Last week, a group of around 25 education advocacy organizations, including influential players such as Democrats for Education Reform and The Education Trust, released a "statement of principles" on the role of teacher quality in the reauthorization of the Elementary and Secondary Education Act (ESEA). The statement, which is addressed to the chairs and ranking members of the Senate and House committees handling the reauthorization, lays out some guidelines for teacher-focused policy in ESEA (a draft of the full legislation was released this week; summary here).
Most of the statement is the standard fare from proponents of market-based reform, some of which I agree with in theory if not practice. What struck me as remarkable was the framing argument presented in the statement's second sentence:
Research shows overwhelmingly that the only way to close achievement gaps – both gaps between U.S. students and those in higher-achieving countries and gaps within the U.S. between poor and minority students and those more advantaged – and transform public education is to recruit, develop and retain great teachers and principals.This assertion is false.
Let me attempt to guess at the reasoning of the statement’s authors. They figured: Teacher quality is the most important measurable factor within schools (at least in terms of test-based outcomes), and therefore schools will never function at their best unless we improve teacher quality. If this was in fact their rationale, then what they should have said is that teacher-focused policy has a great deal of potential, and is worth focusing on (though the research on principal effects is, at best, underdeveloped). What they argued instead is that “research shows” it works, that it works big time and that it’s the only thing that gets us where we need to go.
Sorry, but the research does not “show overwhelmingly” that this is true. Actually, it doesn’t show it at all.
The available evidence, at least as yet, does not even indicate that we can use policy to spur large shifts in the distribution of teacher quality, to say nothing of the kind of shift that would make a large dent in achievement gaps as commonly measured. Frankly, we’re still in very early phases of figuring out how to best measure teacher quality, which is, ironically, the primary focus of the rest of the statement. It’s rather strange for a statement of principles begin by asserting that “research shows” we can close achievement gaps by the process of improving teacher quality, and then go on to lay out speculative guidelines for measuring teacher quality, which is among the first steps in that process.
I am hopeful that these measurement efforts, with some trial and error, will eventually be successful, but they’re a far cry from the ultimate goal of actually affecting teacher quality. Whether or not we can do so at any significant scale is still very much an open question (see this paper for some early evidence), in no small part because how we use these evaluations and other tools is in many respects just as important as how good they are. Moreover, it is highly unlikely that these efforts by themselves, if successful, will be enough to “close achievement gaps," whether those between student subgroups or between the U.S. and other nations. This claim is an inflation of expectations to a level that belies the enormity and multifaceted nature of the problem.
And, even if there was evidence that we could drastically improve teacher quality on a large scale via policy, the claim that it is the “only way” to do so is just wrong. There are many different types of education policies, a few of which have a far more impressive track record than the “new wave” of teacher-focused policy (which barely has any concrete track record at all). For example, the benefits of well-designed early childhood intervention programs – not just pre-K but also ages 0-3 – have been shown to be large, persistent and extremely cost-effective (see here, here, here, here and here). This is especially true in the case of reading achievement, which is often more resistant to other forms of intervention, given that the broad background knowledge required for reading begins growing with a child's first year.
Listen – I’m all for improving teacher quality. I personally support many of the general policies outlined in the statement, most notably new evaluations, including those that use growth model estimates in a responsible manner (which is thus far not the case in many places).
Sure, I often disagree with the market-based crowd on the specifics of their proposals, such as the statement’s recommendation that growth-based measures should be the “predominant factor in a teacher’s evaluation” and the idea of predetermining the number of "performance categories" into which teachers must be sorted. Then again, I think there are some good specific recommendations in there, including the call for a five-year implementation of new evaluations (with two years for design and piloting), linking evaluations with professional development and its emphasis on continually examining the distribution of teachers across schools with differing student populations. Disagreement on the details is a normal, often productive part of policy debates.
But let's stop making strong, sweeping arguments that have no empirical backing, especially when those arguments are presented (without citation) as the “overwhelming” conclusion of evidence that doesn’t exist. Teacher-focused policy may have a great deal of potential, but we need to be honest about the fact that we’re still very early in the process of policy development. And we certainly need to avoid – at all costs – minimization of how much we still have to learn, unjustified inflation of expectations, and discounting other types of interventions that are also critical. Those too are principles worth stating.
- Matt Di Carlo