The Greening of Education
Right now, somewhere in the United States a newspaper or television show is talking about Jay Greene’s education research. And, what he would consider even more important, it’s safe to say that right now somewhere a researcher is replicating, criticizing or citing Greene’s research. His goal – and passion – is to remake public education into an effective system, able to meet the needs of each individual child. As the first head of the university’s department of education reform, he is in an enviable position: he has the resources – starting with a $20 million endowment – to assemble a band of colleagues who share his goal and bring their own energy and vision to achieving it.
Greene came to the University of Arkansas in 2005 with experience both in academia and as a fellow with the Manhattan Institute, a position he retains while leading the department of education reform. His research has been wide-ranging, including, but not limited to, issues in choice, accountability, graduation rates and social promotion. The department hosts its first conference in October 2006, featuring the best ideas for improving public schools, submitted by members of the department’s Technical Advisory Board, which is made up of some of the leading education researchers nationwide. An edited book will follow.
After ten busy months of department formation and continuing research, Greene sat down to discuss his views on education today and his hopes for the future.
Q: What do you think are the most important questions facing public education?
I think the most important question facing public education is how we can improve productivity, that is, how we can get better results for the dollars we spend. The truth is that we’ve been paying a lot of attention to insuring that schools have adequate resources. We’ve double per pupil spending nationwide over the last three decades. We’ve done the same in Arkansas. And yet results have been basically flat. They haven’t gotten worse, but they haven’t gotten better. Anytime you spend more and don’t get more in return, you have a productivity problem, and that’s basically what we’re suffering from in education.
We’re spending more and more because we’re not satisfied with the level we’re at. We want to be better, which is good, but we haven’t figured out how to get there. And simply spending more doesn’t get us there. Unless the next three decades are different than the last three decades, doubling spending again won’t improve our situation.
At the heart of the problem is that we don’t connect resources to outcomes. So schools receive resources regardless of how well they do or how wise the decisions they make are. Essentially this is a lack of accountability. Whenever resources are completely unattached from the wisdom or foolishness of the choices that people make, they are essentially unaccountable and our school systems are to a great degree unaccountable.
Now we’ve started to address this, we’ve started to recognize this problem and in the last few years we’ve introduced accountability testing as one mechanism but it’s a very imperfect mechanism for trying to have consequences attached to decisions. And only when there are consequences will people have incentives to start making wiser decisions and avoiding foolish decisions.
In the past what we relied on was goodwill, and luckily we have a lot of goodwill in public education. We have a lot of good people who care a lot about kids, who care a lot about improving the quality of education, and we’ve relied on that goodwill to produce a quality outcome. The problem is that goodwill is not enough. We see this in many other enterprises where there is a lot of goodwill – medicine, higher education – people who are engaged in these occupations or the professionals working in these fields all care about having good outcomes but that caring is not enough. There also have to be proper incentives for people to make good choices so that results are improved. And we’ve done a better job at that frankly in medicine and higher education that we have in public education.
At a general level there are two broad reform strategies that attempt to increase the accountability of people in the education system, that are strategies for improving productivity. They are what we call accountability which largely consists of testing….
Choice allows consumers of education, children and their families, to move from one school to another and bring their resources with them from one school to another, and this attaches consequences to decisions because schools that make foolish decisions lose students and their revenue, and schools that make wise decisions attract students and their revenue. And both kinds of reforms in theory could improve productivity by attaching consequences to decisions. Now we have tried both of these kinds of reforms on modest scales with encouraging results. So for example we have a fairly large number of studies now on what happens when you increase the amount of choice in education.
It’s important to keep in mind that the real question here is not whether to have choice or not in education. We’ve always had it we always will have it. The most common form of school choice is residential choice. People choose where to live to gain access to desired schools if they have the resources to do so. The trouble is that it’s a relatively inequitable and inefficient type of choice. Only some people have the resources and even the people who have the resources are reluctant to move frequently because there are very high costs associated with moving. It’s disruptive to ones life and you have to pay taxes and pay realtors and move away from families and friends so it’s not something people relish doing so this means it’s a relatively inefficient way to exercise choice. So there have been a number of studies of what happens when we increase the amount of choice in an area. There’s actually a nice review of that research done by Hank Levin and Clive Belfield at Teachers College. And they find that the results are pretty consistent.
When there’s an expansion of choice and competition, public school performance improves and costs tend to lower. They believe that the results are relatively modest in their benefits, but the results are pretty consistent across the research that has been done on this question. In fact, I’m not aware of studies of expansion of school choice in the United States that find the public school performance goes down when faced with increased choice in competition. There are some studies on programs overseas where that is the case but none in the United State that I’m aware of.
So we have some encouraging results there that we might improve productivity in education if we were to expand choice and competition. Just to give you some ideas of where we’ve tried this sort of thing, there’ve been some good studies done of charter competition the competition between charter schools and traditional public schools because keep in mind that when student move from the traditional public school to a charter school, money leaves the traditional school system and enters the charter school. Or if a student leaves a charter school and goes to another charter school, money leaves one organization and goes to another. This by the way is the same as when a student leaves one school district and moves to another. Money leaves one institution and goes to another.
In Michigan, in Arizona, there have been large concentrations of charter schools so it has been possible to study the effects of this charter competition on traditional public schools performance. The results have been quite positive. Similarly we’ve had a voucher choice program, a private school choice program, in Florida and there’ve been four studies of that program, and all four find that public school performance goes up when public schools are faced with increased choice and competition from a private voucher program.
Now keep in mind, researchers always disagree but the overall thrust of the findings is the same. They disagree on how large the effect is and they disagree on whether it exists in all subjects or some subjects, the confidence with which we observe the effect there’s disagreement about that. But it’s useful to step back and look at the landscape and the landscape is that you have consistently positive results from this research, some of which I have done and some of which others have done. That’s the great things about researching in these areas. If you do something that finds a certain effect, others have an incentive to attempt to replicate that and to show that you’re wrong. There are extra brownie points for finding that earlier research was mistaken. That’s the reward system, the status system within academia rewards researchers that find that earlier research was mistaken. That’s what keeps us all honest and also helps us all gravitate towards the truth. We only imperfectly grasp the truth and any one study can be mistaken but one study invites other studies and if the other studies point in the same direction well then we have an idea of which way the truth lies.
Accountability is the other reform strategy out there. Keep in mind that choice is a type of accountability too; we just don’t call it accountability. They’re both accountability systems. One is accountability through market mechanisms and the other is accountability through command mechanism. You centrally determine a goal and you attach rewards and sanctions to meeting that goal. And this exists in the marketplace. Companies do this internally. The University of Arkansas has goals 2010. We have internally determined goals, centrally determined goals, the chancellor has said these are the things we need to do and there are rewards and sanctions attached to progress towards those goals.
And that’s an accountability system. It can create discomfort but that’s generally how organizations make progress. And they make that kind of progress because they are competing with other institutions. They have an incentive to develop these command systems so if we don’t make progress to the chancellor’s goals we’re also in a competitive struggle with other institutions of higher learning and we may lose students and revenue and status to those other institutions, and we don’t wish to do that so we’re trying to figure out ways of preventing that and moving up on others and overtaking them and hopefully we will.
Because No Child Left Behind has made the implementation of accountability testing universal, we only really have research on this question from looking at the variation in adoption of accountability testing that predated the federal mandate that everyone have it. Some states adopted it earlier, some states adopted it only when required to do so by No Child Left Behind, which is the federal legislation. As point of fact, 37 states were already engaged in accountability testing by the time No Child Left Behind was adopted. So really the federal law solidified a developing consensus among the states about what they were going to do. But some of those 37 states were early adopters of accountability testing, like North Carolina, Texas, Kentucky, Tennessee. Others were late adopters or didn’t adopt it until No Child Left Behind.
Some people have looked at whether early adopters made greater gains on measures of achievement than late adopters, to see whether this accountability system was beneficial and the evidence suggests that it was beneficial, that adopting these accountability systems did help improve academic achievement.
But again there’s plenty of disagreement about how large these benefits are, how certain we are about them and so you can’t say in either the area of accountability or choice that the debate is over. The debate will never exactly be over but I think we’ve made progress in understanding how incentives that attach consequences to decisions can improve productivity and we’re seeing some initial signs that these reforms can produce results.
Another large category of questions that I think is very interesting and important is the civic consequences for education. We don’t just have a system of public education to produce high achievement, which then we hope leads to economically useful skills. We also have a system of public education to produce engaged citizens, who will hold the values necessary for our democratic system to function properly. This topic receives significantly less attention than the achievement question but I think it’s also quite important.
Q: You have a rare and exciting opportunity to form a new department with exceptional support for a cohort of endowed faculty chairs. You have already filled several of these chairs. Gary Ritter holds the chair in education policy and directs the Office for Education Policy. Robert Costrell has recently assumed the accountability chair and Patrick Wolf the chair in choice. The chairs in teacher quality and leadership remain to be filled. What mix of expertise and interests do you look for in the faculty who join the department of education reform?
We have hired two new people for the chairs in accountability and choice…The Endowed Chair for Accountability will be Robert Costrell who is the chief economist for the Commonwealth of Massachusetts, an education advisor to Governor Mitt Romney and a professor of economics at UMass Amherst.
The Endowed Chair for Choice will be Patrick Wolf who is professor at Georgetown University and has the government contract to evaluate the voucher program in Washington, D.C. and is also leading an evaluation of the voucher program in Milwaukee. He’s a professor of public policy and his degree is in political science.
Both Bob and Pat have doctorates from Harvard…
We’re looking for extremely bright, hardworking people, and I think we’ve found people like that. But also we’re looking for people whose work is both academically rigorous and easily accessible. This is a very hard combination to find. We want people doing the best, cutting-edge work on these questions and who have an interest and ability in communicating their results to policymakers, educators, and the general public, because after all, finding interesting answer to questions is not very helpful unless you tell people about it and you have to tell people about it in a way they can understand.
It’s very important to our department to find people who have both of these skills. We’re not there to do this work for our own amusement. We are trying to make a difference by conveying what we learn to others so that they can use it. Or not. We’re not tyrants so we’re not trying to impose any particular set of findings on people but we’re trying to add information to the discussion.
There is a difference between doing and knowing, and I think that one of the difficulties we’ve had with education reform and I think one of the arguments for having a separate department of education reform – I mean, arguably the whole college of education is in one way or another engaged in education reform. Everyone’s trying to make schools better, they’re trying to make schools better though in a very direct way, the training of future educators, future principals, and that’s important. But it’s also important to step back and look at the systems that shape the behavior of the educators that we’re training, and so there is something to be said in this department in emphasizing these systematic forces and putting somewhat less emphasis on direct experience in the schools.
So similarly in an army you need very well trained soldiers and you need people to train those soldiers and equip them with all the right gear but then you also need strategists who think about how you use the soldiers for the right effect. I think we need to be doing both in the college of education, and I think this is part of why this new department was created.
Q: Tell us about what research the department has already undertaken and what’s on the horizon.
We have our hands in so many things that I actually had to write down a list so that I wouldn’t forget too many things. I’m sure I’ve forgotten some things even with a list.
One thing that we’ve done already as a department is develop what we call the School Performance Index, which we released in January, and it’s an attempt to measure the quality of schools in what we think is a more precise way using current accountability testing.
The difficulty with current accountability testing is that it often tells us more about the demographic characteristics of the students in the school than about the quality of the school itself. So schools with advantaged students are said to be good schools when they may not be good at all. They may be mediocre but with very advantaged kids. And other schools with low test scores are said to be bad schools but they might actually be good schools bringing up student with severe disadvantages and bringing them to higher levels of achievement.
So we attempted to disentangle these demographic characteristics from school quality characteristics with the School Performance Index. It’s a work in progress. We received a fairly big reaction to it and a fair amount of criticism, but I thought it started a very good conversation about how it is that we really measure school quality, because after all, if accountability systems are going to work they have to attach rewards and sanctions to the thing that really matters and not to the things that are outside of the control of the schools.
Otherwise we simply incentify schools to find advantaged students and avoid disadvantaged students and we don’t want them to do that. That raises serious equity concerns.
So that was, I think, one important thing that we’ve done already, in addition to recruiting. We are now working on revising it and incorporating some of the criticisms to improve it and expanding it to other states, and eventually we hope to have this nationwide. We think it’s a very handy metric for assessing school quality. It’s not perfect, and as we receive more criticism we’ll revise it again and try to make it better. It is important that we have good tools for tracking how our schools are doing. So that’s one thing that we have already done as a department. Everyone in the department was involved in the creation of the School Performance Index.
Also, we’ve done some other research on policies that exist in other states but have potential application to Arkansas. For example, Marcus Winters and I have published research on the effects of ending social promotion or what’s called test-based promotion. This is when students are promoted from one grade to another if they can pass a test or if they receive a certain exemption by a specified process.
We’ve looked at this policy in Florida and analyzed the results and published those results in a journal [http://www.educationnext.org/20062/65.html], and we’re actually in the process of looking at the second year of that program so we can see whether the initial positive effects of that program that we observed hold true.
In the future if they reverse over time and it’s conceivable that they do – there was a result of a study in Chicago of their similar program. There were positive results after one year but the results started to turn negative after two years. We’re very eager to find out whether Florida’s results are more positive or whether they’re consistent with the Chicago results. And we also can see that there are certain differences between the policies in Florida and Chicago so if one is more beneficial than the other that might tell us about how to design these programs more effectively.
Gary Ritter and I have done a study on integration in public and private schools which is an area of research which we each have done separately before joining this new department. In fact, what happened is that I did a study and Gary replicated it and found that opposite result, and that’s how I learned. I didn’t know Gary before then but I came to know Gary because he had refuted a study that I had done with very good data and very good methods but was using a different data set. It was based on a different grade, and so we had possible explanations for the discrepancy in results and we wanted to do a project together to see if we could resolve the question. And we’ve done that.
So we have a new study coming out that will we hope help resolve the disagreement that we had on that question. The sneak preview of the result – Gary has to write this up; he’s taking the lead on writing it – but the sneak preview of the result on that research is that we partially confirm his earlier results and partially confirm my earlier results. We find that private schools are somewhat less well integrated than public schools, as Gary had earlier found but that the financial barrier of tuition seems to explain the difference in integration. So that if we control for that tuition, then public and private schools are equally well integrated. In other words, there’s no special desire on average in private schools to avoid integration, but there is the net effect of financial barriers preventing as much integration as there would otherwise be.
I do have a new book project underway to follow up on “Education Myths” and the tentative title is “The Carrot and the Stick.” If “Education Myths” was primarily about how we misunderstand the nature of the problem in education, the new book is largely about the promise of incentive-based reforms.
Q: In the field of education research and reform, who would you call a mentor or model? Who is a respected critic, one who stimulates your thinking?
My mentor and advisor was Paul Peterson. He was a professor at Harvard while I was a graduate student there and was my advisor. It was with him that I did my initial work on education policy. I feel a very large debt to Paul.
He produced actually a number of students that are involved in this area including Patrick Wolf, who was also a student of Paul Peterson; Rick Hess, who is a fellow at the American Enterprise Institute; Marcie Canstrom, who’s at the Fordham Foundation; Ken Wong, who’s at Brown University; Marty West, who’ll be joining him at Brown University. So he’s produced a cadre of a new generation of education researchers. I think that’s evidence of his excellence as a mentor and teacher.
There are other people in the field that I feel assisted me and inspired me. They include a variety of people: Checker Finn, who is the president of the Fordham Foundation and was assistant secretary of education, is someone that has helped me along in a number of ways. Terry Moe, who’s a professor at Stanford: his work has always been interesting to me and he’s always a helpful person to bounce ideas off of.
But also some people outside of the area of education have been great mentors as well. For example, Morris Fiorina, who was on my dissertation committee at Harvard and is now a professor at Stanford: He taught me a lot about myth-busting, how to take on a conventional wisdom with dispassionate evidence and how that can be done productively. I looked to Moe Fiorina as a model for that.
And James Q. Wilson, who wrote the forward to my book and actually held the same endowed chair that Paul Peterson currently holds. He is the prior holder of that chair but is now in California associated with – I think he’s emeritus now with UCLA and Pepperdine. I don’t know how that works that he is in more than one place but if you’re James Q. Wilson, you can do it. I think that what James Q. Wilson did for me is that he provided a great example for how one can both engage in rigorous academic work and be engaged in public discussions of policy.
Now as to critics that I respect, one is someone I mentioned earlier, Hank Levin, who’s a professor at Teachers College. I always like listening to what Hank has to say on my work and on the questions of education reform. I think he’s been an excellent critic of for-profit education companies and some of their limitations, and I think that he also is just a useful gadfly for those who have too much confidence in market forces. He readily points out lots of inefficiencies in markets, particularly in education. He’s someone whose work I respect a lot. Rick Hess at the American Enterprise Institute similarly has been fairly critical of the ability of markets to operate effectively in education. He’s someone else that I pay close attention to.
Q: In your experience, what are some differences between think tanks and universities?
Actually I think done properly the two are no different. They’re the same thing, if you do it well. That is, you want to engage in rigorous work that is influential in public discussion. That’s what think tanks are mostly about, and that’s what universities ought to be about, try to be about. Now both think tanks and universities fall short of this ideal. Think tanks sometimes are lacking in the academic rigor that they need for their research and sometimes universities are lacking in the ability to insert their research into public discussion. University work done properly and think tank work done properly has a nice marriage of both. That’s our goal in the department, to do both. So I think that there really isn’t that much of a difference between the two.
What are the advantages of being at a university? One is there’s a certain amount of respectability that is attached to universities because of, again, their reputation for rigor. But also, colleagues. There are real benefits to having a critical mass of people all together thinking about related questions and talking to each other. I find that really beneficial.
Similarly, students. Students are colleagues in the making, and I learn a lot from students. I enjoy teaching. We’ve hired a bunch of people, both undergraduates and graduates, to work with us and we’re looking to hire more. They’re extremely helpful and I always enjoy working with students on projects. I regularly co-author pieces with students. That’s how I was trained is that I was essentially in an apprentice program, and that’s how I’m attempting to train our students, and they’re fantastic. That’s working out very well.
One thing that people might think is a particular advantage to universities that I don’t think is an advantage is academic freedom. People think, wrongly, that you’re much more free to say what you want in universities than in think tanks, that in think tanks somehow you’re beholden to someone and can’t speak your mind, while in universities you’re completely free with tenure to say whatever you want. Neither is exactly right.
Universities have all sorts of limitations actually on effective free speech, in that if you wish to continue receiving resources from the university you have to control what you say to some extent. Also we’re a public university, and there is the specter of political involvement in our operations, and while no one is specifically threatening that now, it’s always a possibility so that puts some restraint on our ability to speak freely. On the other hand in think tanks, the thing that gives a fellow at a think tank a fair amount of freedom to speak his or her mind is the fact that you don’t have to work for that think tank. You could go work for another one. There’re lots of them, and they have lots of different kinds of people and endorse lots of different ideas, and if you say something that one group doesn’t like and it’s an important enough difference to them that they don’t you to work for them anymore, then you work for someone else. It’s not such a big deal.
Think tanks are looking for bright, energetic fellows, and they don’t try to control what they say for the most part, at least in my experience. They just try to let you go and say what you want to say. Of course, they select people who they think do the kind of work that interests them, but so do universities. So I really don’t see nearly as much a difference between think tanks and universities as other people might think. My experience—I mean I was in academia for many years before going to a think tank for a number of years and then back into academia, and to me they feel very similar.
Q: When your children grow up and walk into a public school as parents, teachers or researchers, what would you like them to see and experience?
I want them to experience an effective institution, but here’s where I think a lot of people go astray in education reform is that they think that there is a successful model, that if only we were to find it then everyone should do it. I’m actually of the belief that there is no single successful model for schools. There are many different successful models, and there are many different successful models for different circumstances.
Many years from now when my children enter a school as a parent or educator, I expect that they’ll see very different things in different circumstances, if we do it right, because the notion of the best practices, which has been imported to some extent from the business world, can be misapplied in education.
I’m not convinced that there is a best practice. What we end up with if we pursue that too far is we have system that’s one size fits none, and that’s the real danger in education. I know that children have very different needs, and what they need is a school that will be effective for them. It’s not always easy to achieve, it’s certainly not easy to achieve, but that’s our goal.
I guess what I’d want my children to see is just a school that is effective for the students that are there. Exactly how they do that – it will depend.
We increasingly have the technology to customize our education, and we’re not using that technology at all. And why are we not using it? Because there are no rewards or penalties, there are no consequences, out of bothering to innovate, to use the technology effectively. So we have computers in schools, we just don’t use them for anything useful, because it doesn’t matter whether you use them well or not. Again, it’s not that the people in the schools are bad; it’s that it requires some motivation to do something different.
A typical teacher, unfortunately, will do the same thing for the rest of his or her career that he or she did the first year of teaching. The same lessons, the same approaches forever. No matter what kind of school they teach in, no matter what kinds of kids. They’re comfortable with that. It works for them as they see it. But the goal is not to have it work for them; it’s to work for students. It’s useful to be reminded of that by external rewards and sanctions.
All of us, even the best willed of us, do better when properly motivated, when recognized for putting in that extra effort.