the gains. The Houston Chronicle:

The main lesson is clear: Math tutoring really, really works.

In sixth and ninth grades, all Apollo 20 kids spent an hour each school day in an intensive, two-kids-one-tutor session. And according to Fryer’s analysis, the results were astounding.

In math, sixth-graders gained the equivalent of six extra months of schooling; ninth-graders, between almost five months to nearly nine months.

Fryer’s stats show that Apollo-20-with-tutoring makes far more difference in kids’ math performance than almost any other school intervention ever measured.

Let me quote from one of my older blogs for the stats-hounds amongst you.

Grade 6 math tutoring has an effect size of +.484 standard deviations…..So gains that are more than 400% bigger. It’s double a typical KIPP effect.

Grade 9 math tutoring has an effect size of (a whopping) .739 SDs….

Has there every been an effect size of .739 in K-12 on any intervention? I can't recall one. But I'm young! At heart.

d. Roland then brought it to Denver Public Schools for this year.

High-dosage tutoring by Denver full-time tutors, via a nonprofit called Blueprint, is going so well, they too will soon unveil spectacular gains, and try to grow it aggressively.

4. More proof points are on the way!

Alan will lead a Match team in an ambitious effort this coming year to replicate those gains in a new urban district.

Our friends at AUSL in Chicago will soon roll out their own such effort as well.

Our friends Mike Duffy and Jared Tailefer brought this model to Great Oaks charter school in Newark. We'll see what they find.

Hopefully we'll have 3 more major proof points soon.

5. Teacher-friendly

Not only do kids love an hour a day of tutoring in a 2-to-1 ratio from the same tutor every day at a time built into the regular schedule (not after school)...

And not only do they make huge gains...

But the salutary effect on teachers is not to be omitted. In many schools, it's implied that each individual teacher should somehow remediate all 100+ of their kids. It may be true that there exists a small cadre of teachers so skilled that they can pull that off, and of course we should try to grow the numbers of those folks. But remediating 100+ kids who arrive way behind, in 10 months, is not a job for mere mortals.