ENLOW, 1:37 p.m. Monday

So, Jay Mathews and I are slated to have an online discussion this week about school vouchers (thank you Edspresso). For lack of a better analogy, I have been asked to fire the opening salvo.

Rather than pen an emotional piece on all the positive things about school vouchers, of which there are many, I thought I would shoot off a series of propositions and invite Mr. Mathews to fire back. These propositions form the backbone of why many of us to support school vouchers, and I would be interested to know if Mr. Mathews disagrees with any of these propositions, and if so why.

Proposition one: The system is broken.

The system of delivering government funded K-12 education in America is fundamentally broken. At almost every level, public schools, even in vaunted suburbia, are under-performing. Spending has doubled since the 1970s, but test scores have remained flat. Too many children are failing to graduate (these numbers are much worse than people realize). And grades are inflated so much as to almost be meaningless. Everyone who studies education or writes about it knows the system is broken. Few in the public understand how broken it really is.

Proposition two: Most reforms don’t really reform.

Well-meaning education reformers, business leaders and legislators have tried numerous reforms of the publicly run K-12 education system. The public is led to believe these reforms work. They rarely, if ever, do. More money doesn’t lead to better schools. Smaller class sizes don’t lead to higher test scores. Children who go to all day kindergarten don’t perform better in high school. And public school choice doesn’t produce more than a trivial amount of migration between public schools (being told you can eat dinner at any McDonald’s you want isn’t any good when you want to eat at Wendy’s).

Proposition three: Structural reforms that don’t offer private school choice aren’t getting the job done.

Efforts to reform government-run education through charter schools have met with mixed success. There are some good examples of charter schools around the country and some not so good examples. There are problems in securing the up-front financing to build new schools, as well as significant concerns about quality. Moreover, while charter school laws allow some parents some different options they don’t give parents a real choice. That’s because in most places the existing system can veto the creation of charters schools (thus ensuring that no real alternative can thrive) and because the flow of money is still controlled by the same institution that got us into the mess we are in already (see proposition one). At least charter schools can be closed if they don’t do a good job. Another reform, standards and high-stakes testing, has also shown some success, but not nearly enough. And unlike vouchers, this reform is already just about as scaled up as its ever going to get. We have seen the best it can do, and it isn’t going to get the job done.

Proposition four: Limited voucher programs have been shown to work.

Seven random assignment studies of voucher programs around the country have found that children receiving vouchers do better than children who apply but don’t get a voucher. These gains are particularly strong for African-American children. Additionally, a large body of evidence is finding that public schools improve and improve faster in states and cities where voucher programs exist. Also, parents are happier and more involved in their children’s school. And, oh by the way, vouchers also stop the immoral practice of forcing parents to go to terrible public schools by virtue of where they live.

Proposition five: Since limited voucher programs are proven to work, larger and more universal voucher programs are a more promising strategy for large-scale reform than any of the other available options.

Vouchers work the best of any reform because they force improvement in K-12 education by breaking the bonds between the government financing of education and the government administration of schools. We are among the only western countries that explicitly tie the funding of education through taxation, which almost all of us agree to, with the government operation of schools. Basically, we have a government run monopoly. And we know that no matter how many good intentions we have, monopolies simply don’t work. On the other hand, if we ramped up voucher programs to include middle and upper-income families the effects would be dramatic. More children would do better in class and parents would be more involved. Schools would be more responsive to parent needs and more public schools would do a better job of educating our children.

Conclusion: The system is broken. The reforms we have tried so far aren’t working, and some of the newer structural reforms like charter schools have limited effectiveness. Limited voucher programs, on the other hand, have been shown to be one of the most effective reforms ever. Why not try vouchers for everyone?

MATHEWS, 3:53 p.m. Tuesday

I share Robert Enlow's view that low-income students would benefit greatly if significantly more of them had vouchers to attend private schools. I also think our nation's social security system would be much sounder if well-off geezers like me had to pay taxes on our social security checks, and I think my tennis game would be several notches above its sorry state if I went out and practiced occasionally. Unfortunately, because of strong voter preferences for public schools, political fears of offending us old folks and my own congenital laziness, none of those good things are going to happen.

Voucher supporters like Enlow are entitled as Americans to entertain their fantasies and push for a major extension of vouchers. I am not clairvoyant, so they may find a way to convince voters that letting everyone choose private schools with public money is a good idea. But the electoral track record on this is not good, and I think we would be much better off if they gave up that fight and turned their considerable intelligence and energy toward increasing the number of charter schools, which in my view have several advantages over voucher schools besides just political viability.

To take Enlow's good propositions one by one:

1. Anybody who believes the system is completely broken has not spent much time in suburban private and public schools occupied by the children of the great American middle class. Go visit those schools, check out their statistics. Compare a public school---lets say McLean High in McLean, Va.---to a private school with kids from the same kind of families---such as the Potomac School, also in McLean. You find the same great teaching, the same high achievement, and the same flaws of all schools---some not so good teachers, some dumb rules, some resistance to opening up AP to all students. I have compared the ratings of private and public high schools on the AP and IB participation index I use in Newsweek and the Washington Post, and there are no significant differences between schools of similar socioeconomic characters. This is true, by the way, on both the top and bottom of the socioeconomic scale, although of course we cannot go too far down the scale because there are very few private schools that cater to low-income children. Upper middle class kids are learning a lot more than upper middle class kids of my generation. Any parent with kids in AP or IB knows that. The stagnation is in the middle, among average kids of average income not being challenged, and low income kids. But sending them to private schools won't help, since the private schools at that income level do not do any better.

2. Most reforms don't really reform? That's a surprise? That rule covers all human endeavors, including education. Change is slow and erratic. There is nothing about vouchers that is going to change that. The marginal benefits of vouchers in the studies cited do not suggest big gains in a voucherized America.

3. Charters are a better structural reform because they will draw the most motivated teachers and principals. The people who have made the KIPP charter schools work so well are there because they are deeply committed to public schools. You simply cannot get them as excited about private schools, so in this current epoch, you are not going to get the kind of teachers you need in voucher schools to make a difference. That is the charter movement's greatest advantage over the voucher movement.

4. Vouchers work, but not enough to impress very many people. I don't want to ban vouchers. I think they are a good idea, and morally sound. But they just don't have much growth potential.

5. Larger and more universal voucher programs are a sweet dream, but nothing more. Smart people like Enlow should direct their good ideas toward the charter system, and help that grow. Charter schools have left voucher-supported schools far behind, and that gap is only going to grow wider. Maybe in another century or so the political and economic systems will have changed enough to make their vision a reality, but I will be long dead by then, so I am putting my money on charters.

ENLOW, 12:20 p.m. Wednesday:

I want to first say thank you to Jay Mathews for his thoughtful responses and for his kind reference to me as one of the "smart people." Moreover, we don't know each other but I am fast becoming one of those "old folks" he mentions, although I must say that I probably practice my tennis game a bit more than he does.

However, after reading his comments I once again have to say, give me a break.

As I understand it, Mr. Mathews thinks vouchers are a good idea. They have produced some benefits and are certainly the moral thing to do. The only reason that he thinks we should give up the fight is because he thinks they are a pipe dream with a political track record that isn't great, particularly when compared to charters.

As I thought about whether universal vouchers were a pipe dream, I tried to imagine Dr. Martin Luther King, Jr. saying the same thing to Rosa Parks. But my mind rejected it immediately. King didn't think that justice should be denied just because it was difficult to achieve or just because he was facing terrible odds. He knew that it simply wasn't enough for just one person of color to sit at the front of the bus, or even 200 or 2,000 for that matter. Every person of color–every person for that matter–needed to be free to sit at the front of the bus. Like King, Milton Friedman knew that it wasn't enough to offer a voucher to just one poor person, or even 10,000. Every child, regardless of race, income or any other criteria deserved a quality education at a school of his or her parents' choice. This is the systemic challenge we face and the goal we are fighting for, namely educational justice for all. The only way to fully achieve educational justice is to do what Milton Friedman said, separate the government financing of education from the government administration of schools.

So, vouchers would be worth fighting for even if they really were as politically disadvantaged as Mr. Mathews thinks they are. But vouchers, and even universal vouchers, are much more viable than Mr. Mathews thinks. I suggest that he become more familiar with the tremendous success of the voucher movement in the past decade. The idea that vouchers aren't politically viable is ancient history.

When the Milton and Rose D. Friedman Foundation started in 1996, there were only five school choice programs operating in five states. Today there are 22 programs operating in 11 states plus the District of Columbia, and Georgia is about to make it 23 in 12 states. In 2006 alone, five new programs were enacted and six existing programs were expanded. More importantly, in 2007 the state of Utah passed the nation's first universal voucher program, showing once and for all that even though it is difficult to achieve politically, it can be done. No doubt we have a long way to go. There are 122,000 children using vouchers and tax-credit scholarships right now but that is not near enough.

Moreover, the only reason charter schools are more politically viable is because they are no threat to the status quo. Charter schools have left voucher-supported schools in the dust in terms of sheer numbers, but that's because they aren't much better than regular public schools. Dirt is more common than gold, but that doesn't make it worth more – or, indeed, worth anything at all. The proven academic gains from even limited voucher programs are significantly greater than those from charter schools. While the gains from limited vouchers are not revolutionary, the gains from charters are extremely small. There is a lot of high-quality research on charters and it consistently finds that charter schools are only marginally better than regular public schools. That's because at the end of the day, the processes by which charter schools are created is controlled by the same monopolistic institutions that have failed us so far.

I'd like to revisit my original propositions one by one, in light of Mr. Mathews' comments.

1. Mr. Mathews asserts that suburban public schools are just as good as private schools, and that upper-middle class kids are learning more than they did in his generation. And he seems to very strongly suggest that therefore we shouldn't be bothering about trying to improve suburban public schools.

For his first assertion – that suburban public schools are just as good as private schools – the only hard fact he cites is that AP and IB participation rates are similar in schools with similar characteristics. That is just not an adequate measurement of academic outcomes. It is useful to know that kind of thing if we can put it in the context of more important measurements, such as whether students are actually learning to read and do math. But by itself it provides no real basis for any kind of judgment about academic quality.

The 2005 NAEP data don't back up Mr. Mathews' assertions. Comparing suburban public schools with suburban private schools, NAEP shows that private schools outperform public schools in all tested grades (4, 8, and 12) and in both reading and math. The gaps are virtually always greater than 10 scale score points, so the difference is non-trivial.

Having said that, let me also acknowledge that the gap between public and private schools is definitely smaller in the suburbs than in city centers, which are educational disaster areas. The reason for this is relevant to our discussion in a larger way. Suburban schools are not better because they have more money; urban schools now have far more money (adjusted for inflation) than suburbs had ten or twenty years ago, but they have made no progress toward becoming as good as suburban schools were ten or twenty years ago.

So what accounts for the higher quality of suburban schools? A very large body of research has established that suburban schools benefit from the competitive incentives of residential choice – suburban parents have the luxury of choosing to live where the schools are better, so suburban schools can't afford to neglect their students as flagrantly as the schools downtown do. But that's only a very mild form of competition for students, because parents can't choose their homes exclusively on the basis of schools. What if the link between residence and schooling were severed altogether? The healthy competition for students that has made suburban schools reasonably good would then be multiplied ten or even a hundredfold.

I haven't forgotten Mr. Mathews' other assertion, that "everybody knows" upper-middle class kids today learn more than they did in his generation. Like I said, I'm becoming one of the "old folks" myself, so I know better than to ask him exactly which generation he's talking about. But the data show, as they do so often, that what "everybody knows" is not true. The only decent measurements of academic outcomes we have that can track changes over time are the graduation rate, the NAEP and the college admission exams (SAT and ACT). The graduation rate peaked in the 1960s and has been in a slow decline ever since. The NAEP is flat as far back as it goes. The college admission exams – which are only taken by about one third of all kids and are thus the closest thing we have to a measurement of "upper-middle class" kids over time – are flat, except for SAT verbal scores, which are slightly down.

Federal data do show that since 1990 students have been taking more "core" academic coursework in high school – English and math rather than drama and shop. Participation in programs like AP is up. No doubt this accounts for Mr. Mathews' observations about kids today having what appears to be a more rigorous academic curriculum. But the end results of this coursework are still flat on all the measurements over time that we have. They're more likely to go to math class, but they're not more likely to learn math.

2. Mr. Mathews says change is slow and erratic. Of course it is. But he is mistaken in suggesting that vouchers are no more promising than any other reform. And he is dead wrong in saying that the "marginal benefits of vouchers in the studies cited do not suggest big gains in a voucherized America." They suggest exactly that. If these limited programs, mired with all the restrictions that were necessary to get them past suspicious, often union-controlled state legislatures, can produce gains at all is amazing. It suggests that if vouchers can produce moderate benefits with a ball and chain tied to them, they will more likely than not produce significant benefits when allowed to flourish more broadly.

I would also ask: If the proven benefits of vouchers are too small to merit our attention, then what should we say about the much smaller academic benefits of charter schools?

3. While I disagree with Mr. Mathews that charters are a better structural reform than vouchers, I think the debate about which structural reform is better is somewhat silly. We are in an educational crisis and the charter folks should support the voucher folks, and the voucher folks should support the charter folks. (As an aside, opponents of real structural reform love it when reformers take sides against each other and bicker, because they know that a house divided against itself cannot stand.) As to the specifics of Mr. Mathews comments, I disagree that the main motivation of the KIPP folks is their deep attachment and commitment to public schools. Seems to me that public schooling, in the form of a charter school, merely provides the means by which to achieve their end goal, which is the education of children, particularly disadvantaged children. If vouchers of substantial dollar size were available, let's say the same size as the per-pupil amount provided to charter schools, then I think the KIPP folks would be just as excited. They would lose nothing, and would gain much more freedom from the stifling control of the monopoly system. As to the "charter movement's greatest advantage over the voucher movement," namely attracting teachers, I think that this is like saying we more people drive Volkswagen Beetles than next year's yet-to-be-released Subaru. Of course there are. One car is already in circulation and the other isn't yet.  Additionally, as I understand it, there is a major problem in securing new teachers and new talent in the charter school movement. This is something the entire charter community recognizes is a problem, and it seems to undercut Mr. Mathews' argument.  

4. I am obviously overjoyed that Mr. Mathews supports vouchers and thinks that the only problem is that they don't have much growth potential. I wonder how many people said that to Bill Gates when he dropped out of college to start his own software business, or to Steve Jobs when he was toiling away on the Apple in his garage.

5. As the new universal voucher program in Utah proves, larger and broader programs are not a pipe dream. They can be achieved. But still, I agree that voucher supporters should also be willing to work with charter schools. Charters schools prove that schools can be created quickly in a more market-like setting. I won't be putting all my money on charters, though. It's more likely that I will put about 25 percent of my assets on charters, knowing the large cap fund will always get me a small but steady return. But I will put the rest of my money into vouchers, because their return will always be higher.

MATHEWS, 10:20 p.m. Thursday:

I am the last person to want to get in the way of a young man's dreams. I like Enlow's enthusiasm, and wish him well. But he has a long way to go to convince me that the voucher movement has anything like the potential of the civil rights movement of my youth. I had friends in college, as did many people of my generation, spending their summers registering voters in the South. There is nothing like that interest in the voucher movement among college students today. Instead, those kids, like my daughter's roommate, are joining Teach For America, a program that is aimed just at public schools.

He should reflect on why that is. It is more than just the structural and political forces trying to keep vouchers down. Most Americans, particularly young Americans, prefer public schools, and think private schools are for rich kids, or very religious kids. This is, of course, untrue. So is the notion that lottery tickets are a smart investment. But such ideas are part of our culture, and are not easily rooted out. When I see a few hundred college students sitting down in front of a state legislature to protest its failure to create a voucher program, I will change my mind. Until then, I am afraid we are going to have to wait another generation or two before vouchers become very powerful, and I haven't got that long.

As for the differences between suburban public and private schools, it is all about family background. Of course the suburban publics in general do not score as high as the suburban privates. The suburban private schools have almost no low-income students. The suburban publics have a lot. That is why their test score averages differ.

Those stagnant educational outcomes over the last few decades are true, but not the result of schools getting worse. Instead, the number of poor and minority test takers---particularly immigrants---has been growing, pulling down the average on tests like the SAT that used to be given mostly to white boys with college educated parents like me. If you simply look at the level of learning in our best suburban public schools and our most selective colleges these days, producing a boom of bright executives in a dozen new industries, you can see that we middle class types are doing well with the schools we have. We don't need vouchers, which is one more barrier in the way of their future growth.

Remember, I was saying you had to compare public and private schools with the same demographics, like McLean High and the Potomac School, two high-income peas in a nice Potomac River side pod, with less than 10 percent low income students in either one. I wish suburban schools DID fear losing their good students and thus worked hard to keep them, but most of them don't. They look good because they have students whose parents have more money than the kids in the inner city.

There is nothing about the way private schools teach that is noticeably better than the way public schools teach, which leads me to fear that a rapid growth of voucher schools would be as disappointing in many cases as the rapid growth of charters.

I confess to being ignorant of the big gains for voucher students, compared to similar students in a control group, that Enlow mentions. My last look at this was several years ago, with the Peterson study of voucher kids in DC, NY and, I think, Dayton. Blacks in DC showed gains compared to the control group, but there wasn't much of a significant gain in other groups or other cities. Lay those numbers on me, and if they are as impressive as you say, I will write a column about them. For the time being, I like your passion, but I still think it is likely to only bring heartache. We can help all those kids that deserve better schools, but this is not the surest way to that goal.

ENLOW, 11:33 a.m.

It could be said that heartache is the currency of youth – the author and poet Thomas Hardy thought so. I certainly know that my heart aches, as I know Mathews’ does, every time we see just how bad the education is for so many of our nation’s children.

Is there a difference between the civil rights movement and the movement for educational freedom and justice? Sure there is. The times are different. The political leaders are different, as are the self-styled leaders of the minority community – all of whom supported the civil rights movement but few of whom, despite the desperate need in some of their communities, support school vouchers. The demographics are different. The culture is different. And as Mathews says, college-aged children are different.

I guess the boomers are older.

The youthful passion they once had for justice now seems to play second fiddle to their mature desire to add a new kitchen to their three-bedroom homes in suburban Virginia.

But still, there are some similarities. Certainly they have comparable philosophies. The civil rights movement wanted to end years of oppression based on skin color and the movement for school vouchers wants to end years of oppression based on where you live. Moreover, we have seen a growing number of protests and rallies for school vouchers in the last year. Thousands of people in Texas, North Carolina, Ohio, Florida, Utah and South Carolina – much more than the “few hundred college students” Mathews demands to see – have come out to voice their support for school choice, so I guess Mathews will have to change his mind. Additionally, we are seeing growing bipartisan support for school choice, something that also characterized the civil rights movement.

Moreover, I wonder how many of Mathews’ college buddies who joined the civil rights movement were talked to the way Mathews is now talking to me.

Upon reflection, I agree that there is a mystique to the institution of public schooling in America. This is a point that Terry Moe makes quite eloquently. It is an institution that, despite its numerous failings, we have imbued with cultural and historical value. And yes, it will take a great deal of work to root out that mystique.

Public schools were at the center of the civil rights movement. It is where the integration battle was fought. Moreover, it was one of the first institutions to bring diversity to its workforce. This is a main reason why, I believe, many of the traditional leaders in minority communities have been so slow to warm to vouchers.

At the same time, because assignment to public schools is based on residence and because the funding of public schools is tied to property, public schooling afforded many middle- and upper-income families the opportunity to flee the city and create isolated enclaves. These are the public schools that so many of us voters remember fondly and that we have imbued with virtue.

But the idea that people prefer to have their children in public schooling is wrong. They may instill the system with value and believe in its mystique, but in numerous public opinion polls, even the PDK poll as I recall, the public say they would choose a private school for their children if money were not an issue. A recent poll in Georgia, for example, makes this point clearly. When asked “If it were your decision, would you select a public school, private school or homeschooling for your children,” 43 percent of parents chose a private school, while only 27 percent chose public school (16 percent chose homeschool).

As for the data, Mathews makes a number of mistakes. Measures of “low-income” students are not consistent across public and private schools. As Harvard’s Paul Peterson has explained at some length, the number of low-income students in private schools is catastrophically underestimated by most research. All the standard demographic measurements of student income levels depend on participation in federal programs, like the lunch program. And it’s much easier for public schools to participate in those programs than for private schools – the programs are even designed to make public school participation easy and private school participation hard. Peterson has shown that when we use the best available demographic measurements that don’t rely on participation in federal programs, private schools decisively outperform public schools. The seven random-assignment voucher trials establish the same thing, as does a much larger body of other studies using other methods. The old canard that private schools only do better because of demographics survives because it serves some people’s political interests, and for no other reason.

Mathews is right that stagnant outcomes are not the result of schools getting worse. That’s what it means to call them “stagnant.” I didn’t say schools got worse; I said that their results were flat. And I said that because Mathews said schools were getting better, at least for the upper-middle class.

To defend that claim he once again invokes the demographic boogeyman. But, once again, the facts don’t back him up. Yes, the demographics of the student population have changed. And a few of these changes have tended to make schools’ mission harder. But many more of the changes have made schools’ mission easier. If Mathews will search his archives, he will find that he once wrote a column about a Manhattan Institute study by Jay Greene and Greg Forster showing that the demographic trends are much more positive than they are negative – parents’ education levels, parents’ wealth, parents’ residential stability, teenage birth rates, health, crime, drug use, and preschool attendance had all improved. If we adjust our expectations for public school outcomes based on demographics, the adjustment needs to be in the opposite direction.

Mathews also writes that, “If you simply look at the level of learning in our best suburban public schools and our most selective colleges these days, producing a boom of bright executives in a dozen new industries, you can see that we middle class types are doing well with the schools we have.” I have no idea how to test the truth of this claim. Were there no bright executives a generation ago? How do we know that there are more of them now? Were there no new industries being formed a generation ago? How do we know there are more now?

But even if I granted that we’re “doing well” with the schools we have now, that wouldn’t be an answer to my point that vouchers would make those schools even better than they already are. I’m still waiting for Mathews’ response to that.

He writes that “there is nothing about the way private schools teach that is noticeably better than the way public schools teach.” Well, there is at least one thing about the way private schools teach that is noticeably better: they produce significantly higher academic outcomes. The evidence on this point, which Mathews challenges me to produce, has been in the public record for a long time. I’m happy to recall the data and bring him up to speed.

A random assignment study in Milwaukee found that voucher students outscored the control group by 6 percentile points in reading and 11 percentile points in math after four years. A follow-up study by a researcher hostile to vouchers still found an 8 point gain in math. In Charlotte, combined reading and math scores were 6 points higher after only one year. In New York, math scores were 5 points higher after only one year.

These results are in addition to the Howell-Peterson research Mathews cites. He dismisses that research without even considering the size of the gains it found, on grounds that the positive findings of that research were “only” statistically significant for the black students. But he neglects to mention that virtually all the students in those programs were black. If Mathews knows a way to get a statistically significant finding for non-black students when the sample barely contains any, I’d like to hear it. If not, it’s patently unfair for him to throw perfectly good research into the garbage simply because it doesn’t do the impossible. (For the record, in all the other voucher studies cited above, the results are for all students, not just black students.)

The actual results of the Howell-Peterson research were that voucher students in New York had combined reading and math scores 9 percentile points higher than the control group after three years, 9 points higher in D.C. after two years, and 6.5 points higher in Dayton after two years.

These results, as I originally said, are not revolutionary. But given the limitations of existing voucher programs, it’s impressive that they produce these moderate gains. If we took the handcuffs off, vouchers could do much more.

Mathews says vouchers aren’t the “surest” way to improve schools. But surely they are a more sure way to improve schools than by relying on the mystique of the public school and charter schools, which have even smaller gains than vouchers. For myself, I’ll rely on the empirical data – which establish vouchers as by far the most promising reform.

In the end, it is still my hope that mature wisdom of Mr. Mathews will join forces with my youthful enthusiasm to do everything we can, including supporting vouchers, to truly change the system for all children.

MATHEWS, 12:43 p.m.

Enlow is a wonderful advocate for the voucher movement, and this is the strongest case for it I have ever read. But I spend all of my days in regular conversation with educators and parents all around the country, and they are just not talking about the issues he raises. The voucher results are not impressive enough. The programs are not numerous enough. And the people who run private schools, who you would think would embrace such a movement, are just not enthusiastic enough to fuel the kind of change he wants. You can, I suppose, get people to say they would prefer a private school if money were not an issue, but it is. And those same pollsters report that when you ask those same people to rate their local public school, they say they are quite satisfied.

I think Paul Peterson is one of the most brilliant American scholars, and perhaps the best writer in academia, and I always get excited when I read his work. But his effort to break through the considerable difficulty of assessing how many low-income kids of the public school sort actually attend private schools does not quite make it for me. We need more research on that question. And the latest federal mega-study on this issue seems to buttress the notion that public and private schools produce pretty much the same results when they have the same kids.

I am a journalist, not a scholar. My passion for the last 25 years has been to find schools that do the best job in raising the achievement of low-income children. The overall data that Enlow cites is useful, but tends to homogenize many different schools into one big average. I prefer to look at the star schools, the ones that break new ground, and describe what is going on inside them. On that reporter's measure, the voucher schools are doing great work. I have visited several of the D.C. private schools serving low-income kids with vouchers, and there is no question that they are raising achievement and attracting good young teachers.

But they are also struggling with all the downside issues of private schools, particularly fundraising, salary scales, leadership, class bias and unhelpful traditions. When I compare what they are doing, and struggling with, to the dynamo atmosphere and even better results at the best of the charter schools, like the KIPP schools here in D.C., it leads me to think the KIPP approach has a better future. They can pay their teachers more. They can support a longer school day. They can attract more good leaders and many more bright young aggressive teachers. And since they are creating whole new schools with whole new cultures, they do not have to wrestle with the way things were done in the past. It is in the KIPP schools, and the Teach For America membership from which they recruit many of their teachers, that I find that civil rights fervor most alive and well in this country, and that makes me feel very optimistic about the future.

That is the way things look to me from the classroom level. I admire Enlow's grasp of the major statistical data---which I would like him to send me, as he has promised, to further my education. I think there is a future for vouchers. But it is going to be a long slow climb, similar to what women's rights advocates faced in the 18th century, and I don't think either of us will live to see voucher schools blossom in the way he hopes.

I am almost always the most optimistic person in any group. My wife refers to me as the Pollyanna from Hell. But in this case, although I will continue to cheer for vouchers, I am betting my money on charters, and hoping in some ways both movements can reinforce each other.

ENLOW, 4:50 p.m.

I want to thank Mr. Mathews for a wonderful debate, which was full of spirit and civility. I know I learned a great deal, and I certainly understand being the most optimistic person in any group!

It is certainly my hope – and I will do everything in my power to act on this hope – that both movements can reinforce each other. More choice is better than less choice. Good schools are better than not-so-good schools. And, in my opinion, more choice means better schools.