It seems to me that the whole point of every school reform measure that’s come along within the past 25 years can be distilled into these six words: A Good Teacher in Every Classroom.
It’s a simple mandate, but with several implicit complications. First of all, what do we mean by “a good teacher?” And who gets to decide which teachers are good? And once we decide who’s good and who isn’t, how do we make sure each classroom gets a good one?
Underlying these complications is a fragile balance between job security and accountability. Teachers, like me, tend to lean a little more toward job security at the expense of accountability. We’d rather tolerate the fact that there might be some horrible teachers if it means that our own jobs are relatively secure. Parents, like me, lean more toward accountability; we’re willing to tolerate a few unfairly fired teachers if it means that our own children have zero chance of spending the year with a bad teacher. (That’s right, I’m schizophrenic.)
So what is a good teacher? And who does the labeling? And what do we do after the labeling? These were some of the questions addressed in a recent discussion hosted by Education Sector, featuring three panelists: Scott Thompson from IMPACT, the new teacher evaluation system for the Washington, D.C., public schools; Brad Jupp, senior program adviser for teacher quality initiatives at the U.S. Department of Education; and Jen Mulhern, from the New Teacher Project, who worked with New Haven on their new evaluation system. Ed Sector also wanted to bring the voice of teachers into the discussion, so they included a separate panel with four teacher bloggers, one of whom was me.
The discussion began with each of the three panelists describing their work and what they’re doing to connect teacher evaluation with professional development. The second half featured a question and answer period.
I listened carefully and even got to ask a few questions. As the only panelist from the West Coast, I wasn’t actually on-site (the event took place in Washington DC) but I was connected via telephone and email. I also watched the whole thing on-line several days later, and frankly, I was impressed with what I heard.
I heard three educational experts, each of whom is in a position that matters, who seem to “get it.” They understand the complexity of teaching; they know that there’s more to teaching than the production of high test scores. But they also understand that accountability entails the use of data, and that the data has to come from the students we teach. Brad Jupp, who was the mind behind Denver’s complicated Pro-Comp system, talked about the Duncan Administration’s Blueprint for Education and Race to the Top Initiative, both of which are also incredibly complicated, as they should be. If the last administration acted like they were doing brain surgery with a backhoe, this administration looks like they’re at least using the right instruments, which is refreshing. He also made the observation that a better evaluation system and more relevant professional development means more money, which means new money, not money taken from other programs. The other two panelists described the evaluation and PD programs in their respective districts, both of which sounded a lot better than what my district offers. DC has independent, master teachers observing and evaluating every teacher five times a year. Five times! And then they sit down with you and share what they liked and what you might want to work on. If there exists a better job-embedded professional development model out there, I haven’t heard about it.
They use teaching standards and professional expectations that make sense to everyone. They also use tests scores to evaluate teachers, but they seem to do it sensibly. They understand that context matters; test scores that would be abysmal for one school might be phenomenal for another, so the teachers and principals sit down ahead of time and agree on which data will be used for evaluative purposes.
In short, these people seem to “get it.”
But now I’m starting to wonder if we get it. As teachers, are we so freaked out about job security that we refuse to be accountable for student learning? Is it OK for us to ignore the fact that the teacher down the hall isn’t teaching anybody anything, because if we point it out, they might look more closely at us? Are we so worried about the use of test score data that we publicly refuse to even acknowledge it, even though we privately celebrate the good scores and anguish over the bad? And do we actually get everything we can from professional development opportunities, even when the presenter is a little boring, the room a little stuffy and our energy level a little low?
Do we sometimes forget that the whole point of this is student learning, and that the best way to ensure that students are learning is to get a good teacher in every classroom?
I wonder.
Tom, I’m glad you participated in this panel and were able to post some good take-aways that have extended the conversation. One thing I would add is that I’d be glad to embrace more use of data if we could stop equating “data” with “standardized test results.” The reason I’m so dead-set against that idea is that, as a high school English teacher, there is really nothing of any instructional or evaluative use in that data. Maybe in other states or other systems it could be slightly better. However, if we can use create other data, using common assessments and a broader variety of insights into student learning, I’d be interested and willing to get on board. A broader conception of what to look at in evaluation would also suit the goal of improving evaluation for EVERY teachers – not just those that have a standardized test that can ostensibly be tied to their work.
It shouldn’t take three years to remove an ineffective teacher, and under peer review it doesn’t. With the Toledo Plan which was the grandpa of these programs, mentor teachers are taught to not delay the tough decisions. But peer review is the opposite of IMPACT. IMPACT may use the same processes, but it is solely in the hands of management. Just as I want my union to bite the bullet and remove bad teachers expeditiously, I’d want my union to aggressively fight for teachers who are misued by IMPACT. I don’t WANT to be defending ineffective teachers, but the union is defending the process, just like the defense attorney defending a guilty client is defending our Constituional Democracy. For instance, Michelle Rhee was in the news today twice for expressing her disrespect for the law, once ignoring her lawyer’s advice on the Hatch Act, and expressing her disrespect for the property rights of teachers with Effective evaluations. As a citizen she can disagree with the law, but as an educational leader she must respect the law and demand that her people respect the law. You can’t really respect IMPACT when it is a part of an administrative system that has demonstarted repeatedly that they will defy courts and precedents.
I’ll offer some heresy. I disagree with the sentence “The bottom line is that teachers are the center of high quality education.” Education should be a team effort, starting with prenatal and early education. Especially in hardcore schools like mine, we need community schools that gets kids out of the building into the community and the diverse community into the buildings. Rather than trying to reverse engineer superstar teachers, we need orderly and respectful schools where being a good teacher is enough to be an effective teacher. We need great pd and we need to aspire towards greatness, but we also need more team players and fewer one man teams.
In urban secondary schools its a given that teachers must earn “Respect.” Firstly, that sends a terrible message to kids. Respect should be bestowed on everyone. I’m an animated, emotional teacher who puts relationships at the top. As we used to say, I teach the student. But why exclude good teachers who feel more comfortable teaching the subject? Why provide a litmus test that prospective teachers must pass. In fact, one of the most disappointing things at Ed Sector was Jupp’s throway that its a given that systems will want to only hire and retain the type of teacher they want. The risks of an educational monculture are as bad as the risks of an environmental monoculture.
I disagree, Jennifer, with the cynical interpretation that this is about removing expensive, experienced teachers for newer, cheaper teachers. Tom has more closely captured the spirit of this endeavor– get a good teacher in front of every kid.
Terminating teachers, however, is just one part of how we can use this information. Systematic teacher evaluation includes targeted professional development and the use of assessments to improve teachers. We have to come up with a definition of a good teacher and we have to start measuring folks up to that ideal. Peer review using observations with high-quality rubrics and assessment data should be combined to provide comprehensive feedback and intervention to teachers first. The problem is, once we have all this great data on how well (or not so well) folks are doing, we cannot ignore the drastic differences that we observe.
We have to make a choice—use data to drive systematic problem solving that identifies and attempts to address shortcomings or worry about the margins. While most of the information we collect cannot distinguish between an average teacher and a slightly better teacher, we can start to identify the lowest 5-10% of teachers with relatively high certainty. Even if we were unable to distinguish between folks below that line and above the line half the time, nearly every teacher we dismiss would be truly in the bottom 15 or so percent. It’s a question of false positives or false negatives, do we worry about identifying some “ok” teachers as bad or identifying some “bad” teachers as “ok”. I am betting this is partly because I am not an educator, but I would take a few false positives every time because I would rather be certain I am removing nearly every teacher at the bottom of the barrel at the expense of a few percent that are slightly above bottom than have kids in classrooms with our worst teachers.
The bottom line is that teachers are the center of high quality education and we know that we have many schools and teachers that are failing their students. Due process is not an answer—in many states nothing short of criminal neglect can result in a dismissal through due process after hundreds of thousands of dollars and sometimes more than a year of work to remove a single teacher. .
I like bits and pieces of what lots of folks are doing. Rhode Island, for example, has proposed that teachers are removed from their current assignment after 3 years of unsatisfactory performance. Licensure is only lost after 5 years of low performance. This compromise was forged in collaboration of union leaders as a way to recognize that sometimes “bad” teachers are good teachers in the wrong environment for them to personally succeed. Cincinnati uses peer and administrator reviews with a robust rubric, not unlike IMPACT, which seems to provide useful feedback on actionable steps teachers can take to improve and encourages behaviors that can be linked to increased value-added scores in their district. Teacher evaluation is not an “if” anymore, we are at the stage of “how?” and if the teachers and unions reject the process on principle they are going to find themselves left behind. The national AFT has recognized this reality and is attempting to get ahead of top-down developed systems and provide their own guidelines and systems. I fear the NEA is making a losing wager.
The argument really centers here on how to use data. Remember “assessment informs instruction?” Nothing wrong with collecting lots of information–but many of the data analysis models begin and end with the (incorrect) premise that a teacher is either “good” or “bad.”
We need to use data to improve instruction, which in turn will improve student learning. It’s impossible to operate in a climate where fear and monitoring reign, however, so teachers do the logical, protectionist thing: they narrow their teaching right down to what’s on the test.
John is correct–this is about sharing power. And in the very end, it’s about “saving” money, by getting rid of expensive employees and replacing them with (new, enthusiastic–and cheaper) employees.
I also think you can carry data analysis to ridiculous extremes–although most of the people over here think that the more numbers you have, the better: http://blog.coreknowledge.org/2010/06/28/pitchers-teachers-and-data/ Tom should appreciate this.
I’m actually an “old union guy” myself, and I’ve long opposed the use of test data in teacher evaluations. But when reasonable people talk about the reasonable use of data, it seems reasonable. Especially when the only reason to not to use data stems from job insecurity. The processes described during the panel discussion seem plausible to me; they were agreed to beforehand by the teacher and the principal and they used test data for only part of the evaluation.
Furthermore, it seems that the only stakeholders right now in education reform who oppose the use of data in any circumstances are teachers. This worries me. Yes, the data is inherently flawed and shouldn’t be used exclusively or out of context, but does that mean it shouldn’t be used at all?
And my crack about tolerating a few fired teachers was just that; a crack. I meant to express the feeling, shared by many parents, that my children’s education is more important to me than their teachers’ job security. I certainly wouldn’t tolerate, in actuality, unfair firings at any level, but as a parent, nor should I tolerate poor teaching.
If the teacher down the hall isn’t teaching anything, then the administration needs to call them out on that and start them through their due process. Nobody wants a bad teacher in the classroom. Due process doesn’t prevent teachers from being help accountable, it ensures that when a teacher who has been successfully teaching for years is fired they’re fired for the right reasons… not to cut the budget (since they’re finally making an almost respectable salary), not because they’re pregnant, not because their principal is holding a grudge, not because they stood up for what was right in the their classroom instead of what was being handed to them….
I think using data to drive professional development is great. I also don’t think that a bonus (on top of a solid salary schedule) partially determinded by test scores would not be the end of the world. I understand the public is looking for some sort of merit pay.
That said, I don’t think it’s the magic bullet people are hoping for. It’s not going to make the devoted dedicated hard working teachers anymore devoted dedicated or hardworking. As for the bad teachers, well they shouldn’t be getting paid anything. There is a process, it should be used.
Even though I’m known as a “union guy,” I am not a die hard stick-to-the-old-approach guy.
I’d like to see the unions take up (at least partially if not wholly) the responsibility of evaluation, professional development, and retraining teachers. When we take charge and control of our profession, I believe we’ll get the respect and the benefits of the public.
Until this happens, I believe we’ll be stuck in the status quo quagmire.
Tom
I’ll be borrowing your line “surgery with a back hoe.” OK I’ll probably be stealing it. My wife, who first supported Hillary, is getting frustrated with my continued support for Obama even though his education policy is brain surgery with a chain saw. Duncan is doing a lot more good things but I worry that the unintended results of these teacher quality “reforms” will be worse than NCLB. I’m schizophrenic too, because I’ve placed a lot of hope in Brad Jupp.
You write you are “willing to tolerate a few unfairly fired teachers if it means that our own children have zero chance of spending the year with a bad teacher.” Let’s unpack that. If every school has a teacher whose career is damaged or destroyed unfairly, how will schools attract and retain teaching talent. And it will be worse in the high-poverty schools where so many ineffective teachers are already located. These Value Added models tend to be off 10 to 15% of the time, under the best circumstance, but their invalidities are worse in low performing and segregated schools. If you have a one on six or one in five chance of getting unlucky, why accept the stress? Who will ever have any peace of mind on the job? It will be like George Soros metaphor. If you have six bottles of water and only one is poisoned, they are all still worthless. And when the principal of a high-poverty school is under the stress of meeting impossible growth targets, the dangers are multiplied.
And that should be a reminder that Denver’s Procomp roll out took plenty of time and started as a system for incentives not firings. The stakes are not nearly as high if the flawed growth model just gives you more of a bonus or less of a bonus.
I teach in the lowest performing school in Oklahoma and there is virtually no chance that a kid will be stuck with an ineffective teacher. Every kid gets more than one ineffective teacher every year.
Why?
Often the teacher is a good teacher who would be effective in an orderly school, but he or she can’t function in our anarchy. But every year we have probationary teachers, with no due process rights, hired with no chance of being effective who then get a non-continuing contract.
Why?
Why should a principal get rid of one warm body when he or she might not find another warm body who is any better. Besides, you got at the issue with your exclamation about the number of evaluations, “Five times!” Our principals go weeks at a time without letting classroom instruction enter their brains. They are bogged down with doing discipline, or more often, not doing discipline and dealing with the cascade of consequences. And I sure didn’t understand how Jupp, after the issue of discipline was raised, then denied we’re in a zero sum game.
Sadly, this experiment with test scores is mostly unnecessary. We all recognize “the teacher down the hall isn’t teaching anybody anything.” Under peer review, it is not hard to get rid of them. The problem is sharing power.
So, I wish Jupp would accept that test scores in the hands of management is a deal breaker in terms of teacher buy-in, but test scores in the hands of a peer review committee would be a workable compromise. Saying that we’ll figure out how to use test scores for evaluation somewhere down the rainbow, is too cavalier. When, not if, but when some districts do that as a result of the RttT, I hope we ram home a message in Court that they will never forget. But committees of teachers using data that we all know is flawed is the type of imperfect thing that adults do.
I dunno about others, but I tend to agonize over poor test scores and second-guess good ones. *sigh*