Cross-posted from Education Week
There seems to be broad agreement in the United States that improving the quality of the teaching force is a key to improving the performance of our students. But there is no agreement on how to do it. In this blog I want to characterize what I take to be the three leading strategies now on the table. Each is presented not necessarily as the only initiative that should be taken but rather as the most effective or efficient way to go. I will make a case that all are necessary if our aim is superior performance on the world stage. I’m indebted to Dylan Wiliam for this typology. Let’s begin by the considering the three strategies currently front and center.
Fire the worst. First, there is the strategy offered by Eric Hanushek. In a 2009 essay, “Teacher Deselection,” Hanushek presents what might almost be described as a surgical strike on the problem. Identify and then fire the worst five percent of our teachers. Do it often enough, he says, and those who remain will enable our students to perform at world-class levels.
Radically improve the quality of new teachers. Second, there is the strategy offered by my own organization, Vivien Stewart of the Asia Society and others. Observe the strategies already employed by the countries whose students are among the world’s top performers and adapt them for use in the United States, the strategies they point to in accounting for their own success. These strategies concentrate on changing the pool from which we select our teachers, setting much higher standards for their initial education, training, induction and licensure and compensating them at levels comparable to the compensation offered to high status professionals.
Invest in those you already have. Third, there is the strategy suggested by Dylan Wiliam, which is focusing, not on new teachers entering the teaching force, but on those already in it. At the heart of this strategy is the finding of psychologists who study professional competence and expertise that the effect of selection strategies based, in effect, on IQ (see the second strategy above) washes out after a few years and that what really matters is the steady, disciplined improvement of professional skill. This finding, Wiliam says, points to a major effort to support the disciplined development of employed teachers’ expertise as the core strategy for improving the quality of the teaching force.
It is too bad that we are not having a debate among the advocates of these radically different views, so that policy makers and practitioners can evaluate their claims, challenge their logic and make an informed decision based on a reasoned discussion. That debate is long overdue. In the next few paragraphs, I will offer a rapid-fire simulation of what such a debate might look like and then present my own conclusion about the direction the United States should pursue.
In his essay, Hanushek points to the failure of countless strategies for improving student performance and then suggests a different logic for strategy development. Why not, he asks, start at the end? If what we want is a powerful means of improving student performance, why don’t we start with what we know about the variations in school practice that make the biggest difference in student performance? The answer, he says, hands down, is teachers. When we look at variations in student performance, nothing moves the needle more than the differences in teacher competence, as measured by the value they add to student learning. The rest is simple logic and statistical analysis, he says. So let’s construct a dimension line of teacher competence, array teachers along that dimension line, identify the worst five percent of our teachers (those on the left tail of the distribution), and then look at the achievement levels of the students of the teachers who are left. We will find that, if we do this repeatedly, the teachers who are left will equal in quality the teachers in the countries with the best student performance. Problem solved. All we have to do to achieve world-class student performance in the United States is fire the worst 5 percent of our teachers each year, for as long as it takes to reach our objective.
Hanushek gets to this conclusion in the following way. He looks at the statistical distance between American students’ performance on the OECD-PISA surveys and the performance of Canadian students (Canada is among the top 10 and we are not). Then he looks at the difference for the average American students having the average American teacher, as opposed to having an average teacher selected from among the top 95 percent of American teachers. Then he uses this technique to estimate how many years we would have to do this to reach Canada’s level of student performance.
Now let me now go into a little more detail on the second analysis and prescription. It is the product of close examination of the strategies used by the countries that top the international rankings of student achievement. We note that they did not get there by firing their worst teachers but rather by greatly increasing the supply of superior teachers. This was done in the first instance by greatly increasing the appeal of teaching as a career to young people who could have their choice among the high status professions, partly by raising teacher compensation, partly by changing the conditions of work for teachers to make them comparable to the conditions of work in high status professions, not least by awarding these highly competent people the kind of professional autonomy that they would normally have in the high status professions. After radically altering the pool from which teachers are selected in this way, they then employ a much more demanding selection process for young people entering their teachers colleges, producing ratios of applicants to acceptances on the order of 6 to 1, 8 to 1, and in at least one country, 10 to 1. They also insist that their teachers really master the subjects they will take, majoring in those subjects in college if they are going to be secondary school teachers, and at least minoring in them if they are going to be elementary school teachers. And they typically insist that their teachers spend at least a year mastering the craft of teaching, over and above the time they spend on learning the subjects they will teach. Many are moving teacher education out of their third tier universities and into their top ranked research universities. Many are heavily subsidizing young people who choose to go into teaching. Virtually all insist on a period of apprenticeship for new teachers under the close supervision of master teachers for the first year or two after they have first been hired as teachers.
Now lets look at the third proposition. This one comes from Dylan Wiliam, interviewed for the most recent version of our Center for International Education Benchmarking newsletter. Wiliam has been doing a lot of research on the determinants of teachers’ expertise. IQ, he says, is an important, perhaps even dominant, factor in determining teachers’ expertise, but only during the first three years of practice. After that, he says, what counts most is disciplined practice. Disciplined practice is not the same as experience. One can teach year after year without getting any better at teaching. Disciplined practice is an activity that is always tough, no matter how good you are. It is what the best violinists and swimmers and, yes, researchers do. They are pushing themselves all the time to do better than their previous best, to improve their practice through constant learning and constantly applying what they learn. Much of the learning they do has to do with studying what other superb practitioners do, analyzing it keenly and then practicing what is learned in this way, but the best practitioners will take full advantage of every source of available knowledge that might contribute to the improvement of their own practice, often with the aid of a coach. To make Wiliam’s approach to raising teacher quality work at scale, a state or nation would have to develop a set of incentives to induce the current teaching force to want to steadily improve its expertise and it would also have to put in place a set of institutional supports that would enable them to do so.
The cheapest of these strategies is Hanushek’s. But it is not without its challenges. To announce that one’s strategy for improving teachers is to embark on a campaign to fire them is a surefire way to alarm the existing teaching force and to make capable young people choosing a career think twice about choosing teaching. And, sure enough, applications for slots in our schools of education are way down, partly, I have no doubt, because of the widespread impression that the dominant education reform strategies are essentially anti-teacher and anti-teacher union. If your aim is to strengthen the union and soften their union resistance to modern accountability systems, it would be hard to find a more ineffective strategy than firing teachers to improve teacher quality. So it is entirely possible that, by frightening away the most capable young college students, Hanushek’s policy is in practice leading to a lowering of teacher quality.
But, stand back from Hanushek’s analysis and prescription for a moment and compare it to the second prescription. Do you really believe that we can get to a world-class teaching force while we do exactly nothing on the list of initiatives taken on by the top-performing countries to improve their own teaching forces? Do you believe that we can radically improve our teaching force without recruiting our teachers from a higher strata of high school graduates, without paying our teachers at levels comparable to at least some of the high status professions, without insisting that our teachers actually know a lot about the subjects they are teaching, without requiring them to master their craft, without raising the licensure standards to levels comparable to the bars that have to be surmounted to becoming a teacher in the high-performing countries, and so on? Is there anyone out there who really believes that we can have a world-class teaching force without addressing these factors? Is there anyone who believes they are irrelevant? Another very simple way of asking this question is to ask where the replacements for the worst teachers are going to come from when we fire them? If they are no better than what we had before, then Hanushek’s scheme will not work at scale. If other countries are working hard to replace their current teaching force over time with a very highly qualified teaching force, and we are not, then we will never have a teaching force that equals theirs.
But what about Wiliam’s proposal to invest in the current teaching force, creating a system for supporting teachers’ continued growth and development that will incent them to constantly improve their practice while at the same time providing the infrastructure to enable them to do it? This makes perfect sense to me. But nothing about it is easy. We don’t have much experience in creating incentives for our teachers to constantly improve their practice. Nor do we have much experience in creating the kind of infrastructure in schools needed to support teachers who are determined to improve their practice. Both require a culture in the school that is sadly missing in most American schools. In my mind, what is really being talked about here is a transformation in school culture that goes way beyond any specific techniques or content of teacher professional development to embrace a very different outlook on what schools are for and what it means to be a professional teacher. Schools full of teachers who never give up on trying to improve their skills and knowledge as teachers are teachers who not only believe that they, no matter how good they are, can get better, they believe the same thing about their students and their school, and they are prepared to do whatever is necessary to achieve the progress they think possible.
From my perspective, there is merit in all three positions. Who could object to getting rid of our worst teachers? If we can find a way to do it that is fair and politically feasible, we should do it. We owe that to our students, their parents and the taxpayers. But no profession ever fired its way to excellence. The research my organization did years ago showed conclusively that most of the high status professions have very poor procedures for getting rid of their worst performers. They handle the problem of quality at the front end, by restricting admission to their professional schools to top high school graduates, having very high licensing standards and providing very high quality professional education and training, often accompanied by strong mentoring from top professionals once the new recruit is on the job, the same measures that the top-performing countries are now using to produce a superior teaching workforce. My objection to Hanushek’s analysis and prescription is not to the idea that we should get rid of our worst teachers, it is to the idea that doing so is the key to a first class teaching force. It is not. Pursued as if it were, it will simply make bad matters worse.
What about the other options, one concentrating on the incoming work force and the other on the current workforce? The obvious answer is that we need to do both. When you look closely, you will find that the top performers are in fact doing both. Singapore, for example, is not only a world leader in the selection and initial preparation of professional teachers, but it is also putting a lot of money and institutional resources into the constant professional development of its current teacher workforce.
I have often said that the United States was for a long time lucky to have a teaching workforce largely composed of college-educated women who had few careers open to them other than teaching. Their loss was the gain of the students they taught. Many are still in our schools, nearing retirement. But their successors are going to law schools, business schools, engineering schools and schools offering entre to other high status professions. It would, in my view, do us little good if we concentrated entirely on providing a very different kind of professional development to our existing workforce if we neglected what I see as a growing emergency caused by the drying-up of the pool from which we have been selecting our best teachers. And, for all the reasons that Wiliam suggests, it makes no sense to me to concentrate exclusively on the incoming teachers when it will take a very long time for them to make a difference in our schools and there is growing evidence that disciplined learning, combined with experience, could vastly improve the skills of our current teaching force. If we were to put our backs into both of these strategies, and thereby steadily expand our supply of first rate teachers, then it would make sense to get rid of our worst teachers, because we would have a growing supply of great teachers to replace them with, and, no less important, we would have demonstrated to young people making career decisions and good teachers now in our workforce, that we respect teachers and want to support them.