Two Theories of School Improvement – Which Works Better?

Cross-posted at Education Week.

You have just been made commissioner of education in your home state.  Your maiden speech is coming up.  It’s your chance to lay out your agenda and get statewide support for it.  But what is that agenda?

You know your state can provide a decent living in the years ahead only if most of your high school graduates go on to get at least two years of education beyond high school, either in a career and technical field that requires solid academics or in a four-year college transfer program.

There’s only one hitch.  Your students have the profile of the student in an average state.  Half of your high school graduates leave high school without the academic knowledge and skills needed to succeed in the first year of your community colleges.  Most write poorly and the majority have trouble with middle school mathematics—to say nothing of high school level mathematics. Many don’t have the reading skills needed to comprehend the textbooks used in your community colleges, even though those texts are written at the 12th grade level.  But you had better not say any of that unless you have a realistic plan to fix it.  So, what’s your plan?

You need a way of thinking about the answer to my question that will help you develop a coherent set of initiatives that can get your state where it needs to go.  Let me offer two such strategies.  The first, let’s call it the Successful Programs Strategy, is based on the research methods favored by the U.S. Department of Education’s What Works Clearinghouse. They were first developed in the field of clinical medicine to identify effective drugs and procedures.

The second, let’s call it the Successful Systems Strategy, relies on research methods first developed by American companies in the 1970s and 80s to catch up with and overtake Japanese manufacturing companies that were beating them soundly in the marketplace—often putting them out of business—with better ways to organize and manage the work of the organization.

These are very different strategies.

The Successful Programs Strategy assumes that, on the whole, the current set of policies now in place stay in place, schools continue to be organized as they are, they have pretty much the same quality of leadership they have now, we continue to get our teachers from pretty much the same segment of high school graduates, our schools of education don’t change much and so on.  Indeed, the statistical methods used to support the Successful Programs Strategy are expressly designed to measure the effect of the program when everything other than that program is pretty much what you would find in typical setting, so you can see the difference the program makes.  And the researchers tell you that you are likely to get the same results only if the program is replicated in a similar setting.

The logic of the Successful Systems approach could not be more different.  The search for successful systems in industry is as old as industry, but the particular version of it that we at NCEE use in our own work—industrial benchmarking— began in the 1970s when Japanese manufacturers came seemingly out of nowhere to produce many kinds of manufactured products of higher quality than their American competitors using methods that enabled them to sell their products at significantly lower prices, putting many American firms out of business.  The American firms that survived sent their engineers to Japan to figure out how the Japanese were doing it.

They did not find a program installed in a manufacturing system that looked much like the American manufacturing system.   They found a different manufacturing system, organized on different principles.  Though the principles were very similar from firm to firm in Japan, the implementation was often quite different.  As they put the picture together, the Americans realized that some firms were able to implement some features of the overall approach with more skill and to greater effect than others.  Different firms excelled in different areas, which produced their competitive advantages.

The American firms had no interest in copying the Japanese firms.  They wanted to do even better than the Japanese.  They knew they had a lot to learn and they knew they would have to learn it from many different firms.  They would have to adapt what they learned to their own context and they would have to come up with a creative mix of the best of what the Japanese had to offer and the best of what their own engineers could come up with.  That was why they had no interest in replicating anything.  Many of the American firms that worked hard to benchmark the world leaders and learn from them went on to rejoin the ranks of the world’s leading firms.  Many of those too proud to learn from the Japanese went the way of the Dodo.

You realize that your state is in a position very like that of the American manufacturing firms in the 1970s and 80s. Your typical high school student is graduating two to three years behind the typical graduate in the top-performing countries.  They aren’t spending any more than your schools per student, but they are getting much better results.  It isn’t because they are using better programs.  It is because they have a better system.

The teachers in the top-performing countries are better educated in grade school.  The admissions standards for getting into a university that is allowed to offer a teacher education program are higher.  The licensure standards are higher.  They are compensated at a level that is much closer to the level of people serving in other professions requiring the same amount of education.  They are much better supported when they are first hired, essentially apprenticing to first-rate teachers. Like high-status professionals everywhere, they get more compensation, responsibility, authority and status as they get better and better at the work.  Schools are organized as professional workplaces, where the teachers spend much less time teaching and much more time working together in teams to build effective lessons, consult with each other about students who are falling behind, mentor new teachers and conduct research designed to systematically improve the curriculum, the materials they are using and the techniques they use.  Teaches have time to tutor their students when they need it. The level of trust between teachers and administrators is high as is public confidence in the public education system.

The contrast with the schools in your state is stark.  Applications for admission to teacher colleges have been falling for years.  Experienced teachers are bailing.  Teachers are telling their own children not to be teachers and are going on strike.

There is a proposal on the table in your state to address the literacy problem by hiring 17,000 teachers to tutor the state’s students in reading, at a cost of $1.4 billion.  It is a perfect example of the Successful Programs Strategy.  The schools would be required to select among two or three “proven” tutoring programs and the teachers would be required to use the selected program exactly as designed.  If you were to back that proposal in your maiden speech, you would get a lot of support.  It is easy to understand.  It has data to back it up.  Who could be opposed to the use of proven, evidence-based programs?

But then you think about what you’ve learned about the schools in the United States and the strategies the top-performing countries have used to get so far ahead of your state.  There certainly is tutoring for students who need help reading in those countries, but it is typically not a “program” in the sense in which that word is used in the United States.

Highly educated teachers-in-training in those countries are taught, among many other things, how children learn to read, what can interfere with their learning to read and how deal with those issues.  Among the things that can interfere with their learning to read are cognitive issues like insufficient phonemic awareness and comprehension problems, but also non-cognitive issues like violent environments at home and problems caused by frequent evictions leading to equally frequent moves from school to school.

Teachers in high-performing systems are equipped with a firm grasp of a wide range of effective responses to these and other barriers to learning.  Tutoring, is, of course, one of those possible responses, in its many different forms and intensities.  As they are taught their craft in clinical settings in their professional development schools, these teachers learn how to customize the form of help—including the kind and intensity of tutoring indicated—to the needs of individual students.

When they become fulltime teachers, they find themselves working with teachers in teams.  Among those teams is a grade-level team that meets for an hour or two every week. A good part of that team meeting is devoted to a session in which any teacher can bring up for discussion a student who seems to be having difficulties.  They pool their knowledge about that student, offering ideas about the probable cause or causes of the problem and agree on a strategy for addressing it.  That strategy might be a few sessions of after school tutoring, delivered by the regular teacher or a home visit by the homeroom teacher to follow up on the rumors of violence at home that the student is reluctant to talk about.  The grade-level meeting group will keep this student on their agenda until the student is back on track.  This particular student might end up needing a referral to the family and social services agency to deal with the violence at home or could end up with months of intensive tutoring or a few afternoons of tutoring might do it.  Different students may need different combination of these interventions at different times, not  packaged program.  Figuring out what those combinations should be for different students at different times is the essence of what it means to be a professional teacher.

Your choice becomes a no brainer.  All you have to do is compare what you would get by hiring 17,000 more teachers at a cost of $1.4 billion to deliver a standardized tutoring program to the kind of system just described, which you could get for the same, money.

You realize that you cannot build an effective system just by piling up effective programs.  It does not matter how effective any reading program is if your teachers were poorly educated, poorly trained and poorly led.  It makes no sense to give them a standardized program for all your students if, for the same money, you can help them develop the skills they need to draw wisely on a wide range of tutoring strategies and in a mix of their own design to address the wide variety of needs your students will have at any given time.  Nor does it make any sense to assume that all reading problems are problems of phonemic awareness and comprehension when we know that, in many cases, especially in our most challenged communities, students have been so traumatized by their experiences that their teachers have to work hard to get their students to trust them before they can teach them anything.

With $1.4 billion, you can do a lot to improve the quality of high school students who choose teaching as a career, strengthen the quality of their preparation as teachers and implement other measures to restore the morale of a profession whose members are telling their own children not to become teachers.  You will be able to give teachers more time to pool what they know about their students and work together in teams to come up with individualized, highly effective solutions to the problems they face.  With this approach, your teachers will have both the time and the skills needed to provide all the tutoring their students need without creating a tutoring program.

When you look into it further, you discover that many officials in the top-performing countries have said to researchers that much of what they are doing is based in part on American research.  Some of that research is Successful Programs research.  But they rarely replicate the programs that were researched. Instead, they inform their teachers about that research so that those teachers can draw on it as they decide how to meet the needs of their own students. Teams of teachers, drawing on research from many sources will come up with their own ways of addressing the specific challenges their own students face, do their own research on their own adaptations of research done elsewhere, combined with their own ideas, and share the results of that action research with their colleagues in refereed journals.  In this way, “Successful Programs” research can make an important contribution to improvements in the top-performing countries.  But this is a very long way from the “replicating proven programs” approach to improvement.

It occurs to you that there is an irony here.

In effect, the results of the American “Successful Programs” research show what happens when the “proven” programs are used in a pretty dysfunctional system.  There is no guarantee that they will work the same way in a much more effective system.   Given the characteristics of the American system vs. the kind of system found in the top-performing countries, it makes sense in the United States—where the quality of teaching and school leadership varies widely, resources tend to be unequally distributed, policies are relatively incoherent and so on—to mandate the replication of proven standardized programs.  At least the most disadvantaged students are likely to get a consistent offering that has been reasonably well designed.  But that is second best.  In the top-performing countries, those same disadvantaged students are getting top-notch teachers and school leaders who can and do customize solutions for each student, solutions that draw, in a deeply informed way, on the best research worldwide.

You are not willing to settle for second best. Your speech almost writes itself.