School improvement is more about execution than innovation, but doing what works -- like tutoring struggling students -- consistently across a network seems
pretty innovative in education.
In 2010, Houston Independent School District
(HISD) Superintendent Terry Grier called Harvard’s Roland Fryer who had just published a report outlining the
five factors that contribute the most to the success of high performing school networks. These five factors became the backbone of the Houston turnaround
effort called Apollo 20:
1. Start with high expectations
2. Use data to improve instruction and provide teacher feedback
3. Increase instructional time, and
4. Provide targeted tutoring.
This is the first of two posts on Houston as part of Smart Cities, a series examining how
and where innovations in learning are happen and how they spread.
A few school visits and a first year review of the data suggests that Apollo has taken off, “the average
impact of these changes on student achievement is 0.277 standard deviations in math and 0.061 standard deviations in reading, which is strikingly similar
to reported impacts of attending the Harlem Children’s Zone and Knowledge is Power Program schools.”
The turnaround strategy started with hiring great principals. Apollo principals then replaced between 30 to 80 percent of the staff.
Last Monday, Secretary Duncan said
, “We don’t have one district that systematically identifies top performers and matches them with students most in need.” Superintendent Grier makes a
compelling case that Houston has done just that. Duncan said, “We have been uncourageous on addressing this opportunity gap.” That certainly does not apply
to Grier whose team exited 800 teachers from Houston last year.
Principals in Houston receive extensive training in teacher evaluation. With three members of the central office team, Principals participate in a Staff
Review of each evaluation -- including academic results -- for every teacher and are required to place their teachers in one of four categories ranging from
high to low. Principal evaluations hold accountability for retaining top teachers and exiting low performers.
Initially it was difficult to attract great teachers to previously failing and sometimes dangerous schools. Grier and team turned that around with the
largest Teach for America (TFA) partnership in the country by offering big signing bonuses and by marketing the opportunity. They made visible changes at
each school and invited candidate teachers to visit and talk to students who could attest to the turnaround underway.
Sharpstown High School
is safe and orderly, a big improvement over two years ago. A boost in math scores is due in large part to the one-on-two tutoring for students struggling
in math. Tutors, called Math Fellows, are paid a $20,000 base with the potential for a $5,000 bonus. In addition to new or retired teachers, professionals
have been attracted to the tagline, “Give a year, change a life.” A sophomore said, “It’s not that I learn more with the tutor than I do in class; it’s
that he helps me understand the math.”
Tutoring low performing students is a great idea. That’s why it ended up as a step in NCLB. Tutoring appears to work particularly well if it is fully
integrated with classroom instruction, but it will be expensive to continue post-grant. The good news is that adaptive and personalized tutoring math
software is making it easier to blend small group instruction with online learning.
There is a well-intentioned data room in the high school that hosts staff conversations about data-driven instruction, but it’s old data from old fashioned
state tests that are manually transcribed. They are paying attention to every student and making sure they have a shot at college. And there’s been a 50
percent jump in college going rates as a result, but in 2012 we should have daily achievement data with powerful visualization dashboards (like the data I
had in retail 20 years ago).
There’s not a lot of instructional technology in HISD. There is an opportunity to turbo charge and improve student engagement, increase personalization,
and boost sustainability of this effort.
The credit recovery lab at Sharpstown is doing its job but the widely used brand name content is disappointing--flat and sequential with multiple-choice
end of unit quizzes. Struggling students deserve better.
The school visits gave me the overwhelming sense that despite big investments in the last 36 months, the sector is more than five years behind in the
development of basic tools, including simple translations of tools widely used in other sectors.
Fondren Middle School
has a clean, pleasant environment. Like Sharpston, attention to building and grounds paid off in contributing to a serious academic atmosphere.
As a new principal, Charles Foust found education malpractice widespread in December of 2010, “As principal, you do what needs to be done.” He replaced 35
teachers.
Science classes incorporated blended learning strategies with students rotating from BrainPop on laptops, to hands-on experiments, and to a teacher led
discussion at a smart board.
With little prompting, students launched into KIPP-like chants about their academic expectations. College banners hang everywhere, another trick borrowed
from charters.
Learning objectives and assignments were clearly identified in every classroom. Paired reading in social studies resulted in shared comments in preparation
for a Civil War essay. Common classroom management strategies were evident school wide.
Robinson School of Global Studies
is “an amazing school,” in the words of Seth Andrew, Democracy Prep, and a critical classroom observer. Common literacy strategies were evident across the
elementary school. It’s hard to believe this positive and cheery place was among the worst of the 170 elementary schools in Houston a few years ago.
Bilingual classrooms appeared to be driven by confused state bilingual laws and related local politics. Spanish speaking students learn and are tested in
Spanish until fourth grade and then dumped into mainstream classrooms. I’m all for bilingual learning, but it would clear things up if students were just
tested in English as soon as possible.
HISD raised $17 million to support the 20 Apollo schools, which is about $1,800 per student. This is the last year of the grant.
EdLabs
is an interesting combination of instigator, technical assistance provider, and evaluator. Rowland Fryer’s center at Harvard is a hands-on, learn by
testing, make-a-difference kind of shop. Fryer and team have been in Houston dozens of times over the last two years. Fryer is intrigued by the potential
of working at scale, “In our second year in Houston, we were working with 20,000 kids--about half the number of the kids in the whole KIPP network.”
You don’t really expect a tenured Harvard economist to muck around in schools at this level of detail with an interest “in taking instruction from good to
great.” It’s obvious that Fryer appreciates Grier’s willingness to do whatever it takes to boost achievement. Fryer’s shop was launched with a big grant
from Eli Broad.
Houston gained national notoriety in the late 90s under Rod Paige’s empowerment and accountability agenda as Houston superintendent. The improvement agenda
co-crafted by Fryer and Grier suggests that Paige was right about empowered leadership but that a big portfolio like Houston warrants powerful improvement
capacity like Apollo schools.