Harvard Graduate School of Education Logo

Education Reform, Accountability, and the Achievement Gap

Associate Professor Jal Mehta explores the various factors that drive — or impede — sound educational policy

April 23, 2009
associate professor Jal Mehta

“We need to think of each school as a kind of puzzle that needs to be continuously solved by the people in it,” says Harvard Graduate School of Education Associate Professor Jal Mehta in this thought-provoking interview. Mehta links to bold examples as he discusses solving this puzzle through raising school standards, ending "pass the buck" accountability, creating change through education reform, and closing the achievement gap.

You've done a lot of work on the history of American educational policy, and how educational policy has been shaped by the rise of standards and accountability. Could you describe how this rise in accountability came about, and what future trends we might expect to see in educational policy?

My research suggests that the rise in accountability is a result of a number of factors, some specific to the field of education and some not. Surveys show that the public's trust in major professions has been falling since roughly the mid 1960s, which has led to increased external accountability across a variety of fields. Education is particularly vulnerable to external accountability because of its dismal record with poor students, its organizational weakness as a highly feminized “semiprofession,” and the centrality of education to many non-educational concerns (economic, social, cultural, etc.), which leads a wide variety of actors to want to regulate it. No Child Left Behind is an outgrowth of a two decades-long push for external accountability that grows out of primarily economic concerns. I don't see the desire for external accountability reversing itself, but it is possible that the ends for which educators are held accountable could shift. In particular, the emphasis on “21st century” skills — defined by their advocates to be things like creativity, teamwork, and conceptual thinking (skills which, not coincidentally, have always been valuable) — could conceivably get those who want schools to improve for economic or social reasons on board with a menu of pedagogical objectives that would be much more congenial to many educators than what we see today..

Accountability is frequently taken to be the solution to the challenges we observe in the field of education. Why is this perception problematic?

It's not always problematic. In a project that I'm working on examining accountability across a variety of professions, I devote one chapter to accountability “success stories” and one to the “downsides” of accountability. The success stories tend to be internal to the field under consideration. An organization or even an entire sector embraces a certain set of outcomes as within its control to achieve, and then uses its accountability to those ends to drive an intensive examination and revision of practice. Atul Gawande tells an inspiring story of a doctor treating cystic fibrosis who helped his patients live 14 years longer than average by assuming that patient lifespan was highly affected by his medical choices and engaging in relentless experimentation until he created more effective practices. Good schools do this: they are determined to serve their students well, and they hold themselves accountable for ways that they are failing in their mission. Students who don't go to college, or who don't write well, spark these schools to re-examine their practice, to think about what they might do differently to improve outcomes. In my view, one of the most heartening educational developments of the past 20 years has been this embrace of internal accountability, as some good schools have tried to move from the “individual discretion” view of professionalism that has traditionally characterized the field to the more “collective view” of professionalism that internal accountability implies. The problem comes when we have what we've seen under NCLB, which is what we might think of as “pass the buck” accountability. Here a higher power asks a lower power to do something that neither the higher power nor the lower power knows how to do, and then proceeds to publicly embarrass the lower power for failing to do it. Coupled with the narrow metrics of success and rapid expectations for improvement, the result has been predictable demoralization and resistance at the school level, and little of the desired widespread improvement of practice. Policymakers and teachers are caught in a downward cycle where failures in past practice lead to increased external regulation, which in turn makes fewer talented people want to teach in the highest poverty schools, which in turn invites even greater regulation. What we want in the longer run is the opposite spiral characteristic of good organizations, where improved practice generated internally leads to increasingly lighter levels of external regulation. One more thing: accountability is just one dimension in the overall picture of producing organizational quality. Organizations improve by effectively training their practitioners, by increasing the stock of basic and applied knowledge, by inventing new practices and more consistently applying existing good ones. Medicine is a field that works relatively well where most of the improvement processes come on the front end (extensive training, clinical science building on basic science, ongoing dissemination of clinical science) with little emphasis on accountability for results on the back end. Education is a field that works relatively badly where there is little on the front end and lots of accountability on the back end. I don't want to send everyone to four years of ed school, nor do I think the science model applies to education as well it does to medicine, but I do think that in education it is the problem-solving side which is underdeveloped and the accountability side which is overdeveloped.

Your background in sociology and current research in education suggests the importance of having different types of expertise involved in educational reform. Why is it important to consider both local practices, “top-down” knowledge, and their relationship when implementing reform measures?

There is a pretty significant literature on this, which has been almost entirely ignored by policymakers in recent years. Those who actually sit in schools know a lot more about the problems they are facing than those who sit in statehouses: they know their kids; they know what techniques do and don't work with them; and they necessarily have had to develop a kind of craft or experiential knowledge needed to get through the day. To use an analogy: if you had to navigate a ship through a tight spot, would you want a Nobel Prize-winning physicist or an experienced boat captain? At the same time, local knowledge can have a kind of complacency or parochialism about it: the way we did it yesterday is the way we do it today and the way we will do it tomorrow. So, when things aren't working, policymakers are right to try to stimulate change (usually by creating incentives or rules that push more schools towards the characteristics of existing successful schools). But what they need to realize is that the impact of a policy is inevitably going to be an interaction of the new policy and the existing norms and routines of how things have previously been done. How that will play out in any case is much more difficult to know in advance than policymakers like to admit. Some of the wisest writing on this has suggested that we think of policy as a kind of initial hypothesis, which needs to be refined and adapted as we learn more about how it works in an actual context. The takeaway for me is that we want to combine an aggressive effort to combat educational inequality with an epistemological modesty about the likely results of any given policy choice. Pragmatism, as developed by Dewey and others, takes this general stance towards solving public problems and it still seems to me the right one.

How might we organize voices at the school level to make them heard at the policy level? What can teachers and parents do to ensure that their concerns are addressed by policymakers?

People at the level of the school have considerable unappreciated power in shaping the direction of school reform writ large. A working example is worth a thousand policy papers. Consider the success of “high poverty-high performing schools” like the Knowledge is Power Program (KIPP) charter schools, or the University Park Campus School in Worcester, MA (regular public). Over 10 years, schools of this variety have essentially changed the school reform conversation from “can schools overcome the effects of family background?” to “are we willing to go to the lengths that these schools do to overcome the effects of family background?” I think it is safe to say that Mike Feinberg and Dave Levin, the founders of KIPP, have had more influence on the school reform debate than any academic or policy wonk over the past 15 years. Similarly, if schools like the Expeditionary Learning Schools, which achieve similar results to KIPP using much more progressive pedagogy, continue to replicate their success and become more widely known, they could move the debate forward another step by showing that there are other ways to achieve the same outcomes.

In a paper on the achievement gap with Harvard Kennedy School Professor Ronald Ferguson, you argue that the achievement gap has narrowed in the recent decades, this progress came to a halt in 1990. Why is this the case? How can changes at the policy level help schools close this gap between the different student ethnicities?

We don't really know why progress in closing racial gaps slowed after some progress in the 1970s and 1980s. One theory is that it was easier to improve basic skills with policy from above because it requires little new learning on the part of teachers and limited advances for students. Our current challenge is to help all students gain higher order and more conceptual skills, which is a much more daunting task, one which the American school system has never accomplished. There are a number of things I could say about policy, but I will focus on just one. Policy would do well to remember that the key unit of improvement is the school. Going back to at least the 1970s “effective schools” literature we have examples of really successful schools in high poverty areas, but we still have no examples of similarly successful districts. There is a reason for this — the school is the level at which the working climate is set, it is the place where teachers meet to improve their practice, and it is the unit of which students are a part. We need to think of each school as a kind of puzzle that needs to be continuously solved by the people in it. What policymakers can do is to support that process in a variety of ways: altering human capital policies to bring more talented people into schools (and get less talented people out of schools); giving people in schools control over ends as well as means; giving principals and other teacher leaders within schools a variety of tools they can use in the process of organizational problem-solving; closing down failing schools; and developing curriculum, providing non-academic services, and doing the variety of tasks that can be done more efficiently outside the school. Essentially, we need to invert the pyramid, putting the school at the top, and thinking of policy's role as in support of the problem-solving that inevitably will need to happen at the school level.

See More In
Education Policy