4 ms·
Report: Use student performance to rate teachers
- tptacek 17y agoIn other words, if you want to make more money as a teacher, make sure you're not posted somewhere where they really need good teachers; your best bet, for maximum student payoff for each extra hour of teacher effort, is a rich suburb.
- tokenadult 17y agoWhat if what is looked at is IMPROVEMENT?
- tptacek 17y agoWhat if improvement is much harder to achieve in poor inner city schools? Wait. Take the "what if" out of that sentence. Replace the question mark with a period. Incentive pay: good. Reason we don't have a good incentive pay system to argue with the teachers unions about: because incentive systems for teachers are hard. Parent of two school-age kids speaking now: teach-the-test pressure is already bad enough without paying teachers to ratchet it up.
- tokenadult 17y agoBut are you saying that the reform proposed in the originally submitted link would make things WORSE for children in poor inner city schools? Right now their teachers are, for the most part, not evaluated for effectiveness by any standard. And we all know that the kids in the suburbs are better off (not least, because they live in suburban homes in suburban neighborhoods, but also because of schooling quality) than the kids in poor inner city schools. I haven't seen a proof here yet in your kind replies and Alex's that the reform proposed in the submitted link would schools worse for any kind of student. I'm a parent of four school-age children, and living in a whole neighborhood full of children, and I do hope the next generation sees some improvement in education. After edit (and your initial reply to this post): My apologies, I forgot that the submitted link here was a news source, rather than the full report http://www.policy2.org:85/images/stories/documents/policy2.0policypaper.pdf http://www.policy2.org:85/images/stories/documents/policy2.0... of the organization advocating this change in policy. The stakeholders consulted in preparing the report may not all enthusiastically support student testing programs (some of them surely do not), but the report claims that the trade-offs of linking teacher evaluation to student achievement are better than the trade-offs of not linking teacher evaluation to any criteria in particular. I think that is a reasonable view.
- tptacek 17y agoYes. That's what I'm saying. In the absence of any reasonable notion of what teacher performance means and how to measure it, tying teacher comp to arbitrary standardized performance metrics is a bad thing, for exactly the same reason as it's bad to comp developers based on number of bugs opened or lines of code committed. The only difference I see between the two instances of this "solution" are that the incentive comp schemes dev teams come up with are simply fodder for amusing DailyWTF posts, and the punitive "incentive" schemes on the table now for schools have already begun screwing over a whole generation of kids. We all hope the next generation sees some improvement in education, don't we? [ps: I took the time to actually read the "Policy 2.0" report that is the subject of the AJC article. It contains no data. It builds a case for formal teacher evaluation by using the same logic you do: "studies teachers are important, so it follows that a plan to segment teachers by performance would be beneficial". I'm particularly happy to report that almost none of the study's contributors are even named, nor is a methodology specified beyond "we have an internet site, we recruited volunteers, and we led discussion groups and documented the "results".]
- tokenadult 17y agoLinks about a current problem (that a substantial number of teachers are just flat-out incompetent, by any standard of evaluation), which seems to have been going on for a long time (to judge by the dates of publication of the various links): http://www.ericdigests.org/pre-921/incompetent.htm http://www.ericdigests.org/pre-921/incompetent.htm http://www.amazon.com/Incompetent-Teacher-Managerial-Responses-Education/dp/0750700505/ http://www.amazon.com/Incompetent-Teacher-Managerial-Respons... http://www.schoolmatch.com/audit/jacksonville/articles/teach1.htm http://www.schoolmatch.com/audit/jacksonville/articles/teach... http://web.missouri.edu/~podgurskym/papers_presentations/presentations/aei_paper_5.pdf http://web.missouri.edu/~podgurskym/papers_presentations/pre... http://reason.com/archives/2006/10/01/how-to-fire-an-incompetent-tea http://reason.com/archives/2006/10/01/how-to-fire-an-incompe... http://www.maa.org/features/083104pointcallforaction.html http://www.maa.org/features/083104pointcallforaction.html http://www.newyorker.com/reporting/2009/08/31/090831fa_fact_brill http://www.newyorker.com/reporting/2009/08/31/090831fa_fact_...
- tptacek 17y agoYes, we get it. You're a libertarian (Podgursky, Reason, even the Tennessee math professor you cited). Also: one can simultaneously believe that incompetant teachers are too hard to fire and believe that "Race to the Top" and NCLB and charter schools are terrible, counterproductive ideas. Two other comments about your links: * None of them address the concern I raised, which is not a liberal one but a simple market/economics issue. * The last link is particularly interesting because Joel Klein, the NYC superintendent who is its subject, is hugely controversial.
- Alex3917 17y agoWhat about all the evidence saying that when you pay teachers for students performance, students learn less?
- tokenadult 17y agoCitations, please?
- Alex3917 17y agoAlfie Kohn cites a bunch of studies in Punished by Rewards.
- tptacek 17y agoHere's a good starting place: http://www.alfiekohn.org/teaching/edweek/meritpay.htm http://www.alfiekohn.org/teaching/edweek/meritpay.htm Also note that Kohn's ideas on extrinsic versus intrinsic motivation are hardly original. Most people who work at startups should find them collegial, since they're one of the key ideas in books like _Peopleware_.
- tokenadult 17y agoI hope there is somebody besides Alfie Kohn writing about this. http://www.britannica.com/blogs/2009/02/alfie-kohn-is-bad-for-you-and-dangerous-for-your-children/ http://www.britannica.com/blogs/2009/02/alfie-kohn-is-bad-fo... But what about the point in the originally submitted link that in most states NOTHING in particular is done to evaluate teachers?
- tptacek 17y agoThat's what you have to contribute here? A blog post taking Kohn to task for his tone? It's definitely true: Kohn is an archetypical bleeding heart liberal. I'm pretty liberal, and he makes my eyes roll. But that doesn't make him wrong. What he's saying about incentive pay harmonizes with what we as technology entrepreneuers already know about incentive comp, and with what we're already observing in school districts around the country. The teachers in Oak Park don't even get a pizza party for jacking up ISAT scores, but those scores are still a top-of-mind issue for them year round. A reasonable person could argue that it's better not to do formal evaluations at all, rather than smothering the whole school system with meaningless standardized tests.
- schindyguy 17y agoWouldnt that just skew all grades higher? Why wouldnt I want to give out all A's to my B and C students if it meant hire pay? Or am I missing something
- tptacek 17y agoYou're not missing something. You just spot it quickly because the idea of metric-based performance pay has been such a disaster in the technology industry (pay extra for devs not to "make bugs", and they simply stop filing bugs in the bug database).
- pbhjpbhj 17y agoYou don't have to do that, just assess them at a lower level at the start of the course!
- techiferous 17y agoI'm a programmer but I took a three-year hiatus from programming to be a science and technology teacher. So I speak from having some teaching experience. Objective measures like the ones proposed are a siren song. They sound really good on the surface, but a deeper look reveals these problems: * You only get what you measure. Whatever is not measured falls through the cracks. So this means that the teaching goals have to be really good. Historically, they are not, as they are driven more by politics than by people who are talented at teaching. * Some of the value that teachers add is in intangibles which are harder to measure. These important intangibles will fall by the wayside. Also, the administration has a large impact on student and teacher success. Imagine being a programmer and having your performance judged on a project that was mismanaged by an incompetent project manager. I'm not saying we should have no objective measures; I'm just giving a word of caution that this is a complex issue that's not easily solved. Reductionist approaches like these are bound to fail; more holistic approaches that focus on the student-teacher-parent relationship are better.
- teeja 17y agoNonsense. You can't rate the countless dozens of things good teachers do with some phony metric. Any more than you can measure 'good' programming, 'good' science research, or any other human endeavor that requires creativity, tailoring responses to the situation, etc. "Objective" measures are a pipe-dream that's decades old. It's constantly thrown up as a 'solution', almost always by people with little or no education experience. Teachers have little control over the attitudes / motivations of students. You can bring a horse to water. Again, the evaluation of experienced professionals is worth something ... but good teaching is at least 50% art, and what works is a mystery that has to be solved one class at a time.