A snark-free critique of Ladner’s “New Millenium Schools”


by David Safier
Matthew Ladner released his 28 page pamphlet, "New Millenium Schools" last week. I find it to be a serious, intelligent, well researched document which, while I disagree with it in a number of places, presents a plausible alternative educational strategy. This is an action document that suggests a model for a charter school that would attract top quality teachers by paying them high salaries while increasing class sizes. There's no politicking by Ladner here. This is an honest proposal, so I'll give it the respectful, careful scrutiny it deserves.

Ladner's basic proposal is for data driven merit pay on steroids. He wants to reward teachers based on the amount their students grow academically while in a teacher's class compared to their progress during previous years. Teachers whose students show a high rate of growth are rewarded with higher salaries, and on top of that, all teachers are rewarded if an entire school meets an overall growth goal. Teachers who have proven themselves to be "master teachers" can request extra students in their classes and get a bonus for each additional student. In this school, teachers have the potential of earning six figure salaries.

The hoped-for result is a school which is a magnet for excellent teachers. Since class sizes will increase, teachers will earn higher salaries than in other schools without an increase in the cost per student.

This is a simplified but, I hope, reasonably accurate summary of the basic idea Ladner proposes. I would be genuinely interested in seeing an Arizona charter work toward this model. If the theoretical model works in the real world of the classroom — and that's a big "if" — it could provide a viable educational alternative for some students. That's one of the strengths of charter schools. A potentially good idea doesn't have to be good for everyone. It only has to work for the students enrolled. And if it doesn't work well, the charter school can shift gears, or close.

That being said, I want to look at Ladner's ideas more carefully and with a more critical eye after the jump. Click on the link if you're interested.

Ladner's premise is based on the philosophy of fiscal conservatism. He wants to improve education without spending any more money. This tax-and-spend liberal, on the other hand, believes that improving education will take money combined with intelligent improvements. Our philosophical divide drives a great deal of my criticism of Ladner's plan.

We both agree, there's nothing more important than a good teacher. If I had my choice of a great teacher in a mud hut with 40 students or a mediocre teacher in an educational palace with 25 students, I'd choose the mud hut every time. I taught in schools whose faculty contained both gifted, intelligent teachers and mediocre, not-so-intelligent teachers. Believe me, there's a huge difference in the education students receive from the two types of teachers. True, brilliant people can be lousy teachers, and some people with average intelligences have a gift that makes them good teachers, but among those who pursue teaching, you're going to find that high intelligence trumps average intelligence 95% of the time.

We also agree that the U.S. schools have far too many mediocre teachers. Our teacher pool draws heavily on the middle of the college pack, or lower. It's damn near impossible to make significant educational improvements without bringing more of our best and brightest into the field. Not only that, but we have to make sure the top teachers are in schools with students from all economic and social situations. It's unfair to have the best teachers concentrated in schools with upper income students, as happens too often today.

It's when we get to the topic of class size that Ladner and I begin to part company. He's careful not to say that class size is irrelevant. He just says it's insignificant when compared to the quality of the teacher.

Ladner implies it's common knowledge that class size doesn't matter much, while the truth is, the class size battle rages on in educational circles. A researcher Ladner cites on a non-class size issue, Erik Hanushek, is the patron saint of the "class size doesn't matter" school. A few decades ago when most people agreed that smaller class size made for better education, Hanushek gathered some data, crunched numbers and "proved" that students' achievement wasn't affected significantly by the number of students in a class. A cadre of researchers formed around him, citing his studies, then citing each other's papers on his studies, until it began to look like their research had reached critical mass — that is, until another reseacher found she could crunch the same numbers a different way and "prove" that class size does indeed matter. Since then, more carefully controlled studies have indicated that class size is important, and it's most important for students on the low end of the educational spectrum. Note I say "incidated," not "proved." My point is, many people, myself included, believe class size is one of the important determining factors in high quality education, especially with those who need lots of personal attention to overcome educational deficits.

Ladner wants to increase teachers' salaries to attract higher quality teachers. So do I. The fiscal conservative in Ladner wants to increase class sizes to offset the salary increases. Not me. I want to spend more so we can pay teachers higher salaries at the same time we make teaching more attractive by improving conditions for teachers, which includes lowering class size.

Another area where we disagree is in the importance Ladner puts on data collection. (Here, I have to admit, he's in the company of Arne Duncan, Obama's Ed Sec, who is also big on data driven education, while I'm a data skeptic.) Ladner wants test scores to be the main force driving teacher salaries, both the scores of an individual teacher and of the school as a whole, with the prinicipal having the authority to boost salaries a bit based on more subjective decisions about teacher quality.

If you think our schools are driven by testing now, it's nothing compared to what would happen if students' scores determined teachers' paychecks. High stakes tests, meet high paycheck tests. Expect non-testable material to be treated like unwanted step children by teachers trying to get ahead on the salary schedule. Expect drill, drill, drill, followed by practice test after practice test.

Ladner praises South Korea's education system and the high test scores of its students, but a closer look reveals a test-driven pressure cooker, with parents and students reaching unhealthy levels of obsession and outright panic about test scores. That's one reason why some South Korean mothers take their children to other countries to be educated, while their husbands remain at home. The parents complain about South Korea's "rigid educational system," the schools' "educational pressure cooker" atmosphere and their failure to teach "other skills crucial in an era of globalization, like creative thinking."

Merit pay sounds like a good idea in theory, but deciding what is meritorious in teaching is unbelievably complex. Our ability to create standardized tests that gauge a student's educational attainment is very rudimentary. They are a crude tool, a blunt instrument. Standardized tests have a place on the educational table, but we put them in the CEO's chair at our peril.

Finally, I'm not sure if this theoretical school will be as good in the classroom as it sounds on paper. With a select group of cooperative, motivated students — you can be selective at a charter school if you use finesse — it could possibly be quite successful. If the school were more heterogeneous, drawing from a wide socio-economic and educational range, I'm not sure things would go quite as well as Ladner hopes. But I would like to see this model succeed, especially if it demonstrates the importance of drawing more of our top level college students into education.

A less important comment.

Ladner doesn't say this specifically, but I think his model envisions a K-6 school. His assumptions about testing and salary seem to be based on one teacher with a single class of students for the entire day.

If that's true, I question his statistic about class size in South Korea. He says they have 49 students in each class. I've found articles that say their elementary school classes have more like 30-35 students. He also uses 20 students as the average class size here. I'm not sure, but this figure may come from confusing the student/teacher ratio with class size. The term "teacher" usually includes curriculum specialists, librarians, counselors, etc. The average number of students in the classroom is always higher than the student/teacher ratio. When I retired, I had 30-35 students per high school English class. I'm willing to be corrected on either of these assertions, but if I'm close to right, that means the difference in elementary class sizes between South Korea and the U.S. is less than Ladner claims. (It may be that he's using actual class size in South Korea and student/teacher ratio here.)

And a few last thought questions.

1. How can you use data to judge the effectiveness of teachers in the earliest grades — K, 1 and 2? It would be difficult to judge students' progress over such a limited time span.

2. Many schools have large transient populations where as many as 50% of the students at the end of the year weren't there at the beginning. This would vastly complicate the use of data to evaluate teachers. This probably wouldn't be a factor with individual charters, but if this method were generalized to school districts, it could cause problems.

3. I don't see how this system would encourage top teachers to teach at low performing schools. It's a hell of a lot easier to show progress when most students are at or above grade level, so if teachers want that merit raise, their odds are still better if they teach at high performing schools. To make a valid comparison of teacher quality, you would have to compare, say, a year of progress at a high performing school with 2/3 year of progress at a low performing school. (A third grade student with a second grade reading level has made 2/3 of a grade level progress each year, while one with a third grade reading level has made one year's progress per year.) So top teachers in the Foothills district, for example, should increase their students' performance by 1 1/3 year, while the equivalent teacher in a lower performing urban school should only be expected to show one year's progress. Ideally, you want the children who are below grade level to make large leaps forward, but it isn't fair to base a teacher's salary on that expectation unless you want to scare them away.


  1. David-

    It should not be thought of in purely monetary terms. Teaching is a profession with many intangible rewards, and summers off. If I can couple a low-six figure salary along with these rewards to get the kids who would make more in law or business, I think it’s a good deal for kids and taxpayers.


    There is a charter school in NYC that announced it would be opening and providing not only $125k base teacher salaries, but also up to $25k merit bonuses. New York spends a good deal more per pupil than Arizona does, but I suspect that they are doing something similar to what I describe, but I haven’t tracked it down yet.

  2. I finally got a chance to read through the full report.

    I think if this system would work well her in the US, you would probably see it being used in private K-12 schools. Are there any “real-world” examples of this in the US for K-12?

  3. Matthew.

    Why should reward and punishment be thought of in purely monetary terms? After a really good day of teaching when I had a clear breakthrough with a student and/or a class, the high I experienced beat what I would have felt if someone slipped a hundred dollar bill into my hand. Give teachers a respectable salary worthy of their education and importance and give them a bit more respect and prestige, not six figure salaries. I think you’re confusing the more mercenary world of business where profit and financial gain are king with the social sector world where people genuinely want to make a difference. I doubt that the MBA mentality would make someone a “great” teacher, even with merit pay.

  4. David-

    I agree it’s not just about money. Teaching is a profession that does little to reward success, nor much to punish failure. This has got to change, in addition to the pay structure. The essence of professionalism is to be rewarded and punished based on results.

  5. Matthew.

    Not much difference here, except for the fiscal conservative vs. tax-and-spend liberal axioms we apply to the world. Which is, of course, a huge difference.

    Assuming there’s an educational plan, not just money, I would like to see a substantial general increase in teacher salaries, with a genuine bonus attached to teachers going into the low performing schools. I would add a two week paid work session for all the teachers in those schools during the summer, where they could talk and plan without the daily pressure of lesson plans. The bonus would also add a layer of prestige to the choice of teaching at one of those schools, assuming there’s competition for the positions, which I think there would be. If you can prove yourself a good teacher in a difficult situation and get rewarded financially at the same time, you’ve earned a couple of stars on your lapel, which you can wear with pride.

    More people would consider teaching if it weren’t a financial hardship. $100,000 a year would be nice, but not essential. Simply having a salary that allows a teacher to buy a home and support a family along with a pension plan that rewards you for your years of work would go a long way toward attracting people who are drawn to teaching but don’t want to make the financial sacrifice.

  6. David-

    I like you also have reservations about the term “combat pay” but have reluctantly used it because it is a part of the jargon at this point, and “bonus pay for working at a low-performing school” doesn’t exactly roll off the tongue. Although I’m happy to avoid the term, a new one would be good.

    Now to answer your question, I’m asking you to think of the recruitment of high-quality teachers outside of the current limited supply. Don’t just think about the distribution of the ones we have now, think about how many more highly capable and ambitious people you can attract with strong professional wages.

    If charter schools following the model I described opened in low-income areas, they would attract people who would have otherwise not been willing to stay in teaching or never would have considered the profession in the first place.

    Competition for teaching jobs in South Korea is fierce because they pay their teachers 2.48 times gdp per person. If the United States paid 2.48 gdp per person, veteran teachers would be making over $116,000 per year.

    Now of course extra money for children attending schools in low income areas is not something I would advocate, as my belief is that schools have plenty of money as it is. If however you progressives prevailed in providing it, you might not only improve the supply but also the distribution. The main issue however is the supply, although I also think the distribution is a very worthy topic, and in fact a King Kong sized equity issue which we have only begun to recognize.

  7. Matthew.

    I loved that piece on Shane Battier as well, and your use of it to discuss merit pay is right on the mark. You’ve acknowledged how hard it is to create a decent merit pay system. If you add my concerns about the limited and limiting nature of standardized tests, you end up with someone like me, who is very skeptical. Then add my concerns about what a poorly designed merit pay system would do to teachers’ ability to work together, which is phenomenally important in middle and high schools where teachers share students, and my skepticism goes through the roof. I need to see a system that will do more good than harm. I haven’t seen it yet.

    I find a logical fallacy in your “combat pay” comment. Before I begin, though, I think you should find another term. “Combat pay” may be both clever and somewhat accurate, but it also implies that teachers are walking into a war zone where the students are the enemy combatants. That’s exactly the kind of stereotype we want to dispel. Something like “bonus pay for working at a low performing school” or “working with at risk students” (especially since “at risk students” can be at high performing schools) sounds better to me. I don’t think this is being PC. I think it’s being sensitive to the connotations of language.

    But back to the topic. You say a paltry bonus won’t get a teacher to move from Foothills to a low performing school. Then you say the merit pay plus a bonus would. That doesn’t make sense. If, for instance, right now with current salary levels, TUSD offered a $10,000 a year bonus for working at certain schools, this ex-teacher thinks you would get a whole lot of takers. If you made the same offer to teachers in your merit pay system who knew they could make, say, $90,000 at Foothills or $100,000 at a low performing school with the bonus, they might feel the $90,000 was fine, and the bonus wasn’t worth the extra work and psychological drain. So the merit pay bonus actually dilutes the value of the bonus for working at a low performing school.

  8. Note to whoever is posting using another person’s name and making stupid, derogatory comments: please stop. It’s inappropriate and detrimental.

  9. Mr. Huppenthal,
    Motivation is important, but excellent teachers can work around that. The teacher is the most important variable.
    You are referring to Reading First, in your comments about the “national experiment” I believe. My school was part of that grant. Our scores have grown each year for six years in K-2 using DIBELs as the measure. DIBELs is required in AZ in grades K-3.

  10. David-

    I very much appreciate your review of the report. I will try to address some of your concern/comments.

    First, trying to do merit pay is tricky business. One must be very, very careful about one chooses to reward. I’ve written about this since turning in the draft for this study, and I would actually be inclined to have any merit bonuses other than variation in class size determined by schoolwide, communal goals:


    Class size bonuses must be exempted however because we must give more students the opportunity to learn from high quality instructors, and make use of the financial leverage of class size variation to attract ambitious students.

    On your comments on the class size literature, about the only study that I am aware of that shows a significant result for small classes is the Tennessee study, and it didn’t find much of one, and only in the early grades. No need to argue over regression coefficients, however, as American class sizes have declined substantially since the baby boomers went through the system, and our NAEP scores have stagnated.

    I see this helping inner city kids by expanding the supply of high quality teachers. Some want to use “combat pay” incentives to lure the limited supply of high quality teachers into inner city schools, to which I can only say good luck luring those folks out of the Catalina Foothills with your bonus. It’s not enough, the problem is far too large to address with such small incremental medicine. Combine what I am talking about with serious combat pay incentives, now you are cooking with gas…

    On your K, 1, 2 comment- we have got to start testing these grades. Period. These are the grades where kids are either learning to read, or becoming illiterates. It is simply criminal not to test them.

    Transient students are a problem. You can assign them an identifier number and track them as the bounce around the system. Someone or some group smarter than me will have to figure out how to deal with this very real issue, but ultimately you are going to have make judgements based on the data you have available.

    I’ll try to respond to more of your and reader comments tomorrow.

  11. The one way to get better public schools is to elect the nominee from the Democratic Party and hope like hell that John Hopenfail does not get in.

  12. You have a serious unmeasured quality problem that one of the comments alludes to, namely single year academic gains can be pumped up at the expense of a child’s attitude towards school. Since a child’s attitude towards school is the best predictor of future academic gain, even better than teacher quality, such schools may or may not raise a state’s academic achievement if their single year gains are at the expense of attitude towards schools. The historical record of such systems is abysmal. Great Britain ran a system like this for 30 years before it fell apart.

    In addition to running experiemts that we armchair quarterbacks construct, we should examine the schools that are achieving outstanding gains and shine a spotlight on those schools both to spread their practices and to guide parents to them.

    Unfortunately, right now, we (and 49 other states) have little or no information on the academic gains of kindergarten, first, second and third grade. These grades represent more than 50% of all scale score gain in a child’s education career.

    The psychometrics of these early grades is very tricky. We have just, over the last 6 years, run a national experiment on improving reading outcomes in these early grades and, after spending over $6 billion, we have little or nothing to show for it.

    If we are going to run experiments like Ladner proposes, provide accountability to school districts, guide parents to excellent schools, or illuminate true best practices, we are going to have to accurately measure kindergarten, first, second and third year gains.

    Because of the enormous cultural resistance to measurement at those early ages, we will have to start such system on a voluntary basis. We can start by asking the presumed high flyers to volunteer and providing them incentives to be identified as high flyers.

  13. The most important variable in a child’s education is the quality of the teacher in the room. One year’s academic growth should be the expectation; no excuses.
    This is achieved through effective instruction, academic interventions, quality teacher training, student data collection and a partnership with parents.

    That said,
    America is not Korea. We do NOT have a homogeneous population with common background experiences.

    Americans will NEVER accept 49 kids in a classroom. American classrooms aren’t even big enough for 49 bodies.

    Teacher salaries being tied to achievement is okay, as long as the highest salaries go to the teachers and principals who work in high poverty schools and show one year’s academic growth for their students. One year’s academic growth is measured by a pre-test and a post-test, not a test that measures grade level performance.

    Teacher training in America has missed the boat. I contend that teacher training should be a five year program, with the last year spent as an apprentice in a classroom with a proven, effective, master teacher. Apprentice pay should be about 80% of first year teacher pay. Nothing harms a student more than a first year teacher, (especially in the lower grades where children are learning to read) who doesn’t know what he/she is doing.

    I became a teacher in 1971, when there were few career choices for women. As career choices for women expanded and teaching salaries did not rise to match salaries in other fields, students chose other fields of study. At the same time, teacher training programs around the country failed to use science to train people how to teach. Result? Poorly trained teachers. If you want to improve the quality of teachers coming into the field, revamp the teacher training programs at the university level. Teaching IS rocket science.

  14. Actually, Obama is more for merit pay than I am. So is Arne Duncan. Not very socialist of them, is it?

  15. Obama sees this as a problem when teachers can make more money based on achievement rather than time served on the job.

    He talks the blab of the middle class while at the same time trying to destroy it through taxation!

  16. Another plan that would look good in a magical “Randian” land where hard work automatically equals higher pay, every family is stable, and everyone flys around on pegacorns.
    I wonder why the John Galts like Laddner don’t move to Somalia where they don’t have to pay taxes and try out some of their ideas.
    P.S. I used to teach an Earth Science for non-majors lab, and probably 80% of my 35 or so students were education majors. I would say 50% of them had problems doing simple arithmetic with fractions, and even with detailed instructions most could not consistently do conversions between the English and Metric system. This is really 4th or 5th grade math; until our pool of educators improves we can’t improve our system. I really don’t see raising wages by 10% or 20% as changing the talent pool substantially.
    I think there are a lot of people like myself that would love to do something like teach part time or take a year hiatus from our other jobs and teach and see if we like it, but that is realistically not an option the way the system is set up. This is a case where easing the stringent licensing requirements is needed.

  17. This will work fine in areas with high home ownership . In other areas teachers will not have the opportunity to work with the same students two years in a row, and they won’t have proof of their hard/smart work. Think of the consequences.
    I’ll give it a try in a “problem” school in a poor area and if it works…great!