Rating teachers

New York City teachers’ performance will be measured based on how much their fourth- through eighth-grade students improve on annual state math and reading tests, reports the New York Times. But, in response to union complaints, the reports won’t influence formal job evaluations, pay or tenure decisions and will not be published.

The reports include data on the number of low-income and special-education students in each teacher’s class, “as well as the size, race and gender breakdown.”

Using a complicated statistical formula, the report computes a “predicted gain” for each teacher’s class, then compares it to the students’ actual improvements on the test. The result is a snapshot analysis of how much the teacher contributed to student growth.

The reports classify each teacher as average, above average or below average in effectiveness with different categories of students, like those who score in the top third or the lowest third on the test, and those still learning English or enrolled in special-education programs. It also contains separate measurements on effectiveness in teaching boys and girls, though it does not distinguish performance by students’ race or income level. Teachers will also be given a percentile ranking indicating how their performance compares to those who teach similar students and to a citywide pool.

This is a very soft version of the original idea, says Eduwonk.

About Joanne

Comments

  1. The reports “won’t influence formal job evaluations, pay or tenure decisions and will not be published.” However, I wonder how they effect administrators’ opinions of a teacher’s value. And I wonder how those opinions will influence formal job evaluations. Information does not exist in a vacuum.

  2. I wonder why anyone would bother? What’s the purpose of the evaluation if the results are to be kept secret and they won’t have any effect on, well, anything?

    I think this…

    the reports are designed to be guides for the teachers themselves to better understand their achievements and shortcomings.

    ..is supposed to be an attempt at an explanation but it hardly seems worth the effort.

  3. David Cohen says:

    allen quoted:
    “the reports are designed to be guides for the teachers themselves to better understand their achievements and shortcomings.”
    — and responded:
    ..is supposed to be an attempt at an explanation but it hardly seems worth the effort.

    I would say that if such data actually helps teachers understand and improve, that’s certainly worth the effort, but I’d be curious about many elements of this plan.

    – How reliable are the tests? How sensitive are they to instruction?
    – How much pressure will this create to teach to the test?
    – What other types of school experiences and other types of understanding might be neglected to raise test scores?
    – How will they calculate the expected gains for different types of students in a reliable way that takes into account numerous factors that influence growth?
    – How will they adjust these expectations in the face of realities such as student illness, teacher illness, family tragedies, school tragedies, etc.?
    – How will they control for influences beyond the teacher’s control but still within the school? For better and for worse – it’s not too hard to describe which literacy practices outside of tested subject areas are still going to influence student growth during that year. If students in my English class are clustered in a way that sends them to strong literacy teachers in other subject areas, I’ll look better. If students in my English class are clustered in a way that lands them with other teachers who think literacy is only for English teachers, I’m fighting a much bigger battle.

    I’m not against data and accountability, but if that’s only to be determined by limited data gathered at limited times and then over-interpreted, I’m skeptical. If you want to offer me 4-5 measures taken in different times and methods, then I think we can look for more meaning in and across the data and I’m glad to work on improving my practice.

  4. Good administrators already know who the good and poor teachers are. These statistics will enable the administrators to go to the underperforming teachers and persuade them that they really do need to improve. The same sort of thing is already happening with the NCLB test scores in my wife’s district–the teachers can no longer deny that they have to improve. And the majority of teachers want to do better once they realize that they aren’t the great teachers they thought they were.

  5. We don’t need great teachers, having stripped the great ideas from the curriculum.

  6. I’m a curriculum guy, mlu, so I’m inclined to agree. But whether we need “great teachers” or not, we simply don’t have enough and we’re never going to. Talent isn’t that widespread.

    If data such as this is used to hound teachers out of the classroom, I’m not sure we’ve gained much. Identifying mediocre teachers is a start, but finding a way to make them competent — not superstars, but adequate, competent teachers — is the only reasonable solution at scale. As Donald Rumsfeld did not say, but I will: You go to school with the teachers you have, not the teachers you wish you had. National content standards, assessments, and curriculum would be a start. It might allow us to focus our energies on helping teachers learn best how to teach instead of leaving them on their own, trying to figure out what to teach.

  7. David Cohen wrote:

    I would say that if such data actually helps teachers understand and improve, that’s certainly worth the effort, but I’d be curious about many elements of this plan.

    If it were the case that data might help teachers to understand and improve then I’d expect to see the techniques and ideas of such well known lights of the profession as Jaime Escalante and Marva Collins, dissected, discussed and adopted by teachers new and old. That hasn’t happened though, has it? Care to explain why this hasn’t happened?

    I’m not against data and accountability, but if that’s only to be determined by limited data gathered at limited times and then over-interpreted, I’m skeptical.

    I think you are against data and accountability but self-respect or pragmatism demand more acceptable objections to the testing that necessarily precedes the taking of action. After all, if you don’t have to work to any standard other then your own pride, whatever that demands of you, why should you?

  8. I think the hope that nationalization and centralization will fix things is an illusion.

    We have a governance structure that pretty much makes it impossible for anyone to do much of anything, since everything, including the meaning of any test score, is debatable and someone who is feeling left out will have multiple avenues for debating. So we always spend our time arguing about the work rather than doing it.

    I would favor freedom and authority, where local schools (not public) could govern by their own lights (but people only attended such schools of their own free will).

    There would be lots of crappy schools, since there would still be lots of dopes.

    But there would be more good schools, and such schools would set the true standard for what could be done and how, for those honestly interested in such things.

    Actually, I think national-scale curricula and standards would be a good idea, as long as they were unlinked to accountability and enforcement (coercion). Basically, I would say centralize information and disperse decision-making.

  9. Andy Freeman says:

    > If data such as this is used to hound teachers out of the classroom, I’m not sure we’ve gained much.

    We’ll have gained positions to use for good new teachers.

    > Identifying mediocre teachers is a start, but finding a way to make them competent — not superstars, but adequate, competent teachers — is the only reasonable solution at scale.

    That’s true only if there are no good new teachers.

    One might even think about pushing the predictors for failure back into the ed schools and weeding out the poor even earlier. Maybe that opens up space for folks who can succeed or maybe it just means that we waste less money training folks who won’t. Either way, it’s an improvement.

    > As Donald Rumsfeld did not say, but I will: You go to school with the teachers you have, not the teachers you wish you had.

    Rumsfeld worked to change the military that he had.

  10. <<< We’ll have gained positions to use for good new teachers.

    Last time I checked, the line of applicants at my South Bronx elementary school is still not forming. I’m not suggesting we should suffer bad teachers, but to suggest there’s a ready supply of good teachers is to ignore the facts.

    After thinking about this thread all weekend, I posted a lengthy piece about it over at the Core Knowledge blog:

    http://www.coreknowledge.org/blog/2008/10/06/teacher-quality-unintended-consequences-and-the-baseball-achievement-gap/

  11. Good teachers are out there, but they’re way too smart to go teach in a urban school district. We don’t have any trouble attracting good teachers where we are in Upstate Central New York, but we do have to pay a decent wage.

  12. Bill Leonard says:

    The comment by Rex is the only one that makes any sense. At some point it may not matter much how competent the teacher is. Teachers ultimately cannot overcome prevailing social conditions.

    Case in point: my niece, essentially a first-year teacher on her own, is in an urban California setting in which, first day of school, she had a fourth-grader throw a desk at her. The administration, such as it is — who did nothing about the desk-tossing episode — has counseled teachers at her school not to wear red or blue, since significant numbers of the K-6 school where she works are 2d- or 3d-generation gang families (nortenos and surenos, if you understand much of the California paradigm.) Most of the kids have little interest in being there. Those who are interested — a small handful of Vietnamese-American kids — seem scared shitless (who wouldn’t?). No teacher stays after dark — or probably, any longer than they have to.

    Will she stay in teaching? I dunno. I wouldn’t. Greeting at WalMart would be a better deal.

    And so the real problem: How do we deal, realistically, with this social subset — beyond the guaranteed certainty that about 85 percent of her kids will either be dead or in prison by age 24?

    Bill

  13. For what it’s worth, I agree with mlu that hoping that “nationalization and centralization will fix things is an illusion.” Robert P says “it might allow us to focus our energies on helping teachers learn best how to teach instead of leaving them on their own”. Well, it might, but I don’t see how. To me it seems more likely that standards imposed from above would primarily focus teachers’ attention on how they will be judged, with a net detraction of attention on what’s best for their students. I don’t have much direct experience with the impact of NCLB, but from what I read in the blogs it appears that it mainly detracts from what teachers want to do, and should do.

    Allen asks a very good question, why are not the techniques and ideas of successful teachers “dissected, discussed and adopted” by others? I don’t know the answer to that, but it does bring up two lines of thought.

    First, in the field of education we are not good at describing. I happen to think that simple discription is very important. Data, derived by contrived experimentation, certainly is important too, but without a foundation of good description data may be irrelevant, or actually misleading. I have extended these thoughts at http://www.brianrude.com/lackdes.htm.

    Second, the educational “establishment”, meaning ed school, takes an ideological approach to pedagogy, as opposed to a scientific or analytic approach. Thus we don’t do a very good job at dissecting or discussing anything about teaching and learning. I have extended this idea at http://www.brianrude.com/indict-ed.htm.