Weaker teachers leave under new tenure policy

Ineffective teachers were more likely to leave voluntarily after New York City principals got tougher on awarding tenure, according to a working paper by Stanford researchers. After a new policy was adopted in 2009-10, few teachers were denied tenure but many more had their probationary period extended instead of receiving tenure.

Screen Shot 2014-06-15 at 6.49.34 PM.png
“Extended” teachers who were less effective — by principals’ judgments and value-added measures — were the most likely to leave, reports Ed Week‘s Stephen Sawchuck. They were replaced by stronger teachers, on average.

The district started supplying more data on teachers to principals, asking them to weigh performance observations, reviews of teachers’ lesson plans, and in limited instances “value-added” data based on test scores. And it began requiring principals to justify their decisions about whether to grant or deny tenure—particularly if it didn’t match up with the data. Principals could also extend the tenure decision for another year if they weren’t ready to make a final call.

The new policy improved the overall quality of the teaching force, the study concluded.

Teachers in schools with high concentrations of black and low-performing students were more likely to be “extended,” the study found. “We have a chicken-and-the-egg problem here,” said United Federation of Teachers spokesman Dick Riley. “Were people less likely to have probation extended because their kids are more successful, or is it the other way around?”

About Joanne


  1. Jerry Doctor says:

    One day I was talking to the district math supervisor along several other teachers when it was mentioned that I had not been given tenure yet. The supervisor, who had visited my class on numerous occasions, was shocked and announced that was going to change. Shortly thereafter I became tenured. I guess the principal wasn’t happy that I was the only teacher in the school that refused to join the NEA.

    I bring this up to show the importance of having the principal justify his decisions. Too often the administration’s rating of teachers has nothing to do with whether or not they can teach.

  2. Mike in Texas says:

    Another “working paper” released as Gospel.

    • …writeth the man who’s tried to pass off union editorials as credible and objective as well as grossly misrepresented court decisions.

    • You know, Mike, leaving aside your propensity to lie in the service of the greater truth that teachers are special people who deserve, well, pretty much anything you want, what’s your beef with this paper?

      Is it the acceptance that some teachers are bums who, if they can’t achieve the comforting embrace of tenure, will leave the profession?

      I guess I can see that as your objection to the paper and also why you’d want to be less then explicit about your reason. After all, while it’s perfectly obvious to you that teachers are special people who deserve special considerations there is a segment of the population who due to our stupidity, insanity or hatefulness don’t agree and we seem to be proliferating. We stubbornly insist that there are lousy teachers and the best solution for society to the problem they pose is to fire them.

      But that’s the solution for ordinary people who are incompetent and you’ve decided that a refusal to admit that there are incompetent teachers, and if there are that they shouldn’t be fired, represents a principled position on your part.

      It doesn’t.

      • Mike in Texas says:

        What I’ve decided is there are incompetent “reformers” out there. Actually incompetent is the wrong word, as I believe their actions are intentional. They want to privatize education, turn learning into rote practice of meaningless items, and hire minimum wage workers to do it. Google “Everyday Math” or “Open Court Reading” to see just how great that works out.

        My other beef is with “working papers” and how Joanne presents them here as Gospel. They are not work checked for accuracy or validity, so therefore anyone can write one without having to worry about pesky little things called facts.

        You may mock me but you know nothing of my experience in the public schools. I’ve seen the results of these harmful policies and what they do to children and schools.

        As for teachers being special, I don’t know if you have children of your own, but if you did, wouldn’t you want the person who spends 8 hours a day with them to be a special?

        • Roger Sweeny says:

          Mike, almost all scholarly articles begin as working papers. From what I have seen, there are two reasons. One: it allows others to criticize and make suggestions for improvement before the article is published in permanent form in a scholarly journal. Two, scholarly publication is a lengthy process. Working papers get information out quickly.

          Often, the published version is the same as the working paper. Many working papers are of high quality. On the other side of the coin, a large amount of shoddy research makes it past peer review and into publication.

          • Mark Roulo says:

            “…almost all scholarly articles begin as working papers. From what I have seen, there are two reasons.”
            The third reason is that the working papers can be freely distributed (even to folks who don’t have access to the journal). The version in the journal is under copyright.
            Note, too, that in this case the paper has been presented at a conference. It may never get peer reviewed (depends on how the specific discipline works), but might just appear in the conference proceedings and be cited from there.

        • Mark Roulo says:

          Mike in Texas: “They [reformers] want to privatize education, turn learning into rote practice of meaningless items, and hire minimum wage workers to do it. Google ‘Everyday Math’ …”
          The teams can get blurry. Everyday Math comes out of *academia* (specifically, the University of Chicago). It is one of 10 math programs considered exemplary or promising by the non-profit NCTM (National Council of Teachers of Mathematics … a group with 80,000 members … that was founded in 1920). But … the NCTM published a “Curriculum and Evaluation Standards for School Mathematics” document in 1989 that lots of reformers consider a disaster. In fact, a lot of “reform” proposals since 1989 have been direct responses to the 1989 standards. A lot of reformers want to get rid of Everyday Math (see here, for an example of the hostility towards it: “Is Everyday Math The Worst Math Program Ever?” http://www.forbes.com/sites/emilywillingham/2013/12/10/is-everyday-math-the-worst-math-program-ever/).
          So holding up Everyday Math as an example of what “reformers” want is not accurate. Clearly *some* folks want it. But lots of folks that want it are insiders, not those crazy loons on the political right (and don’t get University of Chicago’s econ department confused with the School Mathematics Project from the same university. The politics are quite different). A lot of the reform crowd would kill Everyday Mathematics with an axe if this was an option.

          • Mike in Texas says:


            I’ve used Everyday Math, its crap. Math scores at my school plummeted after its inception. The publisher demanded nothing else be used in our classrooms, so rooms were “cleansed” of other Math programs and resources.

            When I taught 4th Grade Math a common answer to our 5th grade teachers, when asked how the 4th graders’ math skills were, was “I’m teaching division to kids who can’t subtract”. In 5th grade they still couldn’t subtract.

            Everyday Math is just one of the examples foisted on schools by the “reform” crowd. Notice I put reform in quotation marks, b/c I feel “deformers” would be a better term. There may be a backlash against it now (like Common Core) but in the beginning it was a reformers wet dream. Scripted, and made teachers interchangeable (expendable). Never mind that it sucked if we could get rid of those lazy, incompetent teachers (Note for Allen: that was sarcasm).

            And please take note, publishers cannot always be counted on to be scrupulous. I’ve heard lots of stories, from people involved in the process, of jobs and kickbacks being offered to people in the selection process. In my district one person made the decision, against the recommendation of the teacher committee selected to help. Teachers were not given a preview of Everyday Math, nor even told of the decision until it had already been made.

            BTW, Everyday Math is published by McGraw-Hill. the very same people who brought schoolchildren OpenCourt reading and BIG time friends of the Bushes. Opencourt is another of the scripted programs “defomers” use to make highly paid experienced teachers expendable (to be replaced of course with minimum wage workers hired off the street).

          • Mark Roulo says:

            “I’ve used Everyday Math, its crap. … Everyday Math is just one of the examples foisted on schools by the ‘reform’ crowd.”
            I haven’t used it, but I have looked at it (along with many other textbooks … not all from the US). Yes, it is crap. My point is/was that the folks pushing it (even back in the 90s) are very different from the folks pushing for Singapore Math or Saxon Math. The teams are quite different. Even if they want math education to change. The 1989 NCTM document cares about social justice, for example. I doubt that the goal of Everyday Math was to produce lots of minimum wage folks.
            “And please take note, publishers cannot always be counted on to be scrupulous. I’ve heard lots of stories, from people involved in the process, of jobs and kickbacks being offered to people in the selection process.”
            Consider me stunned 🙂 Or not. Richard Feynman, again, has a nice writeup on textbook selection. It is sad. And from the 1960s. This stuff isn’t new. And since there hasn’t been an effective effort to clear it out, I suspect that most of the folks in charge are okay with the current status quo: http://www.textbookleague.org/103feyn.htm

        • You’ve also decided that if you don’t like the implications of the subject you’ll simply ignore it and substitute a topic more to your liking. Such as, for instance, the notion that teachers *aren’t* special people.

          Not quite in keeping with revealed truth according to Mike which is that teachers are owed special consideration due to the crushing burden of responsibility you carry and the stupendous generosity you daily display. Any question of competence is, as a result of those splendid qualities, to be disdained as the improper application of standards to those above such considerations. Certainly no mere non-teacher’s qualified to pass judgment on such as yourself.

          So if there’s some evidence that teachers are motivated by selfish considerations well, that’s not a fit subject for discussion and you’ll try your best to derail that discussion.

          Trouble is, outside this blog you can’t derail that discussion and you sure can’t counter the spreading realization that teachers, at least those represented by your viewpoint, are anything but the noble creatures you’d like to pretend you are. The excuse-making, the poor results, the blaming of just about anything you can think of along with the rejection of any responsibility for those poor results, the avarice are all contributing to the erosion of the public’s view of teachers and far from doing anything to stem that erosion you and teachers like you contribute to it.

          As to what any parent might want in a teacher, that’s immaterial in the system you’re defending. They/we take what we get and if we don’t like it we can beg or wheedle or apply or pray but we can’t make a decision and see it carried through. Turns out that’s not good enough for quite a few parents, more and more every day, and where that process will end won’t be to your liking.

  3. Mike in Texas says:
  4. Mark Roulo says:

    “My other beef is with ‘working papers’ and how Joanne presents them here as Gospel. They are not work checked for accuracy or validity…”
    This is going to come as a disappointment to you, Mike, but peer reviewed papers are usually not checked for accuracy or validity either(*). The peers check for (a) interestingness/relevance, and (b) obvious methodological mistakes.
    Often the peers doing the review do not verify that, for example, math mistakes were not made. Nor do they try to confirm the data. Or reproduce the experiment. And a lot of methodological mistakes get by the reviewers, also.
    And this isn’t just for social science research (though I think it is worse there). The bio/medical folks, for example, are starting to make public that lots of papers related to drugs (e.g. target site papers) and cancer cannot be reproduced by industry. In the field I’m most familiar with (a very tiny niche of computer science), the bulk of the papers leave out important details that make their conclusions bogus. It is common in many fields to omit the data that was used to draw a given conclusion (making it impossible to verify the analysis on that data).
    So peer review is nice … but I don’t think it is the holy grail we all want. And sometimes (in my computer science niche, for example) good papers can’t get past peer review because the truth is not as interesting as the lies.
    Nor is any of this new. Richard Feynman has a paper (not peer reviewed :-)) called Cargo Cult Science. Part of it discusses how to do good science and he describes an exchange he had with a grad student:

    Other kinds of errors are more characteristic of poor science. When I was at Cornell, I often talked to the people in the psychology department. One of the students told me she wanted to do an experiment that went something like this–it had been found by others that under certain circumstances, X, rats did something, A. She was curious as to whether, if she changed the circumstances to Y, they would still do A. So her proposal was to do the experiment under circumstances Y and see if they still did A.
    I explained to her that it was necessary first to repeat in her laboratory the experiment of the other person–to do it under condition X to see if she could also get result A, and then change to Y and see if A changed. Then she would know the the real difference was the thing she thought she had under control.
    She was very delighted with this new idea, and went to her professor. And his reply was, no, you cannot do that, because the experiment has already been done and you would be wasting time. This was in about 1947 or so, and it seems to have been the general policy then to not try to repeat psychological experiments, but only to change the conditions and see what happened.

    So … peer review can be nice, but it often provides very little quality control. And sometimes actively filters out good experiments in favor of bogus ones 🙁
    (*) Two exceptions. Proofs in math journals tend to be checked, but not always before publication. The chemistry journal “Organic Synthesis” doesn’t publish papers until the editorial board can reproduce the result.

    • Mike in Texas says:


      But a working paper can be released as an article, with lazy reporters, or people pushing their reform ideas, as the truth. At least if an article is submitted and published in a journal it can be judged by professionals in the field.

      • Mark Roulo says:

        “At least if an article is submitted and published in a journal it can be judged by professionals in the field.”
        Working papers can get feedback, too (this is often one of the goals when releasing a working paper). And many papers *can* be judged by the folks in the field, but most aren’t … there are just too many papers to deal with so most just sit there, unused, unreferenced, ignored.
        In the hard sciences, there isn’t much seen wrong with circulating a working paper. And Joanne won’t be able to link to a published version …
        I’d be more concerned with the generally low quality of research in general (and, given the topic of this blog, of research in this field in specific) than I would with working papers.