Using test scores to evaluate teachers

by Harry on August 30, 2010

At a meeting of teacher’s union chapter leaders I attended recently to talk about Race to the Top, I was struck by two things: one was how open they were in private about the fact that current ways of evaluating teachers are appallingly bad; the other was how hungry they were for a clearer understanding of how evaluation of teachers using test scores (one of the things States were strongly encouraged to include in their Race applications) would work. I gave my modest attempt to explain how it would work and why it was a bad idea. Now, fortunately, they can discard my critique, and get the real thing. Authors including Richard Rothstein, Helen Ladd, Diane Ravitch, and several eminent psychometricians (including Richard Shavelson, Ed Haertel and Lorrie Shepard) have made an unanswerable (but, as the authors certainly know, eminently ignorable) case against using test scores, even value added modeling methods, to evaluate teachers (here). Here’s the executive summary:

Every classroom should have a well-educated, professional teacher, and school systems should recruit, prepare, and retain teachers who are qualified to do the job. Yet in practice, American public schools generally do a poor job of systematically developing and evaluating teachers. Many policy makers have recently come to believe that this failure can be remedied by calculating the improvement in students’ scores on standardized tests in mathematics and reading, and then relying heavily on these calculations to evaluate, reward, and remove the teachers of these tested students.


While there are good reasons for concern about the current system of teacher evaluation, there are also good reasons to be concerned about claims that measuring teachers’ effectiveness largely by student test scores will lead to improved student achievement. If new laws or policies specifically require that teachers be fired if their students’ test scores do not rise by a certain amount, then more teachers might well be terminated than is now the case. But there is not strong evidence to indicate either that the departing teachers would actually be the weakest teachers, or that the departing teachers would be replaced by more effective ones. There is also little or no evidence for the claim that teachers will be more motivated to improve student learning if teachers are evaluated or monetarily rewarded for student test score gains.

A review of the technical evidence leads us to conclude that, although standardized test scores of students are one piece of information for school leaders to use to make judgments about teacher effectiveness, such scores should be only a part of an overall comprehensive evaluation. Some states are now considering plans that would give as much as 50% of the weight in teacher evaluation and compensation decisions to scores on existing tests of basic skills in math and reading. Based on the evidence, we consider this unwise.

Any sound evaluation will necessarily involve a balancing of many factors that provide a more accurate view of what teachers in fact do in the classroom and how that contributes to student learning.

Read the whole thing for the details.

{ 166 comments }

1

Brett Bellmore 08.30.10 at 10:54 pm

“such scores should be only a part of an overall comprehensive evaluation. Some states are now considering plans that would give as much as 50% of the weight in teacher evaluation and compensation decisions to scores on existing tests of basic skills in math and reading.”

50% isn’t a “part”? They make some good points in the body of the report, but it puzzles me why the executive summary didn’t discuss any of them. You could encapsulate the the report in one sentence: “Changes in test scores do not have a statistically significant predictive value for individual teachers.”

The problem being that changes in test scores are, essentially, what we’re trying to produce. Or at any rate the best proxy we have for it. Do any of the other proposed measures of teacher performance demonstrate significant correlation with what we’re trying to produce? If not, they’re no better a basis for judging student performance.

2

Robert 08.30.10 at 11:22 pm

Hmmm. Condensed, that paper means that one shouldn’t use VAM with small sample sizes over one-year periods as the sole basis for making personnel decisions. Has anyone actually made that particular claim?

This is superficially similar to the debate about whether risk-adjusted outcomes could be used to evaluate quality of medical care.

3

The Raven 08.31.10 at 12:48 am

Brett, Read The Fine Paper. It is damning. If you hominids care about educating your offspring, you will rethink the whole approach.

4

jeer9 08.31.10 at 1:53 am

Excellent paper which confirms most my suspicions about the standardized test mania which has been sweeping our nation for the past decade. At my high school, the administration has completely bought into the PLC (Professional Learning Community) approach, and the consequence is that literature, especially two months prior to the exams and at the behest of the principal, disappears and is replaced by teach to the test drill and kill practice. The result is that our numbers have gone up (though I would argue the quality of teaching has declined – don’t ask me to quantify that!), and teachers are shown lists in August ranking their success (failure?) which is demoralizing when a teacher has busted his ass on a classroom full of behavior problems. Administrators are ecstatic, however, because these scores are the sole measurement of their oversight and indicate great skill and innovation, while the pressure to increase each year’s advance rises exponentially. Short aside: one of our department’s better teacher’s sophomores scored 6 points lower on the vocabulary section than a peer and so he asked the peer what teaching tips she could provide that helped her students on the exam. Since he had taught over 200 vocabulary words, he thought the issue must be one of strategy. The peer replied that she hadn’t actually taught any vocabulary at all last year. As Harry has surmised, this report will certainly be ignored because it contradicts the current data-driven obsession. Merit pay makes so much sense because its proponents remain oblivious to all the diverse factors influencing growth. NCLB and RTTT are Classicism gone mad, clear-cutting trees with reckless abandon while Romantics crawl further and further into their imaginary forest.

5

moe 08.31.10 at 2:21 am

I don’t know any data on what motivates people. But if you wanted to utterly destroy my interests in kids and my motivation to improve their learning, I can’t imagine anything more effective in achieving that than the carrot/stick approach.

My reason for thinking this: Although I’m not a teacher, it seems like it would make kids, particularly difficult to teach kids, my adversaries in a sense. If they do not improve, then I get fired. It would be so hard to keep my eye on a student’s well-being in that context and not see them as little performers who hold the key to my future. If I suspect them of being unable to help being underperformers, there’s the risk I would start to resent them.

People always suppose they can have the beliefs and attitudes that are ethical or rational in any context but there are a lot of contexts that suck the life right out of you and this sounds like one of them.

6

mregan 08.31.10 at 2:27 am

Finding an easily calculated metric to use on short term data to reflect long-term success is a sort of philosopher’s stone for the social sciences, is it not? Long-term studies of controlled populations would be the only definitive answer, but we as a society have no patience for such, and consider them somewhat vaguely unjust in any case. So what’s a mother to do? Are four enough? Are six too many? I know! Let’s try something that seems like a good idea until someone else complains that it isn’t just like education when I was young, and then CHANGE everything before we’ve determined anything, and so spoil any data, and then blame both sides. Perfect democratic solution. Because if we win next time, we can change it all back, and no one will ever know who was right and who was righter.

7

Witt 08.31.10 at 2:33 am

This is really interesting; thanks for posting it.

A related issue that I’m starting to see a lot is schools going to great lengths to avoid enrolling students that they think are going to drag down performance measures. I’ve seen it happen both with students that administrators profile as potential truants, students who are “overage” (17-21), and students who have disabilities or limited English proficiency. As the emphasis on test scores and other data continues to grow, I think we’re likely to see more of this, and a growing group of “castouts” who are unable to access the K-12 system.

8

FDChief 08.31.10 at 2:38 am

When I was an Army sergeant part of my evaluation was a graded exercise called and ARTEP. Several months before the ARTEP I would gather my squad for a friendly talk.

“We’re about to do this graded field problem” I would explain. “We will be graded as a squad but the grade will only reflect on me. The graders will not listen to my explanation of how many of you are gimps, wheezers, chronic self-abusers, morons, gomers, mouth-breathers, learning disabled products of the union between a Marine and a gorilla, the offspring being, of course, a retarded gorilla. They will not believe that the reason we fucked up were because you oxygen thieves were unable to learn. They will blame it on my being unable to teach you.

Therefore, I will carefully explain everything we will do. I will show you how to do it. I will coach you through it. You will then do it for yourselves, with my direct supervision and correction. Finally we will do it at combat speed.

After that you have my personal assurance that any subsequent failure on your part, however small, will result in your horrible lingering death, probably involving a red-hot poker and one or more of your bodily orifices, or a transfer to a posting on the Korean DMZ, whichever you fear more.”

Not surprisingly, I never failed an ARTEP. And this is, in effect, what high-stakes testing will do for teachers and students. “A risk I would start to resent them”? Let’s think about that…

9

M. Krebs 08.31.10 at 2:53 am

I’m sympathetic, but it sure would be nice if there were a few non-Educator types involved in the report’s writing. Like, for instance, a scientist or two.

10

mcd 08.31.10 at 3:00 am

Aside from the problems of testing, the Coleman Report from the 1960s and a study, the Coleman Report, 40 Years After by two Wisconsin sociologists, suggest that teacher attributes are less important than the economic/ethnic makeup of schools,, and the students’ class backgrounds.

NCLB may well be cover for eventually shutting down public schools serving poor children. While piously proclaiming our concern for their well-being.

11

Robert 08.31.10 at 3:10 am

Moe wrote: if you wanted to utterly destroy my interests in kids and my motivation to improve their learning, I can’t imagine anything more effective in achieving that than the carrot/stick approach.

and Witt wrote: A related issue that I’m starting to see a lot is schools going to great lengths to avoid enrolling students that they think are going to drag down performance measures.

As I said above, this is similar to the debate about whether risk-adjusted outcomes could be used to evaluate quality of medical care. Versions of these arguments are made by physicians opposed to their use.

12

me 08.31.10 at 3:15 am

i teach 9th graders in a subject that’s not tested. our school got a grant last year to improve test scores. teachers in tested courses had to improve scores based on the % of their students who passed the previous year. teachers like me, in untested subjects, were assigned a randomized group of 12 at-risk kids to “mentor.” altho we weren’t math/science teachers, we would get $3000 if a certain percent of our group passed the TAKS science test. we were supposed to document when we met with the kids, what we did, etc.
i could never get the kids to come in before or after school, so i never saw them (none of them were my students, only 2 were former students of mine).
long story short, they passed, i got 3 grand for doing nothing.
thats what’s wrong with paying teachers “incentives.”

13

Witt 08.31.10 at 3:28 am

Robert, can you spell out your point a little bit more? I have no background in the issue you’re talking about, but it sounds like you’re saying the argument is, “If hospitals are forced to keep statistics on which patients survive, then they will refuse to admit patients they think are going to die.” Which seems plenty likely to me.

Am I missing the argument? Or are you just drawing a parallel between the educational and medical versions of the concern? And either way, are you suggesting it’s not a valid concern, or what?

Sorry to be dense. (And btw, I’m not really sure what role the “physicians” bit has to play in your point, but I’m not a teacher or connected with schools — I’m an advocate in a different field, and end up dealing with a lot of school-registration problems more or less accidentally.)

14

mregan 08.31.10 at 3:33 am

I am merely the husband of an inner city 9th grade public school science teacher who relates that her every classroom choice is determined by the principal’s opinion of what shall increase the school’s overall grade performance on our state’s standardized test, not a front-line educator, so my opinion is of course at that remove, but from what she will admit frustratedly at night, it would seem to me that a given year’s test scores from teacher to teacher are more a function of the luck of the draw, the assignment of bad apples, the success of the football team (and thus school–and teacher–morale), the incidence of violent crime in the school or neighborhood, the economy in general (as relates to family support of educational attainment), the individual teacher’s rate of improvement at classroom management, and other somewhat difficult of measurement metrics, rather than that of her own admittedly, sometimes downhearted efforts in the face of all of the above listed constraints.

15

Robert 08.31.10 at 6:33 am

I think both that these concerns are valid and that there is a parallel to the way these things have been done in evaluating the quality of health care. However, one of the lessons from the medical world is that *if* one is going to implement something like this then one mustn’t be naive about it. For example, you need large enough sample sizes so that you’re not whipsawed by year-to-year variability; and that any single outcome variable (like operative mortality, or the rate or re-hospitalization, or iatrogenic infections) doesn’t tell you the whole story so you shouldn’t base your entire decision on only a single outcome variable. To do this right, you have to make sure physicians and hospitals don’t turn away the sickest patients in order to improve their ratings. This means that whatever statistical model you develop has to do a good job in adjusting for the patient’s (or, in this case, the student’s) situation at the time when he or she comes under the hospital’s or physician’s care. And, you need to make sure that the physician, the hospital, or the teacher is the right agent and has the ability to improve outcomes. For example, it’s possible to analyze the post-operative infection rates associated with different surgeons but are the surgeons really the “right” focus for infections? In the educational context, are math teachers the only path through whom students learn mathematics? When I was in high school (decades ago) I learned almost as much math from my physics teacher. I probably learned more about English grammar from my Latin and French teachers than from my English teacher. These are all problems that have to be worked out.

16

Dina 08.31.10 at 8:44 am

Yeah, but I want your critique anyway, Harry. Would you be willing to share how it compares to the EPI paper?

17

Zamfir 08.31.10 at 8:55 am

Robert, how well would say these problems are adressed in te medical world? Are they theoretical issues people should take care of, or are they more or less adressedto people’s satisfaction?

18

Brett Bellmore 08.31.10 at 10:18 am

What? Raven, I did read the fine paper. I merely noted that the executive summary didn’t do it justice. And that while measuring the change in test scores of individual teachers’ students may lack statistical significance, whatever measure we DO assess teachers by must ultimately be tied back to student performance at some point.

19

conall 08.31.10 at 12:03 pm

One thing we’ve learned from lotteries for school-places is that schools don’t matter. All that nonsense about schools competing, and therby raising academic attainment is bunkum. (There may be other reasons for choosing one school over another, but better grades is not one of them).

And it will almost certainly be the case that ‘teachers don’t matter’. Unless and until we start allocating teachers at random to their classes, or allocating random groups of students to teachers we will lack the scientific basis to draw valid conclusions.

Details about this can be found in my book “Lotteries for Education” by Conall Boyle, published this month by Imprint Academic, Exeter.

20

Joshua W. Burton 08.31.10 at 12:47 pm

If new laws or policies specifically require that teachers be fired if their students’ test scores do not rise by a certain amount, then more teachers might well be terminated than is now the case. But there is not strong evidence to indicate either that the departing teachers would actually be the weakest teachers, or that the departing teachers would be replaced by more effective ones.

To get rid of the weakest teachers, a more direct approach would be to give the standardized tests to the teachers instead of to the students. A high school teacher whose reading and math scores are below national median for high school students, and who is unable to improve his own score by half a standard deviation in three years, for example, is probably not going to be effective at improving his student’s scores by that amount, no matter what else the school is doing right. In many urban school districts, this cut would eliminate a quarter of all teachers, and refilling the slots under this constraint would force teacher salaries up and nontraditional hiring barriers down, without explicitly lobbying for those long-needed reforms.

21

Joshua W. Burton 08.31.10 at 12:57 pm

Applying the medical analogy, one of the first steps any hospital takes when trying to improve patient outcomes is to make sure their own staff aren’t spreading infection. Long before risk-adjusted outcomes, comes primum non nocere.

22

someguy 08.31.10 at 3:30 pm

This might be a little harsh I cannot read any PDF right now.

The metric[s] really doesn’t matter at this point. No metric that involves elimination of or less pay for lower performing teachers will be acceptable to the Teacher’s Union. No general linkage between pay and performance is acceptable to the Teacher’s Union and thus to the Democratic party.

Ponder NY City rubber rooms being replaced with make work to get an idea of the scope and depth of issue. That is the progress of a very slow moving glacier.

http://www.nytimes.com/2010/04/16/nyregion/16rubber.html

I don’t support testifying education and I don’t think we should do something like drop the lower 10% of teachers every year based on any set of metrics.

Harry is talking about how many angels can fit on a pin head.

What metric would you use to eliminate under performing teachers or what general linakge would you use between pay and performace? No need to be very specific.

Nothing is acceptable. In the NE – You must give an annual 4% raise, 50% pension, medical for life, for 200 or less days a year, and if you don’t you are a dupe for 18th robber barons looking to exploit child labor.

23

chris 08.31.10 at 3:35 pm

No metric that involves elimination of or less pay for lower performing teachers will be acceptable to the Teacher’s Union.

If the teachers are merely being scapegoated for drivers of success or failure that lie outside of the classroom (principally class), this is a completely reasonable position for the union to take. Why should they let some of their members be unjustly blamed and their careers sacrificed so that other people can feel like they are doing something?

The union darned well should demand that the real problems be found and addressed rather than trying to pin everything on “lower performing teachers”.

24

Substance McGravitas 08.31.10 at 3:37 pm

No general linkage between pay and performance is acceptable to the Teacher’s Union

It’s a reasonable bargaining position for any union to take. It’s up to management to hire the right people and do the right paperwork to fire poor performers.

25

bianca steele 08.31.10 at 4:02 pm

I am surprised to see so little interest here (and at Yglesias’ blog too fwiw) in arguments that teachers’ unions can contribute to improving education.

26

Anonymous 08.31.10 at 4:03 pm

>There is also little or no evidence for the claim that teachers will be more motivated to improve student learning if teachers are evaluated or monetarily rewarded for student test score gains.

How can this even be remotely true? Do teachers have downward sloping utility curves?

27

bianca steele 08.31.10 at 4:21 pm

I caught only a few minutes of Emily Rooney’s show on Boston NPR yesterday, but the few minutes I heard addressed exactly this question of how to proceed w/r/t standardized testing given what we now know.

28

Bill Gardner 08.31.10 at 4:22 pm

The report is convincing and does an excellent job explaining the underlying issues. As mentioned, the problems are very similar to problems one encounters in attempting to profile physicians. (I also like what Joshua says @20.)

The report says that because of several important technical and substantive problems, school systems that rely exclusively or primarily on VAM are doing no good and probably doing substantial harm. These problems include insufficient data on individual teachers, insufficient precision in measurement instruments, lack of content validity in instruments, inability to adequately measure student background characteristics, and lack of understanding in the mechanisms that assign students to teachers. They also point to the value of structured observational assessments of teacher performance, and conclude that VAM should be used, if at all, as part of a comprehensive teacher evaluation.

Every part of the above is important. The report does not say that VAM could never work. For example, the report cites the agreement between structured observation and VAM as evidence for the validity of the former; this argument makes no sense unless there is also some validity in VAM. It’s possible that there may be a place for VAM in the (likely far) future, if these specific problems can be addressed.

29

someguy 08.31.10 at 4:24 pm

chris,

That is a very respectable position.

The problem with that and data backs it up is that if the factors lie outside the classroom we don’t need to pay teachers all that much.

Teachers want us to believe that teacher quality is crucial and that it requires high compensation while simutaneously having us believe that teacher performance is un-measurable.

30

someguy 08.31.10 at 4:36 pm

Substance McGravitas,

It is members of the Teacher’s Union that are largely responsible for hiring new teachers and is teh Uion that is completely responsible for not allowing teacher’s, except for most egregious offensives, to be dismissed.

Unless performance has no impact on outcome, I really don’t think it is reasonable for unions to refuse any attempts to measure performance and pay accordingly, and it certainly wouldn’t be ok for management to accede to those demands and in this case all of us are management.

31

Bill Gardner 08.31.10 at 4:37 pm

someguy @28: The problem with that and data backs it up is that if the factors lie outside the classroom we don’t need to pay teachers all that much.

Teachers want us to believe that teacher quality is crucial and that it requires high compensation while simutaneously having us believe that teacher performance is un-measurable.

The report did not argue that teacher performance does not matter, nor did it argue that teacher performance was not measurable. It argued against the exclusive or primary use of VAM for this purpose (see @27).

Think about your own educational experience. Didn’t you perceive important variation in the quality of quality of your teachers, and didn’t it make a difference to you?

32

Bill Gardner 08.31.10 at 4:39 pm

Sorry, the second paragraph of @30 quotes someguy, but the italics to denote this are missing.

33

ejh 08.31.10 at 4:39 pm

So, who’s going to write the amusing comment which retrospectively judges our English teachers by the standard of literacy exhibited by our comments?

34

Substance McGravitas 08.31.10 at 4:41 pm

It is members of the Teacher’s Union that are largely responsible for hiring new teachers and is teh Uion that is completely responsible for not allowing teacher’s, except for most egregious offensives, to be dismissed.

That “largely” is doing a lot of work, and yes, as advocates for their members, unions try their best to keep their members employed.

Unless performance has no impact on outcome

Read the paper, it’s interesting.

35

Steve LaBonne 08.31.10 at 4:49 pm

I am surprised to see so little interest here (and at Yglesias’ blog too fwiw) in arguments that teachers’ unions can contribute to improving education.

That’s because the usual run of “education reformers” don’t really care about education (or understand much about it), let alone about improving it; they just see an opportunity for union-busting. It’s to Ravitch’s discredit that she got mixed up with that crowd in the first place, but kudos to her for now doing everything she can to right that wrong.

Meanwhile, of course constructively engaging the unions is a critical part of getting teacher buy-in to genuine improvements. In fact, I think willingness to acknowledge that point is a very good proxy for assessing whether any given would-be reformer is worth taking seriously.

36

someguy 08.31.10 at 5:03 pm

Substance McGravitas,

I am pretty sure principals are members and they have most if not all of the say regarding who gets hired. but maybe I am wrong. Please let me know.

Whatever Unions should and shouldn’t do we as management should make an attempt to make sure performance and pay have a connection. Assuming performance impacts the outcome.

If performance has no impact and outcome and again that is respectable position we should at the very least consider whether we need to pay teachers all that much.

Teachers want us to believe that teacher quality is crucial and that it requires high compensation while simutaneously having us believe that teacher performance is un-measurable.

37

Robert 08.31.10 at 5:04 pm

Zamfir asked: Robert, how well would say these problems are addressed in the medical world? Are they theoretical issues people should take care of, or are they more or less addressed to people’s satisfaction?
I don’t think these issues are universally well-addressed — however, I would say they’re better addressed. For example, I don’t know of anyone who uses single-year risk-adjusted outcomes (the analog of educational VAM) as the sole determinant of decisions; they’re used as part of a larger decision, as perhaps they should be.

This debate has been going on for years and years in health policy. Bill Gardner might be able to flesh this out. Professional medical societies (like, those for the surgeons or cardiologists or obstetricians or anesthesiologists) have ended up collecting their own data so that they could figure out whether the outcomes people have been focusing on, and the risk-adjustment models that are being used, make sense. A good side-effect is that the act of data collection allowed the societies themselves to examine differentials in the quality of care, and to begin the discussion about how to improve quality and reduce differentials. I’d think that if VAM becomes more common, teachers’ unions might want to do the same thing.

38

Steve LaBonne 08.31.10 at 5:09 pm

I’ve never met a teacher or a teacher educator who believes there’s no effective way to assess teacher performance. This is a straw man erected by, again, people whose interest is in union-busting, not in education.

What any person who’s not an idiot DOES realize is that there is no simple, quick, magic-bullet way to assess teacher effectiveness, least of all a poorly designed multiple-choice test.

39

Substance McGravitas 08.31.10 at 5:29 pm

I am pretty sure principals are members

I certainly can’t speak to the whole of America but in my experience principals are representatives of management and therefore non-union.

40

bianca steele 08.31.10 at 5:33 pm

Steve LaBonne:
I doubt that’s the case. I assume they’ve simply taken as their own the general disdain for teachers’ unions among large parts of the public that has become general in the rightmost sections of the Democratic party, and that they’ve never gone looking for data on the actual history of the unions. IIRC you’ll also find the assertion that class size doesn’t matter, when I would guess what they have in mind is the difference between 20 and 15 students per class. I don’t think this is traceable to their not really caring or to a belief that there are no constraints on what “correctly” trained individuals can do if they really try. But it’s almost as if they live in an alternate universe in which there has never, ever been any significant support for unions in the United States, and they therefore have to build a union movement up from scratch.

41

Sebastian 08.31.10 at 5:35 pm

“Meanwhile, of course constructively engaging the unions is a critical part of getting teacher buy-in to genuine improvements.”

Hasn’t that been demonstrated to be much like constructively engaging Republicans as a crucial part of getting buy-in for genuine improvements?

42

someguy 08.31.10 at 5:36 pm

Bill Gardner,

Like I said I didn’t get a chance to read the report. I must need some kind of PDF install.

Teachers refers to teachers as a general whole not to the folks who wrote the report or all teachers.

I agree that measuring teacher peformance based mostly on test scores is probably a poor approach.

And in general measuring teacher performance is not easy and linking performance to outcomes is harder.

I am fairly sympathetic to what I believe Harry and say Matt Yglesias
want. They desperately want better schools. They want to better invest current resources and spend even more money in order to get better outcomes. They think this can be done.

They think teacher quality is crucial and that it responds to incentives. [At least Matt does and I would guess Harry does as well] Maybe or maybe not but the white elephant is this

http://www.nytimes.com/2010/04/16/nyregion/16rubber.html

whatever they think is largely immaterial.

Poor public incentive structures are a feature and not a bug of US public spending.

In the meantime all indications are that we over spend on education. That we could spend less and get the same outcomes.

43

Steve LaBonne 08.31.10 at 5:38 pm

Hasn’t that been demonstrated to be much like constructively engaging Republicans as a crucial part of getting buy-in for genuine improvements?

No. This has been another edition of simple answers to simple questions.

44

someguy 08.31.10 at 5:40 pm

Steve LaBonne,

“I’ve never met a teacher or a teacher educator who believes there’s no effective way to assess teacher performance. This is a straw man erected by, again, people whose interest is in union-busting, not in education.

What any person who’s not an idiot DOES realize is that there is no simple, quick, magic-bullet way to assess teacher effectiveness, least of all a poorly designed multiple-choice test.”

It is a straw man and yet the very vast majority of public teachers are not paid based on relative performance let alone student outcomes.

45

Substance McGravitas 08.31.10 at 5:42 pm

yet the very vast majority of public teachers are not paid based on relative performance let alone student outcomes.

The vast majority of people are not paid based on relative performance let alone outcomes.

46

bianca steele 08.31.10 at 5:43 pm

someguy,
If the principal is motivated to raise test scores, she certainly has carrots and sticks to motivate her teachers even if she does not have the direct power to raise or lower their pay.

47

Steve LaBonne 08.31.10 at 5:44 pm

It is a straw man and yet the very vast majority of public teachers are not paid based on relative performance let alone student outcomes.

Indeed, both of those statements are true. Is that what you meant to say?

Do you think teacher pay should be based on ineffective assessments that only pretend to measure teacher effectiveness?

48

Bill Gardner 08.31.10 at 5:55 pm

Robert @36:

Agreed, but I would qualify what you say in that doctors are, unfortunately, sometimes profiled on insufficient samples of data and inadequate measures of outcomes.

Using data to evaluate and improve complex services is still a black art. Many people think that Brent James, MD, and InterMountain Health Care know as much about it as anyone.

49

Robert 08.31.10 at 6:28 pm

Bill wrote: Using data to evaluate and improve complex services is still a black art.
Yeah, but the data evaluation part is probably less black art than using the data to improve services part. I’d say James’ particular skill is in the latter.

As I said, much of the EBM push was enabled by having the evidence available. On the one hand, the educational system would do well to start collecting the data they need for similar kinds of quality improvement efforts. On the other, the political environment facing teachers is different enough from that facing physicians that the analogy isn’t exact. You can see some of it in this thread.

On a completely different tangent, I’m sorry you gave up your blog.

50

Alex 08.31.10 at 6:43 pm

Imposing performance-related pay without identifying what it is the good teachers are doing right and vice versa is not only futile, it’s evil – it’s like taking 10% of the rations from the bottom 10% and giving them to the top 10%. You’ll kill the bottom 10% all right…but there’s always going to be a bottom 10%. Also, it’s profoundly un-, indeed anti-scientific – correlation without any attempt to identify mechanism of action, let alone to demonstrate causation. Cargo cult stuff.

Further, there’s a Goodhart effect. In the absence of information about how to improve, what gives you the best payoff – trying to do better, or cheating?* It’s the Enron fallacy – create some right-wing prick’s idea of a fitness-selecting environment and you end up selecting for sharks if you’re lucky enough to get multicellular complex life, smallpox viruses if you’re not.

*Further detail: if you were constrained from cheating by social norms before, you may not be if given enough coercion.

51

Bill Gardner 08.31.10 at 6:50 pm

Robert @49: the data evaluation part is probably less black art than using the data to improve services part. I’d say James’ particular skill is in the latter.

Right on both counts, and I was specifically thinking of James in the latter context.

And the blog has risen from the dead! Thanks for the kind words.

52

Dr. Hilarius 08.31.10 at 7:24 pm

Joshua@20 is on to something important. Many teachers lack a good foundation in their own subject matter. Part of the problem is the whole “School of Education” model in which would-be teachers are channeled into “subject matter for teachers” and away from regular math, biology, and history classes. Education students are then subjected to the latest fads in teaching methods. (dear god, how many decades of debate over phonics vs. whole word reading must we endure?)

When I still strolled the halls of university I met many students who had contemplated a teaching career, only to abandon it once they checked out the curriculum for certification. In teaching biology at two universities, teachers were some of my least prepared students.

53

Robert 08.31.10 at 7:28 pm

Bill wrote: “And the blog has risen from the dead!”
Excellent. I’m sure I’ll enjoy following it.

Back to topic: while I think it’s important to understand that the analogy isn’t exact, there are still lots of parallels between “pay for performance” in health care and education. If the educational world is going to move toward VAM they could certainly learn from what went on (and is still going on) in the health care world.

54

Steve LaBonne 08.31.10 at 7:31 pm

Many teachers lack a good foundation in their own subject matter.

This is something I agree with, though it’s primarily a problem at the elementary level (but has disastrous effects on elementary math and science education). I observed this myself with several of my daughter’s teachers. On the other hand, she had some outstanding high-school teachers (moreseo in science than math, though.)

But it’s important to note- and not always realized by would-be reformers- that having only subject-matter training is not adequate preparation for the classroom. (It isn’t for college teachers either, as I can testify from my own experience.) Ed schools need to be improved, not abolished.

55

chris 08.31.10 at 7:32 pm

@Alex 50: Furthermore, even if you could confirm that there was something some teachers are doing wrong that is limiting their performance, the better way to deal with it is not to fire them, but to _retrain_ them. Perhaps seminars on teaching techniques taught by the top 10% of teachers and attended by the bottom 30-40% would have some value, provided the actual measurements were valid and reflected something meaningful. The top teachers could receive a bonus for teaching such a seminar (which, of course, they have to score at the top to be eligible to do, so this program would also double as incentive pay), if any funding for carrots could be pried out of the stick-clutching fists of the “reformers” — for my money, the fact that “reformers” speak first, last, and frequently only in terms of punitive measures reveals their actual motivation to be hostility towards teachers and teachers’ unions.

P.S. One obvious problem with year-over-year measures is that a teacher would benefit statistically from having students that had an unusually bad (for them) year last year, whether that was due to a bad teacher or some external stress such as family problems, as those students will tend to regress (upward) to their personal performance mean, and the opposite goes for students who are depressed or have family trouble *this* year. In theory, a large, careful, longitudinal study might be able to correct for that; in practice, anyone looking for a quick statistic they can brandish to decide which teachers to smite isn’t going to stop long enough to even look for that kind of effect, and given any reasonable class size, only a few such cases is going to make that teacher a temporary outlier.

56

Sebastian 08.31.10 at 7:38 pm

“No. This has been another edition of simple answers to simple questions.”

Then why has it taken more than 30 years of resistance from the teacher’s union to even get to the point where we might be willing to consider *some* method of evaluation? They’ve been resisting the very idea for decades. The general strategy of the teacher’s unions in the United States has been to say “no, no, no” to any possible proposal for evaluating teachers and god-forbid firing bad ones. Just like Republicans in Congress there is no useful proposal, merely a never-ending series of “we don’t like it” and “that’s unamerican”. The ATF had a short period to the contrary, but the NEA hasn’t even gone through that.

57

Steve LaBonne 08.31.10 at 7:39 pm

They’ve been resisting the very idea for decades.

And the reason why is clearly illustrated by some of the simple-minded “contributions” to this thread. I’d think less of them if they didn’t resist when that’s what’s on offer.

58

chris 08.31.10 at 8:13 pm

any possible proposal for evaluating teachers and god-forbid firing bad ones

A perfect example of my point @55: the *first* reason given for evaluating teachers is so that we can decide which ones to fire. The idea of improving their skills (which are, like most skills, learned, therefore can be relearned) doesn’t even cross his mind.

P.S. Of course once we start firing teachers, better teachers will fly out of our butts to take their places. There couldn’t possibly be any downside to destaffing an already understaffed educational system.

59

Harry 08.31.10 at 8:27 pm

Rule of thumb: when somebody says we need to fire teachers, they have never fired one, and have no intention of doing so. People who actually want to fire teachers do not talk about it.

60

Steve LaBonne 08.31.10 at 8:27 pm

There couldn’t possibly be any downside to destaffing an already understaffed educational system.

Not to mention demoralizing an already demoralized educational system. I’m sure the “god-forbid” types would respond really well to that kind of “motivation” in their own workplaces. (Assuming they’re not bosses who are inflicting it on their own employees and then bellyaching about how you can’t get good help these days).

61

Alex 08.31.10 at 9:53 pm

Also @55: quite a lot of universities already have the practice of having teachers sit in on each other’s classes.

It’s also very possible, perhaps likely, that the best teachers may not be aware intellectually of why they are doing well, in which case there’s no point in asking them to teach the others.

There is an excellent reason to get value-added numbers sorted by teacher, but it’s not to demonise any particular group of teachers. It’s to get some actual information about the trade, teaching. But as long as the aim is to bully somebody, it’s entirely right to treat the whole thing as more rightwing bullshit.

It is strange that people who affect to love the ideas of game theory, deterrence, and rational signalling don’t seem to apply them to their own politics. If you want the teachers to agree to this stuff, you need to make a costly signal in that direction – like *not being such a cunt about it*.

62

piglet 08.31.10 at 9:55 pm

The education debate in the US is bizarrely fascinating. I don’t think there is another profession that is as frequently maligned in this country. Wall Street bankers aren’t even close to getting the amount of blame that is routinely dumped on teachers. The myth of performance pay should have been irreversibly crushed by the spectacle of failing finance executives paying themselves bonuses in the face of the global financial crisis. And still people rant about teachers being overpaid.

Like any profession, teacher quality is likely to describe a bell-shaped distribution. Like in any profession there will be some really bad individuals and some really good ones. those probably don’t matter much for the overall outcome, it is the average that counts. That handful of teachers that they try to fire in New York and that have been condemned to these bizarre rubber rooms until their fate is decided, they don’t matter for the grand scheme of things. Likewise, and it is amazing how people get this wrong, the few extraordinarily brilliant teachers are not making much of a difference because teaching doesn’t scale. Any one teacher can only teach so many students. The education system depends on the many who are average. All the magic solutions completely ignore that fact.

63

Sebastian 08.31.10 at 10:10 pm

“The idea of improving their skills (which are, like most skills, learned, therefore can be relearned) doesn’t even cross his mind.”

The training and retraining has been done for years. There are already all sorts of incentives in place–encouraged by the union–to get your second and third Master’s degree and this or that and the other certificate. What you are suggesting is already part of the game. To the extent that you believe there is a problem, that isn’t the solution. (At least so long as the union resists evaluation–because we can’t know which ones work and which ones don’t without evaluation.)

“The myth of performance pay should have been irreversibly crushed by the spectacle of failing finance executives paying themselves bonuses in the face of the global financial crisis.”

I don’t think that myth teaches what you think it does. The problem of failing finance executives and performance pay is that there was little downside for poor performance. Which is exactly what people here seem to think is a good isea in this case.

64

Alex 08.31.10 at 10:10 pm

And the only way for the average to improve is…learning.

65

Alex 08.31.10 at 10:12 pm

At least so long as the union resists evaluation—because we can’t know which ones work and which ones don’t without evaluation.

They will continue to resist evaluation until you give them some reason to think your motives are honest.

66

Sebastian 08.31.10 at 10:44 pm

What is an honest motive? This is exactly why I compared the teacher’s union to the Republican Party. They have a dishonest interpretation of what striving for good policy would look like which causes constructively engaging them to just let them cause endless delays. So long as “evaluation must never cause teacher firings” is their mantra, they aren’t engaging in good faith.

67

Substance McGravitas 08.31.10 at 10:46 pm

So long as “evaluation must never cause teacher firings” is their mantra

What evaluation are you referring to?

68

dsquared 08.31.10 at 10:52 pm

I have no idea, no idea at all, why the question of performance-related pay for teachers (which is a not awful idea, kindasorta looks like it worked when tried in Northern Ireland, clearly not a panacea but probably a sensible idea in moderation) has got caught up with the question of whether or not it’s possible to create an econometric model for the problem of extracting a pure and coherent signal of teacher quality from the noisy indicator of grades (my assessment – maybe, but why would you want to do that?).

Were my school days completely unusual? I had some teachers who were naturally gifted, some who were competent and put in a lot of effort, some who were competent and put in a reasonable amount of effort, some who were lazy, some sadists and one alcoholic. Furthermore, everyone knew which teachers fell into which categories. The purpose of a PRP system is simply to try and move some teachers from categories three and four into category two, plus possibly to try and retain the ones in category one from going off to more profitable employment. It’s never going to revolutionise anything, but paying people for putting the work in is pretty well proven as a technology. All that’s needed is to give the headmaster a budget for bonuses, and tell him to hand it out. If you don’t trust the headmaster to do that, well that’s your problem right there and you probably need to solve it before you do anything else at all.

The really extraordinary thing is that this is actually how all sorts of non-monetary rewards are actually handed out in the education industry – promotions in particular. Nobody (I hope) would start to claim that promotions, headships etc ought to be awarded purely on econometric grounds, but nobody also would suggest that they should be given on the basis of strict seniority or something. What I never understand is why there is so much resistance to handing out a bit of cash in the same way in which head-of-department status (which amounts to a bit of cash, plus a title, plus a rebalancing of admin vs classroom) is actually awarded. The studies on the effects of the Northern Irish experience, IIRC, attributed nearly all of the positive effects of performance related pay there to the fact that it created a way to allow talented teachers who had reached the top of their scale to continue to increase their earnings without taking promotions out of the classroom.

69

Alex 08.31.10 at 10:58 pm

So long as “evaluation must cause teacher firings and that is its purpose” is your mantra, you aren’t engaging in good faith.

70

piglet 08.31.10 at 11:21 pm

An observation: performance-based pay is not really used much in most of the private sector. There are some areas, like sales, where there may be a direct feedback between some measured performance metric and pay but it is not I think the case in much of the private sector. During my years in the IT industry, I didn’t encounter performance metrics except in the most rudimentary sense (namely, the yearly evaluation by the supervisor mostly based on subjective judgments and team politics). Never did I get paid for the amount of code written or the number of bugs registered (and these are metrics that are easier to quantify than teaching success, although code quality is not). Bonuses are usually based on some combination of company success (profit) and team success (usually vaguely defined except in sales, and often based on politics). The often heard claim that pay is normally based on performance, “everywhere except for teachers”, is just bogus. Least of all is it true in finance, as even Sebastian concedes.

A bonus system in education would be either largely symbolic (as it is often in the private sector), or it risks being counterproductive by creating dissatisfaction, a feeling of being treated unfairly, jealousy etc.

71

piglet 08.31.10 at 11:31 pm

To be more precise, many nominal performance pay or bonus schemes in the private sector are not, or only weakly, based on the actual individual contribution. (CEO pay based on share price is an obvious example – the CEO has actually little influence). In most cases the individual contribution is difficult to measure and attribute.

72

bianca steele 09.01.10 at 12:11 am

piglet,
I don’t think pay and benefits issues in the software services or engineering/manufacturing world are much like those in schools, basing this on my awareness of what went on in my dad’s school and gossip among high school students. There’s nothing in that environment like teaching, and when people are doing real work there isn’t time for bs to the same degree.

Alex,
I don’t see that bullying has anything to do with it. It might be bullying if there were no feasible or infeasible way the teachers involved could even try to meet the demands, but AFAICT there is a well accepted and easily monitored method for getting test scores up–teach to the test–with little concern about what cannot be tested with bubble sheets.

73

Steve LaBonne 09.01.10 at 12:14 am

People in the “engineering and manufacturing world” are doing “real work” all the time whereas teachers aren’t? One word: Dilbert.

Most people who talk like this wouldn’t last a week in a typical high school (or worse yet, middle school) classroom.

74

Robert 09.01.10 at 12:26 am

piglet observed: “An observation: performance-based pay is not really used much in most of the private sector.

Well, there’s always steak knives.

75

hix 09.01.10 at 12:58 am

“Furthermore, everyone knew which teachers fell into which categories”

Definitly not the case during my schooltime.

76

Witt 09.01.10 at 1:53 am

Furthermore, everyone knew which teachers fell into which categories

I strongly suspect that this becomes less and less true as you move to large, urban districts with heavy turnover and long supervisory distance between the teachers and those making decisions about hiring or assigning them. (Most of the schools I’ve dealt with, the principal has no power to choose his/her teachers — it’s all done at the central office.)

It also bears consideration that the U.S. is a very mobile society, and a lot of districts see large batches of new students throughout the school year, including both native-born families moving around and new immigrants arriving. So the student/parent side of “everybody knows” might be true of some subset of plugged-in students and parents, but definitely not of the whole universe, especially in terms of newcomers, emancipated teens/those coming out of foster care, people with limited English, etc.

77

Harry 09.01.10 at 2:13 am

Just to be clear, I agree with everything dsquared says (above, not everything he’s ever said). And, indeed, not trusting the principal to i) have any idea which teacher falls into which category and ii) to award the money according to those cattegories even if he/she did know is a good strategy in almost every american high school I have had any experience of. And, indeed, since their training teaches them nothing about teaching and learning, and the selection process and culture of administration militate against knowing anything about teaching and learning, and 1/3rd of all principals in the US are former gym teachers or athletic coaches (when I told my dad that he started giggling), it makes sense why unions would resist the strategy that dsquared suggests. Perfectly sensible in UK schools. And, as dsquared implies, the problem is the quality of management, not the idea of having prp that openly relies on human judgments rather than on fake econometric precision.

78

Billikin 09.01.10 at 4:06 am

me: “teachers in tested courses had to improve scores based on the % of their students who passed the previous year.”

This sounds like what I have heard, that the test scores compare apples and oranges. The proper scores to compare are before and after scores for the same students, not the after scores for two different groups of students.

79

Sebastian 09.01.10 at 5:44 am

And to be completely clear, unions in the US are not even remotely interested in anything like what dsquared is talking about.

80

studentee 09.01.10 at 5:53 am

“A bonus system in education would be either largely symbolic (as it is often in the private sector), or it risks being counterproductive by creating dissatisfaction, a feeling of being treated unfairly, jealousy etc.”

why should stewards of public funds be concerned about potential feelings of jealousy? obviously, it would be nice were it to be avoided, but once again *incentives* — there should be some

81

Alex 09.01.10 at 7:58 am

AFAICT there is a well accepted and easily monitored method for getting test scores up—teach to the test

Which immediately renders the whole project of evaluation futile, as it then becomes a question of who is cheating more effectively.

What is the opposition to even trying to identify good (or bad) teaching, as opposed to good or bad teachers?

82

dsquared 09.01.10 at 8:20 am

and 1/3rd of all principals in the US are former gym teachers or athletic coaches (when I told my dad that he started giggling),

good lord. Is this a recent development?

83

Alex 09.01.10 at 8:22 am

It makes sense; they have fewer outside options, so they have to pour their energies into that particular greasy pole. Right?

84

Lulu 09.01.10 at 8:22 am

I feel that in a sense, this is the teacher version of college applications. I’m a Senior in high school and well, I have always asked myself, “how important are my SAT and ACT scores?” I always wonder how of a role that plays in my applying for college. My test scores are definitely not representative of me as a person or a student.

I feel that there is no cause and effect relationship between teacher and test scores. They may be correlated, but there are so many factors that we can’ t account for on the students’ part. I don’t know which states are considering for test scores to be 50% of teacher evaluation, but I don’t want to be there. Having students with low test scores doesn’t necessarily make the teachers bad. A lot of teacher valuable things that are not test by standardized tests. In English class, we write essays and analyze books that we never get tested on, ever.

Many of my teachers say that they don’t like to teach to based on a test (i.e.: AP tests) and I don’t agree with that. I want them to teach based on the test, but also combine it with other materials. If we didn’t have to take the AP tests, then I would say screw teaching what the Collegeboard wants. I would only want them to teach based on the AP b/c I need good scores for college. If college wasn’t involved, I would love to read more books and spend more time having discussion instead of going over AP test formats. In addtion, the AP tests do ask a lot of stupid questions that are so irrelevant or unimportant in that subject (i.e.: what type of art is on a soup can? that was a questions in the APUSH test).

I diverged somewhat, but considering making 50% of teacher evaluation based on test scores is ridiculous. The quality of teaching cannot be reduced to dead statistics that do not paint an accurate picture.

85

dsquared 09.01.10 at 8:37 am

I suppose so. Given that, my new policy on teacher evaluations in America is “do what you like, you’re fucked anyway”.

86

Alex 09.01.10 at 8:47 am

It does explain an awful lot.

87

Bill Gardner 09.01.10 at 9:52 am

Harry @77: 1/3rd of all principals in the US are former gym teachers or athletic coaches

(Gasps.)
One possible escape from the implications of this fact. In the schools I attended, most of the sports were coached by teachers of academic subjects, not gym, as an additional job. Could that account for this?

88

Steve LaBonne 09.01.10 at 10:31 am

And to be completely clear, unions in the US are not even remotely interested in anything like what dsquared is talking about.

And apparently neither is dsquared now that he’s heard something about the competence of the people who’d be handing out the carrots and sticks.

Even by ed school standards “educational leadership” programs (which provide the courses needed to qualify as a principal) are often regarded as a joke. This is a problem. It urgently needs attention.

But the Sebastians of the world are not interested in education, only in union busting.

89

Brett Bellmore 09.01.10 at 10:39 am

“AFAICT there is a well accepted and easily monitored method for getting test scores up—teach to the test

Which immediately renders the whole project of evaluation futile, as it then becomes a question of who is cheating more effectively.”

I don’t know, it would at least identify who was incapable of even teaching the test… And, unless you’re just memorizing the sequence of letters on a multiple choice test, you’re going to learn something if successfully taught the test.

90

Steve LaBonne 09.01.10 at 10:43 am

And, unless you’re just memorizing the sequence of letters on a multiple choice test, you’re going to learn something if successfully taught the test.

“Learning” that’s little more than memorizing factoids can actually be cognitively detrimental, so the “something” can be worse than useless.

But of course, finding teachers to punish takes precedence over improving education, if the latter is even really a goal at all.

91

engels 09.01.10 at 11:34 am

‘All that is needed is to give the headmaster a budget for bonuses’

Obviously it makes sense take the world of finance as our model but why stop halfway? We should go the whole hog and award successful teachers champagne, coke and strippers. People respond to incentives, don’t they? [/modest proposal]

92

alex 09.01.10 at 11:48 am

I’d say that was a very immodest proposal, and I’m all for it.

93

Substance McGravitas 09.01.10 at 1:53 pm

I don’t know, it would at least identify who was incapable of even teaching the test…

Would it? There is a linked paper in the post at the top that suggests otherwise.

94

joe koss 09.01.10 at 1:53 pm

caught this, thought it was quite good:

http://download.publicradio.org/podcast/americanradioworks/2010/testing_teachers_radio.mp3?_kip_ipx=1552865659-1283347800

Lots of good points. The Benwood schools model sounds particularly effective, and exactly what (good) educators talk about what is severely lacking and wanting.

The LA Times has a page dedicated to their on-going series on value-added analysis:
http://www.latimes.com/news/local/teachers-investigation/

Good stuff!

re: harry’s rule of thumb — except Michelle Rhea…

95

piglet 09.01.10 at 1:56 pm

studentee: “why should stewards of public funds be concerned about potential feelings of jealousy?”

Every supervisor should be concerned about “dissatisfaction, feelings of being treated unfairly, and jealousy” among his/her subordinates. Also, teachers at a school should be working as a team and this requires fostering an atmosphere of collaboration, not antagonism. Again, in my own private sector experience, supervisors ARE concerned about these things. It is amazing that common sense observations like this are disregarded by some commenters when it comes to public education.

96

Steve LaBonne 09.01.10 at 2:00 pm

It is amazing that common sense observations like this are disregarded by some commenters when it comes to public education.

It makes sense when you realize that these bastards don’t think of public employees as human beings.

But the criminals who destroyed the economy? No bonus (funded by taxpayer-financed bailouts) is too big for them. This is class war, pure and simple. Nothing to do with education at all- wingnuts don’t WANT a genuinely educated populace.

97

piglet 09.01.10 at 2:15 pm

Another observation. The issue of teaching to the test and has been mentioned. The format of standardized testing is rarely questioned since there seems to be no practicable alternative to multiple choice testing. Well take this. In my whole 19 years of education in Germany, I have never ever taken a single multiple choice test. Zero. Every test, every exam was handcrafted and hand graded by a teacher who had an educational purpose in mind. Graduation exams like the Abitur, on which university admission is based, are statewide exams administered uniformly on a single day. Again there is no multiple choice. This system is a lot less flexible and more cumbersome than administering a computerized test but it is time tested and works. Importantly, in my view, the student passing such a test has a different feeling of accomplishment. It is a coherent piece of work, not a sequence of disconnected tasks.

Now having some (very little) experience on the teaching side, I am aware that it takes an awful lot more time to do it that way. Under time pressure, multiple choice often seems the only choice. So I wonder whether the exam culture tells us something important about what is valued in each educational system?

98

Sebastian 09.01.10 at 3:43 pm

Your class warfare game doesn’t actually advance your argument. If you truly believe that there is a travesty in finance bonuses (and I’m inclined to agree with you) it does not follow that bonuses should not be given to teachers. I’m relatively sure that we can agree that the problem was that the finance people were rewarded for inappropriate actions which didn’t take into account the risks they were taking. There was a fundamental disconnect between what they were doing and what they were being given bonuses on.

Your apparent argument is that since finance people fucked up and got bonuses, education people should be allowed to fuck up too. If you believe that education tends to be very important, and I do, the fact that high finance crazies got away with all sorts of things isn’t a very good argument to suggest that we should just let US education continue to go down the tubes. Class warfare where we slit the throat of our education system to satisfy your sense of justice about how high-flying financiers got treated strikes me as fun rhetoric, but ultimately ridiculous.

The proper question from the leftist perspective would tend to be “is this institutional structure helping or hindering education”. I would say that institutional structure with teacher’s unions denying that valid evaluation of teachers is even possible, is not helping education of the next generation.

And we can probably learn from other countries here. In general, do other countries tend to believe that evaluation of teachers is not plausible and doable? My sense is no, they think it is possible and they do it. Do other countries use how well students do as a large part of that evaluation? My sense is yes, they tend to.

I also question the reflexive dismissal of teaching to the test. For certain areas like history I can sort of understand the objection–that it will cause teachers to focus only on the historical ideas likely to be tested. But for other areas, teaching to the test IS learning. If I teach to the geometry and trigonometry tests and the kids learn to answer questions correctly about determining areas, angles and volumes, then teaching to the test and learning about geometry and trigonometry have both taken place. If I teach to the arithmetic test, and students demonstrate the ability to add, subtract, multiply and divide successfully teaching to the test and learning about arithmetic have both happened. If can teach to the reading comprehension test, the children will have learned how to look through a paragraph and understand it.

The general complaint about standardized tests is that they aren’t as good as people think at making the fine grade distinctions that they are sometimes used for i.e–on a scale of 1-100 they don’t ensure that a person who scores 97 is really better than the person who scores 96. So in a reading comprehension test, sometimes one of the students will just happen to know something by luck which allows a 97 while the better student doesn’t and gets a 96. But in determining general competency, that criticism doesn’t apply. Standardized tests really are quite good at telling you the difference between a 95, a 75, and a 25. And if we use these tests to determine general competency, we don’t even need to worry about the individual student who scores one point below ‘passing’ by being unlucky. We instead worry about the teacher who year after year has a fair number of students who scored passing grades upon entering their classes and has more who score lower than passing at the end of the year.

Now there is obviously a lot of play in ‘fair number of students’, ‘score lower’, etc. But we aren’t even allowed to talk about important things like that because we are still at the “evaluation of teachers isn’t possible” phase with most of the teacher unions.

And as far as cheating on the tests goes, a huge part of that could be remedied by not letting teachers proctor the exams in their own class–and randomly assigning whose class they will proctor on exam day. In theory you might get the occasional school where every single teacher will go along with a conspiracy to cheat, but that kind of school clearly has problems anyway.

99

Steve LaBonne 09.01.10 at 3:47 pm

Your class warfare game doesn’t actually advance your argument.

I don’t waste time arguing with wingnuts. They never argue in good faith. I just tell it like it is. If you think dumping on the likes of teachers and (now thanks to Simpson) veterans, while fellating banksters, isn’t class war, you’re either stupid, dishonest, or both.

100

roac 09.01.10 at 3:51 pm

Parenthetically, it has always seemed self-evident to me that the vision driving the campaign against public education and teachers’ unions is: Private schools competing on price = downward pressure on teacher salaries = lower taxes = more money for people with money. Anybody not see it that way?

101

Steve LaBonne 09.01.10 at 3:54 pm

roac- That, and the agenda of defunding public education has the added benefit of being a shiny bauble for distracting the religious-right cannon fodder.

102

bianca steele 09.01.10 at 4:10 pm

Oh, be fair, Steve. Isn’t it up to the working class to arrange the kind of education for their next generation that they want? Shouldn’t we be thinking about them kind of like the way we think about Iraqis? It’s not exactly accurate to call it class war.

103

Steve LaBonne 09.01.10 at 4:11 pm

I’m sure the working class is preparing to greet the Koch brothers with flowers even as we speak.

104

yeliabmit 09.01.10 at 4:11 pm

Apparently the editors of The Onion have turned their minds to the question of student performance recently: [NSFW video link]

105

Alex 09.01.10 at 4:22 pm

106

Steve LaBonne 09.01.10 at 4:29 pm

For those too dense to figure out this game (or who don’t want to admit it for ideological reasons), the teabagging Republican Senate nominee in Nevada will be happy to lay it out for you as plain as day.

107

Salient 09.01.10 at 4:33 pm

I’d like to see how we’re going to measure “inspires their students to investigate stuff and enjoy analyzing it” in a quantitative way.^1^

College entrance type concerns aside, I would rather my kid have an 6th-grade teacher who succeeded at this but failed to teach the content knowledge test material terribly well, than have a 6th-grade teacher who cruised everyone through to content-knowledge mastery at the expense of the students hating it.

Folks who enjoy reading books and tinkering with stuff can recover from gaps in their knowledge, and are amenable to actually doing so. Folks who were competently taught 10,000 mnemonic tricks to memorize the geometry content for their 11th-grade mathematics test have since forgotten most of it, and when they must recover that knowledge are liable to think, oh god not this shit again.

On some level I’m not even worried about developing rudimentary low-level understanding versus high-level conceptual understanding; I’m worried about teachers providing students with the ability to better enjoy investigating stuff and getting the hang of it.

^1^This (mostly) isn’t snark. We could hire someone knowledgeable in T&L into an authority position at the district level to interview students and sit in on teachers as they teach and so on, and offer assessments based on students’ reported experiences. Call ‘em a student ombud. I think it would be quite fun to ask struggling students, “how can we help you enjoy this? what would make it more fun for you, or at least more tolerable?” And form a plan based on reasonable responses. Notice response trends in a particular class. Etc. The answers should be qualitatively valuable when attempting to assess how well a teacher helps students understand how to enjoy learning. If we start that kind of interview interaction at a young enough age, it should be possible to avoid the snark you’d get from teenagers. And I bet a full-time ombud could interview every student at least once/year, and struggling students on a more regular basis; they could interact with teachers as a liaison, a guide, and an assessor.

108

Steve LaBonne 09.01.10 at 4:40 pm

Salient makes the crucial points. THAT is what education is about. Figure out a genuine way to promote and reward that, and I’m on board.

Here is a book I’m fond of that takes a similar view.

109

bianca steele 09.01.10 at 5:14 pm

my @102: In fact, a large number of teachers in the Baby Boom generation and older themselves came from the working class, especially in big cities. They pulled themselves up by their bootstraps, with the help of the idealists in the universities and teachers’ colleges, and (like Ravitch, I’d guess) really don’t understand why the kids in their classes are having a tougher time than they did.

110

Isocrates 09.01.10 at 5:18 pm

Echoing Salient’s sentiments @ 107. This book is my favorite handbook on the subject. While it’s geared toward HS / college-level students and so comes much later in the student’s career, I think in principle it’s aimed at the same goal.

111

soru 09.01.10 at 6:09 pm

During my years in the IT industry, I didn’t encounter performance metrics except in the most rudimentary sense

It does occasionally get tried, generally with mildly hilarious results. The thing about being in some ways a simpler industry, with more measurable outcomes, and perhaps less emotion and politics tied up in it is that some of the more obvious bloody stupid ideas do tend to weeded out quicker.

The effect of reviews on morale is lopsided: while negative reviews hurt morale a lot, positive reviews have no effect on morale or productivity. The people who get them are already working productively. For them, a positive review makes them feel like they are doing good work in order to get the positive review… as if they were Pavlovian dogs working for a treat, instead of professionals who actually care about the quality of the work that they do.

at least two dozen studies over the last three decades have conclusively shown that people who expect to receive a reward for completing a task or for doing that task successfully simply do not perform as well as those who expect no reward at all.

http://www.joelonsoftware.com/articles/fog0000000070.html

112

Substance McGravitas 09.01.10 at 6:28 pm

113

piglet 09.01.10 at 7:14 pm

Sebastian:

“Class warfare where we slit the throat of our education system to satisfy your sense of justice about how high-flying financiers got treated strikes me as fun rhetoric, but ultimately ridiculous.”

Hm, slit the throat of our education system? That would be Sharon Angle (thanks for the link Steve 106). Me, I am skeptical of imposing performance-based pay schemes on public education, especially given that schemes like that have spectacularly failed in other industries.

And we can probably learn from other countries here. In general, do other countries tend to believe that evaluation of teachers is not plausible and doable? My sense is no, they think it is possible and they do it. Do other countries use how well students do as a large part of that evaluation? My sense is yes, they tend to.

This is something that I am really curious about – can the US learn from other countries? I am not aware of any other country where the education debate regularly degenerates into teacher bashing. “My sense is”, this is a genuine American phenomenon and this should give us pause. Do other countries try to measure teacher performance ? I don’t know of any example (other than old-fashioned evaluation by peers or supervisors sitting in), do you? Your sense, Sebastian, can that be turned into something a bit more specific? Do you have actual examples that you could share?

114

Steve LaBonne 09.01.10 at 7:29 pm

And not just Sharon Angle. “…the best thing to do for the economy is to let money stay with taxpayers and allow them to consume education as they would anything else: according to their individual priorities and abilities, which they know better than anyone else.” That, folks, is the Koch-funded right-wing agenda for education. That sound like “reform” to you?

115

piglet 09.01.10 at 7:36 pm

“If I teach to the geometry and trigonometry tests and the kids learn to answer questions correctly about determining areas, angles and volumes, then teaching to the test and learning about geometry and trigonometry have both taken place.”

The obvious flaw in your argument is that being able to answer multiple choice questions geometry is not the same as being able to apply the geometric concepts. You learn geometry by doing it, not by learning a catalog of test questions about geometry.

116

piglet 09.01.10 at 7:37 pm

“being able to answer multiple choice questions ABOUT geometry “

117

Matt 09.01.10 at 8:14 pm

The format of standardized testing is rarely questioned since there seems to be no practicable alternative to multiple choice testing.

Some schools in the US have used formats other than multiple choice, though it’s more expensive, I’m sure. Several years ago I had a summer job grading the standardized exams in history and social studies for (I think) 5th and 8th grade students from Massachusetts. These were all short essay questions, and hand-written. They were scanned into a computer and then fed to people like me, sitting in a room of about 50 people, who graded the exams on a 1-5 scale against a rubric for 8 hours a day for 10 weeks or so, for $10 an hour or the like. It was mind-numbing, sometimes very funny, and required huge amounts of coffee. Whether it was more effective in testing people than a multiple choice exam would have been I don’t know, but it was a different format, anyway.

118

Salient 09.02.10 at 12:38 am

“If I teach to the geometry and trigonometry tests and the kids learn to answer questions correctly about determining areas, angles and volumes, then teaching to the test and learning about geometry and trigonometry have both taken place.”

a^2^ + b^2^ = c^2^

% of students who remember this equation, 10 years later: 99%
% of students who have any idea what kind of relationship it’s describing, 10 years later: maybe 4%
(not actual statistics but I would be surprised to be far off…)

Also: SOHCAHTOA SOHCAHTOA SOHCAHTOA. I have never met a returning student who could not say this to me. Some even remembered sine, cosine, tangent. A few even remember something about opposite, adjacent, hypotenuse sides of right triangles! (Those few tend to forget which side is the “opposite” side, though.)

It is literally possible (in the sense of ‘it actually happens’) for kids to leave 10th grade geometry with less understanding of geometry than they had going in, even if their standardized test scores for ‘geometry’ improve.

…I miss the Preview feature, liar though it was at times. Those ^2^s have me nervous.

119

Davis X. Machina 09.02.10 at 2:19 am

“Private schools competing on price = downward pressure on teacher salaries = lower taxes = more money for people with money. Anybody not see it that way?”

Two words: Dotheboys Hall.

120

chrismealy 09.02.10 at 2:57 am

Principals in America are generally terrible. The Wilson Quarterly had an entertaining article (“Our Uneducated Educators”) on that subject a while back:

The academic and intellectual aimlessness of our schools is a direct outgrowth of their leaders’ impoverished academic backgrounds. About one-third of the principals surveyed by the National Association of Secondary School Principals (NASSP) in 1987 held undergraduate degrees in business, education, or physical education. (More than half had earlier worked as coaches, including 28 percent who served as athletic directors.)

Yeah, that’s pretty old stuff.

Anyway, I blame America’s system of franchise sports. Without promotion and relegation our top leagues are too small and our minor leagues are pointless. High school and college sports fill the gap. A lot of our school districts are essentially football and basketball clubs with schools attached. It’d be as if English secondary schools were run by non-league football (soccer) teams. Let’s privatize school sports!

121

Salient 09.02.10 at 3:26 am

Oh, lordy, principals.

I once had a principal tell me he didn’t know what ‘velocity’ was and that never held him back. (It would have been so many times more hilarious if he had said ‘friction’ or ‘inertia’ but you go to anecdotes with the quotes you have, not the quotes you wish you had.)

He was a former coach, too. And he did teach, he was a social studies teacher because if you wanted to coach football everyone knew you were supposed to become a social studies teacher so you could teach civics directly out of the textbook; his energy devotion (by his own accounting) was pretty much 90% : 10% :: sports : teaching.

My social studies teacher in high school was also a coach. He’d leave class at least three or four times a week to make sports-related phone calls in the hall. We learned everything from the textbook. When he ‘taught’ he’d read it to us; when we had ‘in-class study’ we’d read it ourselves. My other social studies teacher in high school was a former boxer. And a coach (weightlifting). He was a very nice and good-natured guy. I don’t remember him ever actually teaching us anything. I think we read from the textbook in that class, too.

Mind you, those were the halcyon days when nobody was testing to see if kids were learning anything about civics. At least I don’t remember any standardized test on the topic outside of the breezy in-class exams. It was mostly a place where kids could sit for a while and not get in trouble, and those who wanted an A or B (or whose parents wanted an A or B) could read the stuff we needed to memorize for the test. I think some kids slept. They got D’s or F’s, probably. For the rest of us, it was undemanding.

And hell, I learned some facts: Executive, Legislative, Judiciary. I also learned some quasi-facts, such as that only Congress has the power to declare wars (the textbook was very quaint). And it never required any thinking. Maybe I slept through the class too, in a manner of speaking. I remember feeling relaxed.

122

Eli 09.02.10 at 5:55 am

@114 – That quote is insane. There’s so much wrong I don’t even know where to begin. So if one set of kids enters school knowing how to read, with a vocabulary 3x another set of kids who don’t know what letter or numbers are, which parents “know better than anyone else”? And guess whose tax rebate would cover private school tuition? And which parent has the time to homeschool?

Craziness, I tells ya.

123

alex 09.02.10 at 8:27 am

Of course, there’s a sound revolutionary argument for doing away with the government-funded sausage-factories that are public education. Activists could go out amongst the workers, create free revolutionary schools, and educate the children in the ways of critical consciousness. Not to mention singing, dancing, historical materialism, and the care and maintenance of automatic weapons.

Or, alternatively, we could wait a few years until the mass of the population is genuinely unemployable, and then encourage them to eat the rich.

124

Alex 09.02.10 at 12:59 pm

The problem with those options, lower-case alex, is that option 1 tends to turn into Montessori as soon as house prices start rising again, and option 2 may lead to the mob eating the wrong people.

125

bianca steele 09.02.10 at 1:10 pm

@alex:
I’m torn between two answers:
You mean social studies in US schools doesn’t include historical materialism? No wonder our kids can’t keep up internationally!
and
That’s exactly why we have to get social studies out of the schools and get the schools back to the traditional education in history we all remember from our youth.

These play out into your second option differently. In the first, we (well, not “we,” because I don’t know anything about historical materialism except that, according to Kolakowski, it’s exactly the same as medieval nominalism) just have to swarm the intertubes and get people to learn about it. In the second, we just have to swarm the intertubes and get people to become more traditional.

126

bianca steele 09.02.10 at 1:14 pm

Shorter me: Big-A Alex@124.

127

chris 09.02.10 at 1:39 pm

Activists could go out amongst the workers, create free revolutionary schools, and educate the children in the ways of critical consciousness.

Who’s going to fund that? Not the government, since you’ve already done away with government-funded education. Not the students or parents, or they wouldn’t be “free” (I assume you mean free-as-in-beer), and anyway they can’t afford it. Damn sure not the Kochs or their ilk, who don’t want any such project to succeed.

ISTM that you need some very well-heeled class traitors to back this project or it goes nowhere. You need a lot of classrooms and chalkboards and teachers to teach tens of millions of students, and good intentions are not going to conjure up those things.

128

bianca steele 09.02.10 at 2:12 pm

chris,
I agree that little-a alex’s theory doesn’t have an explicit place for the concept of funding, and that in some circumstances it would be helpful, when running into someone who seems importantly to have forgotten where funding comes from, to interrupt the discussion and remind them of funding’s importance.

129

Josh R. 09.02.10 at 5:03 pm

Considering the topic of retraining teachers and the necessity of subject knowledge has come, I am surprising nobody has cited this NYT Magazine article on the topic (or I haven’t seen anybody do so):

http://www.nytimes.com/2010/03/07/magazine/07Teachers-t.html?pagewanted=all

As to firing bad teachers vs. retraining them:

To some degree it feel like there is a good deal of talking past each other. Perhaps a needed bridge is greater distinction between sources of “failure” on the part of teachers. We might be able to group teachers into three broad categories: “good” teachers (those who are trying to impact knowledge to their students and are succeeding, for one reason or another, in doing so); ineffective teachers (trying, but failing for one reason or another – the reasons could be located in the individual’s failing or in factors outside their control); and teachers who have essentially given up trying. This last group is in all likelihood the smallest, but anecdotal information suggests that they do exist. One group seems to mainly be focusing on the last set while another is on the second, which is the source of the confusion (for lack of a better term).

If you could discern the sources of ineffectiveness, you would want to retrain the second (provided you could figure out what you needed to train them on) and either fire the last or structure their incentives so that their motivation is kicked in to higher gear (or some combination – i.e. structure the incentives but if that fails then…).

What is interesting though, is that end of year tests probably cannot differentiate between the last two categories all that effectively. It might give you some clues if necessary covariates are included in the analysis (such a the class’ SES background) but it would fail to tease out those teachers trying but failing from those who aren’t trying all that hard since the end result for both teachers is the same. What you would need in that case would be on the spot evaluations from knowledgeable sources with a background in education and pedagogy.

130

Steve LaBonne 09.02.10 at 5:18 pm

What you would need in that case would be on the spot evaluations from knowledgeable sources with a background in education and pedagogy.

Rather than in basketball coaching. ;)

(My high school guidance counselor was also the basketball coach. All these decades later I vividly recall how totally useless he was in the former capacity- I sure hope he was good at the latter.)

131

piglet 09.02.10 at 5:37 pm

Josh R.: see http://crookedtimber.org/2010/03/05/building-better-teachers/#comments.

This bit from the Green article reinforces my earlier comment: “And the gaps were huge. Eric Hanushek, a Stanford economist, found that while the top 5 percent of teachers were able to impart a year and a half’s worth of learning to students in one school year, as judged by standardized tests, the weakest 5 percent advanced their students only half a year of material each year.”

Leaving aside possible methodological issues, what stands out is that such an outcome is not unexpected at all. It is in fact what you would expect from any profession. The fact that some teachers are better and some are worse than average is not a failure of the system, that is simply how reality works. Of course the question whether and how the multitude of average teachers can learn from their excellent peers is totally legitimate but that is not how the debate is usually framed.

132

Steve LaBonne 09.02.10 at 5:40 pm

Of course the question whether and how the multitude of average teachers can learn from their excellent peers is totally legitimate but that is not how the debate is usually framed.

We could start by structuring schools in such a way that it’s even possible for them to do so, which is largely not the case now.

133

chris 09.02.10 at 6:45 pm

We might be able to group teachers into three broad categories: “good” teachers (those who are trying to impact knowledge to their students and are succeeding, for one reason or another, in doing so); ineffective teachers (trying, but failing for one reason or another – the reasons could be located in the individual’s failing or in factors outside their control); and teachers who have essentially given up trying. This last group is in all likelihood the smallest, but anecdotal information suggests that they do exist. One group seems to mainly be focusing on the last set while another is on the second, which is the source of the confusion (for lack of a better term).

First of all, I think part of the confusion stems from the ambiguity of the English word “bad”. Both categories 2 and 3 are “bad [ineffective] teachers”, but only category 3 is *morally* bad, or acting in bad faith. If we referred to “unskilled teachers”, it would be obvious that the appropriate solution is to improve their skills, not to punish them. The flexibility of “bad” allows you to slide from “producing poor results” to “deserving of condemnation” without looking too closely at whether your underlying facts only support the former. So I think your typology is helpful at untangling that.

That being said, I think there is a substantial group in this debate that denies the existence of the second category, and therefore tars the second and third with the same broad brush (and talks about firing them at the drop of a hat).

This flows naturally out of a Green Lantern Theory worldview: if you can do anything you try hard enough at, it follows that anyone who fails at anything isn’t trying hard enough. Failure is not an option, and similar cheap sentiment. But there may also be other paths to the same conclusion – e.g. someone who hasn’t tried teaching may just be convinced that it’s easy, and therefore nobody could possibly be failing by accident.

Regardless of the basis, denial of Category 2 means you don’t have to worry about whether you are equivocating between different meanings of “bad teacher” because they are all coextensive — all bad teachers are bad in the same way. (Crude Platonism rears its stupid head and endorses this conclusion.)

On the other hand, there are those (including me) who feel that any misclassification of a Category 2 teacher as a Category 3 teacher, let alone concrete action taken against them on that basis, is an injustice, and when it’s done by a public instrument like government then it’s an injustice in which we all in some measure participate. And certainly the *categorical* misclassification resulting from denial of Category 2′s existence would emphatically qualify. Thus, we push back pretty hard to defend teachers who are doing the best they know how, even when that isn’t as good as certain armchair quarterbacks might wish.

If you feel that avoiding sweeping Category 2 teachers up in a dragnet intended for Category 3s is a moral imperative, then obviously any system of teacher accountability needs to be designed much more carefully and avoid the use of blunt instruments. This will naturally lead to a clash with people who just want to fire the lazy bastards because their lack of results is proof enough of their lack of effort, and smash any social institution that shelters them, such as teachers’ unions.

(Note: I leave to one side the disputants for whom union-smashing itself is the primary motivation and education is just a sideshow, although there probably are some.)

134

engels 09.02.10 at 8:53 pm

Let’s also leave aside the fact that the people who decide policy for state education often didn’t go to a state school themselves and wouldn’t dream of sending their kids there.

135

Matt 09.02.10 at 9:57 pm

Let’s also leave aside the fact that the people who decide policy for state education often didn’t go to a state school themselves and wouldn’t dream of sending their kids there.

I think that depends very heavily on the state, and is much, much more common in some areas (mostly east coast) than others. In much of the west, there is not nearly as much bias against state schools (and there are usually very few private high schools), and often even a bias against “ivy league types”. For a long time, the majority of state legislators in Idaho hadn’t gone to college at all- I think it was only in the 90′s that that changed, and by far the most common place to have gone is the University of Idaho. Things are similar, if a bit less pronounced, in much of the west, and perhaps in the mid-west, too, though I don’t know. The big bias against state universities, and also against public schools, is mostly a east coast thing.

136

Harry 09.02.10 at 10:19 pm

engels and Matt — engels means state education in England (how many of the current cabinet attended state schools? How many had kids who go/went to them?). It’s true that legislators in US states tend to have gone to public schools, and I imagine they send their kids to them too. Unlike Democratic Congressmen, Senators, and Presidents, of course (whereas senior Labour politicians pretty much have to send their kids to state schools or resign these days).

137

piglet 09.02.10 at 10:23 pm

engels/Matt/Harry, are you referring to High School or University? (The US use of ‘school’ always confuses me.)

138

bianca steele 09.02.10 at 10:39 pm

@Matt:
Certainly there were those among my mother’s friends who were nearly gleeful in their certainty that anyone attending a public high school was getting an education in depravity. I imagine this was worse at the tail end of the 70s than it might have been earlier, and that things in Philadelphia and its suburbs are still pretty much the same in that regard.

@Harry:
IIRC David Brooks has made it clear that nobody in his social circle would send their kids to public school even in the tony suburbs they surely live in.

@piglet:
US “high school” is any school that teaches ages 14-18. US “college” is any bachelor’s degree granting school. (There are high schools with “college” in the name, but that doesn’t change it.) Harry will correct me if I’m wrong, but I understand the work he is citing to involve only pre-university education.

139

engels 09.02.10 at 11:07 pm

‘senior Labour politicians pretty much have to send their kids to state schools’

Token left-wing leadership candidate Diane Abbott being one exception (if not ‘senior’). Ruth Kelly another. How many in Blair’s cabinets were privately educated?

140

engels 09.02.10 at 11:12 pm

Educational apartheid and deficit hysteria are actually two points where Britain leads the US!

141

engels 09.02.10 at 11:23 pm

142

Harry 09.02.10 at 11:36 pm

Completely forgot about Ruth Kelly. Dianne Abbott isn’t exactly senior. One labour councillor in Oxford in the 80′s resigned his city council seat when he sent his kids private.

Yes, I mean k-12. I regard all universities as public, whatever they call themselves.

Bianca — your comment to Matt is interesting. My impression is that, with some exceptions (CA, NY, maybe others) legislators neither come from fancy backgrounds by and large, nor have much money to send their kids private, And remember that the tony suburbs provide elite high spending educations with no poor kids to deal with, all for free as public schools.

Engels – wow, that is a much higher proportion of state educated ministers than I imagined. Curious whether the grammar school kids would have been privately educated if they’d been young enough for comps, or in comprehensivised LEAs.

143

Matt 09.03.10 at 12:04 am

Sorry for the confusion as to what Engels was talking about, and the lack of clarity of what I was talking about. Partly, on my part, that was due to not wanting to take the time to separate distinct cases. So, my impression (from talking to people, not so much first-hand) is that, in the South, many white people who can send their kids to private primary schools, including high-schools, but that public universities there are not looked down on at all, and would even be expected among the “elite”. In the west in the US, private primary schools are quite rare, except for some Catholic schools and the like. Public universities, and often enough not super fancy ones, are not looked down on at all. The worst, in the US, for looking down on public education of all sorts is the north east.
Bianca- the schools in Philadelphia are still quite bad, and many, perhaps most, people who live in the city who can send their kids to private schools. Many schools have a 50% or less graduation rate, and some have had very serious violence in recent years, both inter-racial (black vs. Asian, mostly, recently) and violence against teachers, with some being very seriously injured. It’s a sad situation that ought to be seen as shameful for everyone around.

144

Harry 09.03.10 at 12:14 am

But, let it be said, UPenn (Matt’s empoyer) runs a public primary school (k-8?) which serves a socioeconomically mixed population and which ANYONE would be glad to have their kids in.

145

Sebastian 09.03.10 at 12:25 am

“If you feel that avoiding sweeping Category 2 teachers up in a dragnet intended for Category 3s is a moral imperative, then obviously any system of teacher accountability needs to be designed much more carefully and avoid the use of blunt instruments. This will naturally lead to a clash with people who just want to fire the lazy bastards because their lack of results is proof enough of their lack of effort, and smash any social institution that shelters them, such as teachers’ unions.”

But isn’t the denial of Category 3′s existence (the teacher’s union position) also a little bit of a problem? Hell the frankly deny the ability to effectively tell the difference between category 1 and category 3 much less between category 2 and 3.

Harry, “My impression is that, with some exceptions (CA, NY, maybe others) legislators neither come from fancy backgrounds by and large, nor have much money to send their kids private”

I’m not sure if Michigan is typical ( it certainly isn’t a CA or NY), but apparently there legislators in the Senate are 3 times as likely and legislators in the House are twice as likely as the general population to send their kids to private schools. cite

Hawaii, half of them send their kids to private school, 3 times the private rate. cite

“And remember that the tony suburbs provide elite high spending educations with no poor kids to deal with, all for free as public schools.”

This isn’t true in most of the larger states. California for example is required by the state court system to generally equalize the funding across all the districts. (Serrano v. Priest) and has done so since the 1970s. Unequal spending in school districts by local taxes still exists in some states, but is no longer the big deal that it once was.

146

bianca steele 09.03.10 at 1:37 am

Matt,
I’d argue that things weren’t quite so bad 25 years ago, but in any case what’s happened to the Philadelphia schools illustrates what I said above. The school district did not go to pot because Philadelphia elected a black mayor, or because the PFT made work-rule complaints every time a principal tried to put too many students in a class (which happened every single year in the majority of my classes). It wasn’t the fault of the hippies either. And it isn’t the fault of the people who live in the neighborhood now. It was, largely, the fault of people who took a good city education when it was there for them and then moved to the suburbs, leaving other people to pay the taxes; of people who ignored their neighborhood schools because only real losers attended them but expected them to take their delinquents when the archdiocese couldn’t handle them any more; and of administrators who used “teach the black kids this year” as a punishment.

147

Matt 09.03.10 at 1:59 am

I’m perfectly willing to believe that all of that is so, Bianca.

The school Harry mentions, the Saddie Alexander School, is quite good. It’s one of the most successful of a series of public-private attempts to run various schools that have been tried here. There was talk of Penn running a highschool, too, but I think that’s no longer in the words, at least as far as I know.

148

bianca steele 09.03.10 at 2:03 am

Busing, at least in my neighborhood, may have been an issue emotionally . . . but Philadelphia is so big, and so segregated, and has such lousy bus service, that there wasn’t that much of it. Through the 1980s, it was voluntary only, and largely motivated by setting up charter schools and new selective high schools.

149

bianca steele 09.03.10 at 2:04 am

“charter” sb “magnet”

150

Salient 09.03.10 at 2:10 am

But isn’t the denial of Category 3’s existence (the teacher’s union position) also a little bit of a problem?

Dear god, Sebastian: we’re not representatives of any teacher’s union, here. (Or if any of us are, we’re sure not acting in our official capacity.) If you want to argue with teacher’s union representatives, or hate on them, please find a more sensible place to do so. Some of us have come to the same conclusion as you attribute to the monolithic entity of teacher’s unions, but by way of different arguments; please engage what we’re saying, not what the union bosses are saying in your head when you try to go to sleep at night.

Oh, and this comment isn’t meant to dismiss or disparage teacher’s unions. Teacher’s unions exist to protect their members as a whole from unfair treatment. So sure, this suggests that some individual teachers may benefit from their advocacy who shouldn’t, according to whatever harsh moral standard we might apply. And sure, let’s be analytically hesitant about taking education policy advice from the NEA, just as we might be analytically hesitant taking climate policy advice from Cato, or hell, from anybody. Fine. OK. But we’re not putting forth their arguments, here, we’re putting forth our own.

Of the ~170 school districts in my state, only 15-20 are represented by local teacher’s unions (margin of error because I’m too lazy to check a few of the localities whose status I’ve forgotten since last checking). There’s a state union which I used to belong to and still follow carefully; that state union doesn’t seem to ever say any of the things you attribute to “teacher’s unions.” One of their big fights, recently, was to try and make sure that teachers are having a lunch break in which they are not responsible for student oversight, i.e. an actual break.

151

JanieM 09.03.10 at 3:25 am

Harry: And remember that the tony suburbs provide elite high spending educations with no poor kids to deal with, all for free as public schools.

Sebastian: This isn’t true in most of the larger states. California for example is required by the state court system to generally equalize the funding across all the districts. (Serrano v. Priest) and has done so since the 1970s. Unequal spending in school districts by local taxes still exists in some states, but is no longer the big deal that it once was.

There was some discussion about school funding at Obsidian Wings recently, during which Sebastian alleged that “at this point most large states don’t engage in school funding via local property taxes.”

I linked to Census Bureau numbers that indicated that only 3 of the top 10, and 7 of the top 20, largest states get more school funding from the state than from local sources (Table 1), and also that the vast majority of local money in most states still comes from property taxes (Table 4).

Sebastian being Sebastian, he never acknowledged these numbers, or clarified what he was trying to say. Now he’s over here making a slightly different claim, except that California is still being used to stand in for “most of the larger states.”

It’s probably not news, but despite Sebastian’s assurances some people seem to think that unequal spending in school districts is still a big deal:

The wealthiest 10 percent of U.S. school districts spend nearly 10 times more than the poorest 10 percent, and spending ratios of 3 to 1 are common within states.

— From here.

152

b9n10nt 09.03.10 at 4:56 am

And, regardless of state funding schemes meant to provide equal state funding, there are regular school local bond measures that glibertarians’ wives pass every year in the affluent suburbs that poorer towns can’t get the vote for. And then there are the local/city- level educational foundations which raise 10′s of thousands of dollars a year by fundraising from parents that never occurs in poorer districts.

No, public school districts are funded quite unequally due to these local initiatives. Though it’s also true that wealthy localities do in fact pay for their priveledged public schools and do not passively engulf state revenues.

153

chris 09.03.10 at 1:18 pm

But isn’t the denial of Category 3’s existence (the teacher’s union position) also a little bit of a problem?

It would be if it weren’t a strawman.

Please point out to me where a teacher’s union has advocated that no teacher should ever be fired for any reason, or that no teacher should be fired for any reason other than criminal indictment/conviction.

School districts and administrators *already have* the authority to fire teachers who aren’t making any effort to teach. They don’t need a new program or test scores to do so. At most they need some proof that those teachers really aren’t making an effort and the firing isn’t totally arbitrary; and then they have to file some paperwork.

This is a bit different for *university professors* with tenure, but that’s a contractual provision which I don’t think a new law could screw with anyway (contract clause), and is also becoming much rarer.

Teachers’ unions most likely do deny that Category 3 is of any significant size or effect on the overall performance of the educational system — but anecdotal evidence of a few Category 3 teachers, even if accepted at face value, can’t really refute that position anyway.

154

piglet 09.03.10 at 1:34 pm

Bianca 138: In US usage, state school (and public school) often refers to a state university (random google http://www.scholarships.com/resources/college-prep/choosing-the-right-school/pros-cons-state-universities.aspx), not a school where you’d find children. In fact I don’t ever hear “state school” referring to a secondary school but that may be regional. School without qualifier can refer to both secondary and post-secondary institutions (not to mention law and medical school).

155

JanieM 09.03.10 at 2:23 pm

I don’t ever hear “state school” referring to a secondary school but that may be regional.

This is my experience as well (60 years’ worth, and 10 or so states of residence in different parts of the country). It makes sense, since they’re not “state” schools (as in Nebraska, Ohio, etc.), they’re local schools, controlled by local school boards.

“Public school” in my experience most often refers to local schools for pre-college-age people (grade school, elementary school, grammar schoool, middle school, junior high school, high school, lots of regional variations in what ages go where, and what terms are used), whereas yes, “state school” pretty much always means college/university. I don’t think I’ve ever heard “public school” used for colleges.

156

engels 09.03.10 at 2:37 pm

The Prime Minister’s attitude has been perfectly consistent ever since he became Labour leader: he is a normal, middle-class family man who wants the best for his children. He would like, as a good, liberal-minded, socially conscious sort, to send them to the local comprehensive but it is, alas, a little rough, and it doesn’t get enough students into Oxbridge. So he carts his children across the city to the London Oratory, a school that is technically a comprehensive but is widely regarded as being both socially and academically selective. To this advantage, he adds tuition from teachers at Westminster, one of the country’s most élite private schools. It hardly needs saying that, even if children from the poor areas of London got into the Oratory, their parents couldn’t afford £100 a week for tuition.

OT anyone read St. Tony’s memoirs yet?

157

Sebastian 09.03.10 at 3:57 pm

“I linked to Census Bureau numbers that indicated that only 3 of the top 10, and 7 of the top 20, largest states get more school funding from the state than from local sources (Table 1), and also that the vast majority of local money in most states still comes from property taxes (Table 4).”

Janie, that is because you are citing sourcing, not distribution. If you look, for example, at what you reference (Table 4) for California under your understanding, you would see that about 1/3 of the schooling money is *sourced* through property taxes. That is true, but it doesn’t mean what you imply. Serrano v. Priest said that it was a violation of the California constitution to allow such disparity of funding by district, so California tries to equalize it. So you aren’t using the chart properly. You are using it to imply that 1/4 funding by property taxes in California means unequal spending per student. It does not. (What actually happens is much more complicated than it should be. Essentially California distributes quite a bit of state funds. It reviews how much has been raised in property taxes, sets a revenue limit on the richer schools, and pumps more money into the poorer schools.) Most of the other large states have similar rules. So for a very large majority of public school students in the US, this vast disparity in potential property tax revenue isn’t an important issue.

Now there are states that don’t have such rules. And they exhibit quite a bit of spending inequality. But for the most part they aren’t the states that educate very many of the students in the US. So while it is something that ought to be corrected, it can’t be talked about as if it is a typical characteristic of US public schools. Which is why I said things like “at this point most large states don’t engage in school funding via local property taxes.”. If you want to focus on what is wrong with the schooling for a vast majority of students in the US, focusing on the unequal funding issue isn’t going to get at the problem because it isn’t a factor for the larger states which (by virtue of being both larger and younger than most of the smaller states) educate the vast majority of the students.

California, by way of example, has lots of problems. But unequal funding isn’t one of the big ones. The Census chart can’t be used the way you’re citing it.

And since I would have said the exact same thing in the obsidianwings thread if you hadn’t posted at the very end of the thread as it was petering out, you don’t need to read anything into my lack of response to you there. Unless it makes you feel better, in which case go all out.

158

JanieM 09.03.10 at 4:12 pm

That is true, but it doesn’t mean what you imply.

So you aren’t using the chart properly. You are using it to imply that 1/4 funding by property taxes in California means unequal spending per student. It does not.

I wasn’t “implying” anything more than I said. I made no statement whatsoever about what inequality. You are refuting a figment of your own imagination.

The original exchange at ObWi started with you saying that “at this point most large states don’t engage in school funding via local property taxes.”

The numbers I linked to indicate that that is flat out not true; a very great deal of school funding still comes from property taxes all over the country. That is not a statement — and I didn’t make one — about inequality; to make a statement about that would require digging deeper, which I did not do: nor did you, and that’s my point.

Here at CT you changed your assertion to “that [unequal spending] isn’t true of most large states.” I had seen you make an unsupported and, it turned out, incorrect assertion about “large states” before, so I decided to question whether what you said about “larger states” this time could be trusted any more than last time.

You have still haven’t provided any actual evidence that shows that the California model, or something like it, is followed in other “larger states.” You’ve only asserted it. Repeatedly. That’s not evidence.

159

JanieM 09.03.10 at 4:16 pm

Me: I made no statement whatsoever about what inequality.

Meaning, I didn’t draw any conclusions about inequality from the Census Bureau numbers.

But I’ve had my quota of standing in for straw men for the day. Month. Century.

160

Sebastian 09.03.10 at 4:34 pm

Oh good heavens Janie. The discussion both here and there was about how unequal school property tax funding hurts children.

You know it, I know it, and anyone reading this thread can see it for themselves. I can’t imagine why you would think that the precise origin of the taxes is remotely interesting to the discussion if you divorce it from the discussion of inequality.

It is specifically what I’m responding to and exactly what I’m talking about.

Harry: And remember that the tony suburbs provide elite high spending educations with no poor kids to deal with, all for free as public schools.

Sebastian: This isn’t true in most of the larger states. California for example is required by the state court system to generally equalize the funding across all the districts. (Serrano v. Priest) and has done so since the 1970s. Unequal spending in school districts by local taxes still exists in some states, but is no longer the big deal that it once was.

That is the context of the discussion. And it was the context of all three of the recent education discussions over at obsidianwings where the issue came up. If you divorce it from the context, it makes no sense. If it really is true that you were making no statement whatsoever about the inequality, you weren’t engaging in the conversation. I can’t really be blamed for assuming that you were participating in the conversation, right?

161

roac 09.03.10 at 4:36 pm

As a footnote to the discussion of school finance in California, it used to be possible, and may still be, for industrialists to band together and incorporate as “cities” with minimal population, and therfore avoid being taxed to support anybody’s schools. As the most absurd example, I give you Sand City, pop. 261. But the City of Vernon and the City of Industry, in the LA area, are more consequential (there are likely others I don’t know about).

162

Salient 09.03.10 at 4:38 pm

The discussion both here and there was about how unequal school property tax funding hurts children.

gee, I thought the discussion here was about “using test scores to evaluate teachers”

I wonder what might have given me that impression.

163

JanieM 09.03.10 at 4:43 pm

So, according to Sebastian, making sh*t up ( “at this point most large states don’t engage in school funding via local property taxes”) is a valid way to participate in the conversation, but pointing out when people make sh*t up isn’t.

Salient, my apologies, I’m done now. I should have known better than to start.

164

Sebastian 09.03.10 at 6:09 pm

Salient, I was responding directly to Harry, who as the author of the piece is probably allowed to decide if he thinks he is on topic. My response was directly on point with what he raised. JanieM either seems to have wanted to talk about something else entirely, or had misunderstood the data she cited. Since she quoted from my response to Harry, I assumed it had to do with my response to Harry. Since she now claims that she was raising something that has absolutely nothing to do with the inequality of funding issue, I have no idea whatsoever why she thought it made sense to quote me. The context is clear, and my claim is specifically couched in those terms :

“This isn’t true in most of the larger states. California for example is required by the state court system to *generally equalize* the funding across all the districts. (Serrano v. Priest) and has done so since the 1970s. *Unequal spending in school districts by local taxes* still exists in some states, but is no longer the big deal that it once was.”

If for some reason she believed that the data she quotes is somehow a response, it would be nice to know what she was thinking. At obsidianwings, there were 3 series of comments where school funding, taxes, and California came up, and in all of them the inequality context was prominent. Trying to play gotcha over it by removing the inequality context seems like she was either not paying attention to the discussion or being deceptive. Doing it here, where it isn’t even contextual, but is in fact completely explict in the direct statements she quotes, strikes me as just weird.

165

Harry 09.03.10 at 6:15 pm

Sebastian — what is true is that urban schools with high concentrations of poor kids are higher spenders than most suburban districts. But the top spending 5% of districts in most states are spending more, and are catering to very low need populations. Take a look at the figures for Palo Alto Unified, for example (a public school district in CA). My point was that it is somewhat misleading to compare the way British politicians (who have to go private for an elite education) behave with the way that US politicians (many of whom can get an elite education paid for by the taxpayer) behave.

166

Sebastian 09.03.10 at 6:51 pm

Sure, I have no idea about how British politicians act. It looks like US politicians send their kids to private schools at somewhere between double and three times the rate of their constituents, for whatever that is worth. I suppose the argument could sort of go either way (i.e. an absolute numbering–only about 1/3 to 1/2 of legislators, depending on the state, send their kids to private schools. Or alternatively relative analysis–two to three times as many legislators, depending on the state, feel they need to send their kids to private schools when compared with their constituents.) The funny/sad thing is that in this particular argument both sides are super-likely to seize on the statistics of the style directly opposite their general arguments. Conservatives who always want to talk about absolute poverty standards are likely to cite the legislators who send at 2 to 3 times the normal rate, while liberals who heartily embrace GINI coefficients in other circumstances will want to point out that it is only 1/3 or so who do. And the statistical games go on and on.

Comments on this entry are closed.