Lessons From DISD’s New Merit Pay System: Bad Teachers Go Bye-Bye

Great teachers are finally making as much money as bad teachers

edna-krabappel

Nearly a month ago, the Dallas ISD board was given a briefing on the impact of the district’s revolutionary Teacher Excellence Initiative (TEI), the evaluation and scoring system that determines a teacher’s effectiveness and pay. I didn’t write about the briefing then because I was knee-deep in covering the DISD bond. Now that the bond passed, I want to tell you the important takeaways from the TEI data, because the findings and results are pretty amazing.

That’s not to say that there aren’t ongoing bugs and glitches in the system – important ones that need to be addressed. (I wrote about some concerns last year, and these were largely addressed.) I’ll write about those soon. First I want to talk about the big lessons from the TEI data that should change the way we think about public education and could change the way districts operate.

If you want to follow along, you can pull up the background materials provided to the board and the slideshow of highlights that was presented. If you want to see the presentation itself, go here and click on the “meetings” tab, then click the 2015 tab on the left. Then hit the Oct. 15 briefing agenda and click on the video presentation marked TEI. I can’t link directly to it.

First, some background. We’ve talked a lot about TEI, but as a reminder, it’s a system that evaluates teachers based on student outcomes, peer evaluation evaluations from certified administrators, and student surveys. (I’m simplifying this a lot.) Student outcomes count a lot, peer evaluation counts a little less Evaluations count the most, student outcomes a little less, and student surveys a little less than that. (Except when that’s not true; it’s calibrated differently by grade/subject. Also, as was pointed out in the comments, I’m ignoring something called SLOs, because those are complicated and are a small percentage of most evaluations. The link two graphs above is a deep dive into all this if you care.)

This is a radical departure from the way teacher pay has been determined at DISD since 1894, when Superintendent J.L. Long first instituted a seniority pay scale. So for more than a century, the only metric used when determining pay for teachers has been a calendar. (Well, that and a higher degree; more on that later.) Under TEI, experience counts in certain ways — first- and second-year teachers are capped as to how high an evaluation category they can be placed, for example — but much more data is considered. The hope was that a more equitable system would get rid of bad teachers, as well as attract and retain good-to-great teachers.

What was the result of its implementation? The first important slide in the presentation is this one:

Screen Shot 2015-11-12 at 2.46.11 PM

It looks at the 1500 or so (out of 10,000-plus) teachers who left the district, and shows how the departed graded out under TEI. (Green shows percentage that left in each category, blue shows percentage that stayed in a teaching role, and red shows those who stayed but are no longer in a teaching role, meaning they are now staff or administrators or whatnot.)

Ignore the difference between “evaluation” and “effectiveness” for now. No reason to get glassy-eyed over a small distinction that we’ll discuss later. The point is this: The majority of teachers who left the district were those who were either not graded or were graded as “unsatisfactory” or “progressing.” Those who graded out as “proficient” or “exemplary” stayed in much greater percentages.

Do you know how important this is? If a merit-pay system is identifying and driving out your worst teachers — and therefore is identifying and retaining your best teachers — that is a recipe for a dramatic turnaround in any public school system, let alone a large urban district. (That is if you believe, as studies have consistently shown, that the greatest influencer of a poverty-stricken student is a great teacher, and that the greatest detriment to the outcomes of that same student is a bad teacher.)

I know that the headline “Great teachers remain in DISD as bad teachers flee” isn’t sexy, but it’s important. You should know this is happening.

What does this mean for the pocketbooks of good teachers? Aside from individual and unique flaws in the system — which, as I said, we’ll look at soon — it means great things. Look at this chart, which is farther down in the slide presentation.

Screen Shot 2015-11-12 at 3.07.58 PM

See what it’s saying? Basically, the teachers who scored the poorest under TEI, while still making too much money, didn’t see a collective raise. Meanwhile, it shows that the old calendar-pay system was not rewarding the district’s best teachers, and that now the best teachers are seeing raises commensurate with their excellence. When you realize that every group of teachers in DISD, including its best teachers, were earning less as a group than the WORST teachers in the district before TEI, you see how broken the old seniority system really was.

I think that’s the most important takeaway from the first-year TEI data: bad teachers are leaving, good teachers are staying, and DISD is now paying good teachers more money. Tomorrow, we’ll look at some other interesting data points and what they say about the district’s continued efforts to transform itself.

Newsletter

Get a weekly recap in your inbox every Sunday of our best stories from the week plus a primer for the days ahead.

Find It

Search our directories for...

Dining

Dining

Bars

Bars

Events

Events

Attractions

Attractions

View All

View All

Comments

  • Jeffrey Weiss

    How do you know that the same pattern of retention would not have taken place under the old system? Not a rhetorical question. Under any evaluation system, there’s churn. A lot of it — though hardly all — is relatively new teachers who decide it’s not for them. Likely those are teachers who grade badly under whatever reasonable system is used. Post hoc ergo procter hoc? Or do you have reason to believe — evidence — that the system actually changed what the pattern would have been?

    • EricCeleste

      The difference is you had no idea before. Now you know.

      • Priscilla Clark

        You’re ;wrong. You don’t know. One year is one year. It’s not a trend, or a pattern. It’s just a snapshot.

        • ParleyPPratt

          That’s perfect, actually. Let’s let it run for 5-6 more years and see what happens. Then we can tell if it’s more than a snapshot.

          • Priscilla Clark

            Ok, but if you run it, you have to run it the same each year…. OOOPS… can’t do that! They’ve already made changes this year. So now we’d compare apples to oranges, not apples to apples. I guess if you want, we could keep it the way it is designed TODAY, 11-12-15.. and run it for 5 years…..I think it’s a bad idea…but forget about me, EVERYONE says it needs adjustments (BOT, Superintendent, Todd!, etc.) So maybe we need to wait til we work out the kinks, and then run it for 5 or 6 years….that’ll put us at what? 2025?

    • Amy S

      Based on the mathematics of averages, and the lower wages made by newer teachers, I would think the new teachers would bring the average salary way down if they made up any significant portion of the “unsatisfactory” column.

      • Amy S

        It is what is called a “red flag” seeing the highest average salary in the Unsatisfactory column f/y 14/15. You have to evaluate how a teacher would end up with the highest wages under the old system. Which would be (mainly) due to job longevity.

  • EricCeleste

    Correction: Per usual, I remembered something backwards: the evaluations usually count 50 percent, the outcomes 35, and the student info 15 percent. I mixed up the first two. Also someone points out evals done by a certified evaluator, not necessarily a peer. So noted.

    • Priscilla Clark

      You’re wrong for so many reasons. First of all, it depends if you’re talking about an “A” teacher, a “B” teacher, a “C” teacher, or a “D” teacher. So right away you’re off target when you talk about those percentages. Second, the system changed so much throughout the year, you can’t call it valid. Third, asking students if the classmates behave and then grading a teacher on a student’s perception of his/her peers behavior is absolutely ridiculous. Perhaps you could write about that. I could go on and on,and write a short story about all that’s wrong with TEI, but let me ask you this one question you and other supporters have refused and/or failed to answer. Where has a system like this ever worked to effectively attract or retain teachers?

      • EricCeleste

        You are wrong.

        • Priscilla Clark

          About what?

          • EricCeleste

            Wrongness.

          • Priscilla Clark

            More proof you’re wrong… where’s the 5% SLO in your statistics? Where’s the School Rating in your statistics. It’s sad to be misinformed. It’s worse when you know the truth and spout lies.

            Cowardice, I guess.

          • EricCeleste

            You are right that I am a coward.

          • Tim

            I think he means, I’m not interested in wasting time on lengthy explanation to people whose minds are made up that they are against something.

      • Johnny

        Im so tired of the “where has it worked” argument. Its NEW! According to your logic, nothing new should ever be attempted. Just keep paying the same bad teachers the most money, forever, because anything else would be to try something that “hasnt ever worked before”.

    • miss adieux

      Evals are done by a “certified evaluator…” oh, how august. Certified by DISD TEI people! Hahaha for the third time here tonight. The more I read, the richer this gets.

    • miss adieux

      Ugh @ the endless use of “outcomes” in place of more accurate (and less-holistic-sounding) “test scores.” (35%)
      It does not make people (you, Morath, Todd Williams) sound more convincing when you use marketing ‘n’ messaging buzzwords. Instead, it implies a disingenuousness, hints at a disrespect towards your audience by assuming they lack the intellect to see beyond the language of hype.

      Evaluations are extremely subjective, and where they are not, rely upon the locally-infamous unmentionable (at least, none of you local deform fans will touch it – and we’ve been asking for 3.5 years) known as MRS DOLLO. (50%)

      Have you read the student survey questions? Nancy Bingham says they are creepy. None of the other trustees disagreed. (15% of TEI rating.)

  • Murf the Surf

    A fine example of what Freud calls Kettle Logic. All one has to do is look closely at your reasoning here to see that you are actually saying nothing. What are the metrics being used to evaluate teachers? What counts and what doesn’t? Does the 50% from evaluation mean that teachers who don’t tow a particular pedagogical line or who rub principles the wrong way get marked down? This is so open to corruption and cronyism, the very things you like to claim the new system is correcting. And, the 35% outcomes is based on what? The teaching to the test testing mania that measures something, but I would like to know what? You don’t mention that at all. And the fact that kids are allowed to have a say in their teacher’s salaries is flat out awful. In short, you throw around a lot of numbers,but they are essentially meaningless. I fear we are creating a cadre of teachers who are basically party loyalists, who follow the party line and who live in fear of offending their students and their principles. No educator in his or her right mind could possibly think this is a good idea. I am not at all convinced that the majority of the teachers who are leaving are bad teachers. Maybe they are just bad “test takers!” I would be very interested in knowing how coaches in DISD are fairing under this system. I am willing to bet that a winning coach who is a poor teacher is more likely to get a higher rating from a principal than a good teacher who rubs the principal the wrong way.

    • Todd W.

      The perfect should not be the enemy of the good. Every industry in America has an evaluation system. None are perfect. All try to continually improve it. But it is incomprehensible that because there are concerns over certain aspects of TEI system, rather than work hard to improve it, there are those who advocate we return to the old method where the only input was a calendar and thousands of great teachers over the years weren’t paid fairly (and many quit) because they worked in a system that valued longevity over effectiveness. That’s not fair to kids.

      Pay teachers what they are worth, when they are worth it. If we want to build community/taxpayer will to pay our better teachers more (and we absolutely should) prove to them we have a system that actually determines who are best teachers are. Anyone who comes to a discussion without data is just another person with an opinion….and you can’t manage to success with opinions.

    • EricCeleste

      You are wrong.

      • Murf the Surf

        Your minimalist response has the same analytical heft as your statistic laden original posting.

  • Eric Foster

    Now if such a plan can be applied to Doctors, Lawyers, Police, and Politicians, and get rid of their bad apples. What’s good the for goose!

  • Murf the Surf

    There are way too many inadequate teachers, I have no doubt. But, I am not at all convinced from Celeste’s claims that is necessarily the bad teachers who are leaving. He certainly provides no evidence to support his claim. You have to look much further into the metrics being used to assess teachers–and he does none of that. I am all for making it much more difficult to become a teacher, as I am for increasing teacher salaries. we need more rigor in our classrooms, but I am not sure that the state and city are providing the kind of rigor that matters. I know from experience that assessment regimes from other industries do not necessarily translate to the field of education. This is a mistake that I think well meaning people and deeply engaged people like yourself, Todd, often make. Teachers need to be assessed. The best way to do that is to have other well, respected, senior teachers with proven track records assess them. Not administrators–and certainly not students!

    • miss adieux

      On this:
      “…Celeste’s claims that is necessarily the bad teachers who are leaving. He certainly provides no evidence to support his claim. You have to look much further into the metrics being used to assess teachers–and he does none of that.”

      I offer this from another comment:
      [Todd Williams] “Anyone who comes to a discussion without data is just another person with an opinion….and you can’t manage to success with opinions.”
      (not sure exactly what the wording at the end – “manage to success” – is supposed to communicate, but think we get the basic idea.)

      • Murf the Surf

        A problem is that well meaning “civilians” like Mr. Williams, who to my knowledge has no extended meaningful experience in the classroom, if any at all, has a very limited sense of what constitutes “data.” I would not, for example, call the graphs that Celeste has produced, “data.” It’s a collection of statistics put into graph form that is meant to suggest meaningful information, but which conveys nothing more than a manipulation of jerry-rigged statistics that upon even a cursory inspection does nothing to support his claim that bad teachers are leaving the district. i would imagine in Mr. Williams’ world of financial speculation, there is utility in raw numbers. I have seen little evidence that those techniques transfer well to to education. We can tell a salesman is making money for his employers in terms of the raw dollars they bring to the firm. It doesn’t work that way in education.

        • Miss Eww

          You wouldnt call the graphs Celeste produced data, but you cant stop praising Bill Bentzen enough for his ridiculous child-like charts and analysis. Thats rich.

  • Priscilla Clark

    Dear Todd, I have been following this, and you need to be aware teachers are not advocating only for a return to the old system. MANY are advocating for a change, but NOT THIS change. As you have pointed out, there are OTHER SYSTEMS out there. Why are you and other so stuck on THIS SYSTEM, when there are other, better, PROVEN systems out there?

    • jfpo

      You sound like the new evaluations aren’t going so well for you. Also, SELECTIVE capitalization is the scourge of commenting these days and needs to STOP! We can read.

      • Priscilla Clark

        Actually, my evaluation was wonderful. I have no complaints. Clearly you can’t read, because others and I keep pointing out the same factual errors and misunderstandings, and yet writers and supporters don’t seem to notice!

  • Priscilla Clark

    Dear Johnny, I regret to inform you that it is NOT NEW! It’s another of several variations made to a Pay for Performance system. Pay for Performance has been tried in school districts and businesses. It has NOT been proven successful. There are other options.

    I am tired of the TEI supporters saying it’s either TEI or the “old seniority” system. There are others out there. In fact, the one Todd! most recently referred to is a system that Dallas actually implemented not to long ago… and guess what… it worked. But then it was stopped. Why does no one want to talk about that? Why not bring it back? Data shows it worked in Dallas and in Washiington DC. Why is that (one of many better options) never discussed?

    • miss adieux

      Priscilla is diplomatically omitting to point out that Todd! even deleted a comment and disappeared when asked further about this on DMN blog/comments. He seemed to have dropped Washington DC as an example, then vanished, but I never saw the now-deleted comment (am just noting all other contextual commentary.)

  • miss adieux

    “the district’s revolutionary Teacher Excellence Initiative (TEI), the evaluation and scoring system that determines a teacher’s effectiveness and pay.” – Hahaha!

    Revolutionary? How? This has been used all across the Rhee-formers’ edu-scape. Do you not bother to read what happens in public education in other American cities, or are you just counting on your readership not to do so?

    “This is a radical departure from the way teacher pay has been determined at DISD since 1894, when Superintendent J.L. Long first instituted a seniority pay scale. So for more than a century, the only metric used when determining pay for teachers has been a calendar.

    Hahaha again! So… did you write the argument and talking points for the TEI lovers, like your pal Mike Morath at the TEI-related board briefing, like your pal Todd! Williams in all his DMN commentary? Or, as is more likely, is their specific narrative and its corresponding list of talking points your main source of information? The ideas and even wording are now practically identical. (Evidence: comments made at the October Board briefing; Todd Williams’ DMN comment transcripts for the last several months.)

    “Bad Teachers Go Bye Bye” – your headline is, predictably, the offensive and inaccurate rhetoric of Todd!, Morath, and the rest of the Rhee-formers, worded in your signature nanny-nanny-boo-boo tone.
    Yet again, you parrot the standard deformer’s view of public education, subscribing to ideology that has already been a proven failure. In doing this, you exhibit a reckless disregard for the lives and futures of children within it, while paradoxically trying to claim that this is all “for kids!”

  • miss adieux

    I feel the same way when I read Celeste’s writing on education.

  • miss adieux

    Just making sure I got that right… y o u are tired of particular arguments? Then “it’s new!” (or “its NEW” as you try to argue.)
    Then you’re bringing it on home with the “the same bad teachers” and “you hate new things” arguments… Hilarious!

  • miss adieux

    If the perfect should not be the enemy of the good, why can’t DISD openly and accurately report the amount of vacancies and long-term substitute teachers at each campus? If we want to build community and taxpayer support and engagement, we need to offer this BASIC level of transparency.

    One could argue that this should have been remedied before devising DISD’s own evaluation system.

  • Murf the Surf

    Not sure that is what is going on. For one thing, if anyone’s mind around here is made up, it the mind of one Eric Celeste, blogger for hire. I think a better explanation is that Celeste simply can’t engage in an exchange with people who challenge him on his conclusions and his methods of analysis. He’s such a bully and a wiseacre! And like many bullies, he turns tail when challenged. Fortunately, no one reads him except his small Hallelujah Chorus and those of us who like to mess with him for sport.

  • Michael Dryden

    Todd, I noticed the average age of teachers dropping below 22 in the mid 2000’s to find teachers coming from dubious colleges along U.S. 80. On average I saw 1 in 5 elementary teachers not solve an elementary math problem. I cannot tell you how many times I saw conceptual errors in science classes. However, I also saw some of the best teachers and I taught in 5 different countries. The problem with TEI is that it seems to be removing good teachers which was rarely done before. We need to reward the best, fire the worst but TEI is not even close. Nobody expects perfection but random classification will destroy public education. BTW looking at successful 4 year college graduates of DISD the strongest predictor of success after controlling for prior achievement was teacher experience and education. The goal is to get rid of the incompetent teachers with weak or unrelated degrees and years of poor teaching. Again TEI is not doing that.

    • Todd Williams

      Mike, I am 100% supportive of tremendously increasing the rigor of what it takes to become a teacher. I am willing to make a bet that under TEI, DISD on a net basis (good teachers lost vs. ineffective teachers more quickly removed) is better off with the current system than the old one, particularly if the district moves more and more toward empowering their effective teachers and moving away from substantial observations and scripting for this select group. You had to do this in Year 1. But ultimately, once someone proves that they are effective, empower them and focus your development/coaching efforts on those that need it.

  • Michael Dryden

    As someone who as actually analyzed the data in Dallas I can say the TEI has a very flawed measurement model. The value added model, CEI, was never sensitive to 9 merit scales and now the district eliminated school level analysis. The Colorado Median Percentile Growth model ignores factors outside the school and equates prior achievement into only 4 buckets. The expectation is that a TAG student has the same expectation as a student at Roosevelt even if the Roosevelt student is 25 percentiles lower. The student survey does not contain standard survey writing practices and historically many students just fill down one side without carefully reading the survey. The spot observations are dependent on time of year with fall ratings lower than spring ratings to feign growth. The district does not have the capacity to write, pilot and analyze 255 teacher designed tests every year, ACP, and of the few ACP tests I did analyze most had too many poor items to rate teachers for merit pay. The TEI is probably random and is eliminating both good and bad teachers while rewarding both good and bad teachers. You might as well roll a die and label a teacher.

    • Michael Dryden

      Solutions are needed. Consider reexamining the teacher evaluation system DISD was about to implement after years of community input. Miles dropped it as one of his first acts as superintendent. Look at teacher college transcripts by subject taught. A colleague did this and found someone who failed freshman college English three times teaching high school English. There are many other stories of unqualified teachers but also many stories of the best teachers. Don’t expand VAM/CEI beyond its sensitivity. Reward those in the top 20% for three years and flag those in the bottom 20% three years in a row. Be careful that prior grade cheating is not occurring. Switch the effort into so many ACP tests from summative EOC assessments to diagnostic assessments throughout the semester and hold teachers accountable for understanding impediments to child learning. Develop peer teacher observations but protect the teachers if they demand a poor teacher leave the system. Finally, reinstate some form of curriculum based on a recognized philosophy of how children learn with available resources and materials, and formative/diagnostic assessments. .

      • Todd Williams

        Mike, none of us would ever tolerate our child having a teacher in the bottom 20% three years in a row. We should not advocate for any system which allows a bottom 20% performer to be in front of other people’s children for three years in a row.

        • Michael Dryden

          Todd, that is true if the evaluation system was accurate but even the old CEI system was near random and most below the 20th percentile one year.were above the 50th the next year. That’s my problem. Everyone assumes accuracy that is just not there.

    • Todd Williams

      Mike, all who comment on here on shortfalls of TEI will have much more credibility if they first admit the prior system was absolutely, unequivocally terrible for the kids of DISD. As the highest poverty district in the region, we have to pay our best teachers more and more quickly else the district will never attract enough great teachers in its competition with other districts if all have the same seniority based salary scale. You and many others are on record that there aren’t enough great teachers in the system to fill every seat…the district absolutely has to have a wage based on effectiveness and not seniority. Those are not mutually exclusive and for many are highly aligned.

      • Michael Dryden

        The old system had nearly everyone exemplary but good principals got rid of bad teachers. Unfortunately the district merely reassigned them to schools that deperately needed good teachers. Now the TEI non-renews good and bad teachers. Are there proportionally more bad than good teachers non-renewed? Probably, but the lose of good teachers is almost irreversible. This is so easy to test. Just run the same algorithms for student growth in 2013, 2014, and 2016 and check the stability of the student growth portion of TEI. I did it for the old CEI which was much more accurate than the current algorithms and it was near random.
        http://dfpe.org/documents/CEI_Math_Simulation.zip
        (Windows based Excel was used)

        • Todd Williams

          Mike, if the old system relied on the subjective judgment of those principals considered “good”, why was that any fairer of a system? I do look forward to seeing the analysis you suggest.

          • Priscilla Clark

            Dear Todd,
            You are not the only one who wants to pay good teachers more, and more quickly. Teachers have for years wanted a better evaluation system. In the past you have espoused the Washington DC model. We had a similar model here in Dallas a few years ago. Why do you insist on drawing a line between TEI and PDAS with nothing in the middle? I keep asking why no one has bothered to consider other options. I wonder why no one has bothered to answer. All I get for responses is either TEI or PDAS. Why no conversation on the other options. Why not something more like your favored Washington DC model?

            In regards to principal training, under the current system, the Principal’s evaluation has a direct impact on the Assistant Principal. So if last year a campus had a great Principal and great Assistant Principal, all was well. But if the Principal leaves and the campus gets a terrible Principal, it impacts the Assistant Principal. I don’t understand that. Dallas has so many evaluation systems where people on judged on things they have no control over. You seem OK with that. I can’t imagine why! TEI has so many flaws, it will take years to work out all the problems.Some times it’s better to proceed slowly than to rush in to something.

      • Priscilla Clark

        Dear Todd,
        I respectfully must point out you are not understanding the comments of the TEI opponents. They have never said the previous system was a great system. They have never said it was a fairer system. In fact, most have commented that it was equally as unfair as the current system. The big difference though, is that one unfair system, the current one, penalizes teachers, while the previous unfair system did not. The two systems are highly subjective, and can be manipulated very easily for a principal’s “pet” or a principal’s “peeve”. However, the current system will financially penalize a teacher, whereas the previous one would not.
        As I and others have mentioned before, there are other options out there. Why they haven’t been explored or pursued is quite confounding.

        • Todd Williams

          I guess my immediate reaction is twofold. One, almost every industry in America involves subjectivity of a manager in personnel evaluations…just as it did in the hiring decision to begin with. Second, an effective teacher doing the same work and being just as effective as a teacher with twice the years (yet being paid less under a lockstep compensation system) is clearly being penalized….as is the child that would benefit from their teaching if not for their decision to leave the profession due to working in an industry which makes little attempt to recognize their individual mastery of their craft at an earlier age than others.

          Teachers are absolutely the solution to our educational challenges. I just wish I could better understand why I’m the one that wants to create a system that pays the great ones more money and quicker and is politically realistic to understand that making years of experience a huge formulaic component will be a non starter for the bulk of the tax paying populace.

          • Michael Dryden

            Todd and Pricilla, I only go with the data and the so far the TEI in Harrison 2 and DISD is a disaster and much worse than the problematic old system in DISD. Besides the random nature of the CEI VAM model, I believe the new CEI dropped the school level analysis thus treating TAG and Roosevelt the same. The main criteria for teachers is not CEI now but either absolute scores or a gain model called the Colorado Student Median Percentile Growth Model. The CSMPM does not adjust for any factors outside the school and uses only four wide achievement bands. It makes teachers at high performing schools look good in large part due to parental influence, not teacher effectiveness. Look at Holly Hacker’s recent list at the DMN. The magnets are at the top. That might be correct but I would like to make sure the TEI does not correlate to emphasis of education in the home, which I suspect does.
            The merit pay system in Harrison 2, according to the Colorado state data, made the teacher turnover rate explode, lowered the years of teacher experience, discipline issues skyrocketed, and the AP program collapsed. All that would only occur if the more competent teachers left. Unfortunately, DISD is starting to mirror this trend.
            The district lost an appeal and the AG made them release the item level data on ACP. The district now creates 255 tests per year without the time or expertise to write and analyze properly. I was sent that file and analyzed a few tests. Each had too many errors to justify use for merit pay. One ACP, 5th grade ELA in fall 2014, was so bad 90 percent of the items had 90 percent of the students answering correctly and the other 10% of the items were so poorly written they confused the brightest kids. How can teachers be judged on a test almost everyone scores high and the remaining items are invalid?
            You cannot use the argument that no evaluation system is perfect therefore the best we can do is tweak the TEI. Advocating higher pay for better teachers is good but if the identification process is so far off then the better teachers will be falsely rated ineffective and the system suffers. The TEI is so flawed it needs to be scrapped. The data overwhelmingly suggests it to be arbitrary, capricious, biased, and naive both in measurement design and learning research. It is indefensible.

        • Todd Williams

          I would also reiterate that one of the efforts that I am highly focused on is improving principal training because I believe an effective campus leader is critical to retaining great teachers. People don’t quit school systems; they quit their leaders if not treated and supported well.