If it's Tuesday, it must be...merit pay

I met this morning with two veteran teachers who work in a local middle school in a district considered one of Maine's best. They had heard about the possibility of Maine Enterprise School in their community and wanted to learn more about our approach. One of the teachers was talking about the math curriculum - how she's been ordered to abandon her successful, project-based, integrated math lessons in favor of an approach designed to raise test scores by scripting each lesson each day for the entire school year. Consistency, this is called. I'll admit that I was surprised, in a district that has such a history of "best practice" and test scores that are generally thought of as high, to see exactly what she meant - but literally, that is what it was. If you are in sixth grade in this school, the only variable is when the first snow day will come and throw the whole shebang off.

She had mentioned earlier that she was coming back to school after having suffered a serious brain injury, and we all laughed when she said, "at least I don't have to worry about whether my brain is up to this, since I won't have to use it." She went on to explain that she's having to re-teach herself certain basic things - like how to recognize and then attach meanings to words - but that she'd found nothing in the math lessons that actually required her to know anything about math.

When I got home, I read the NY Times article about the states that won the Race to the Top money and the emphasis on systems that enable schools to tie teacher evaluation to to standardized test scores. Maine, like many other states, changed the law to allow this, and has commissioned a panel to figure out what that will actually look like. Two of the gubernatorial candidates have come out strongly in favor of "merit pay." My conversation with these two teachers this morning has me thinking deeply about how a seemingly hard to argue with concept - paying more for better results - could go profoundly wrong.

Before I start, though, let me make clear that I think that standardized test scores can be a valuable source of data, that higher standards in general are crucial to meaningful reform, and - though I'm a former Teamster and union rep for my NEA affiliate in Wisconsin - I firmly believe that our teachers' unions, Maine's included, are protecting their membership at the expense of the kids they serve. If we ever hope to make real change, we must make getting and keeping a teaching job as hard, or harder, as it is to get and keep a job as a doctor, lawyer, plumber or electrician - all of which currently have far more genuine standards for admission and ongoing performance assessment that does teaching.

That said, I have rather serious questions about whether what is called merit pay - a mishmash of incentive plans that purport to pay "better" teachers more - will do that. For instance, "The scoring system gave credit to states in which unions signed onto the application, but New Jersey’s union withheld its approval after Gov. Chris Christie decided he would not compromise on some issues, including merit pay for individual teachers. The union wanted merit pay allotted to whole schools, to be spread out among all teachers at the schools." There is a good idea here - incentivize whole school performance and systemic improvement rather than treat teachers as disconnected variables in a larger system. But, as anyone who has spent time in some of the most chronically under-performing urban schools in the US will tell you, any system that does not create serious, fair,and immediately actionable processes for showing the least effective teachers the door is protecting the wrong thing.

In New York, "The new teacher evaluation system, tied to test scores, could make it easier for principals to single out teachers deemed ineffective, although state laws still make firing such teachers so arduous that only a few are forced out each year." Ignoring that last part, that sounds like a good idea, right? Well, it depends on what is meant by "tied to test scores." If it means what I think it means, then it could be a terrible idea. Remember our brain injured friend? We know how to raise test scores, especially in elementary and middle school. Do just what her school is doing - create teacher-proof lessons and let talented (or perhaps just feared) classroom managers lockstep as many kids through them as we can. In most classroom in most schools, standardized scores will rise - for a time, anyway.

Unfortunately, there is no correlation between high test scores and any socially useful function. There is some correlation between some things that can be measured by standardized testing - such as reading comprehension - and other kinds of success. But there is a far greater correlation between things that can't be measured in standardized tests - resilience, problem solving, social aptitude - and the kinds of success that matter the most. And there is a profound downside to the drill and kill approach to raising scores, which is starting to show up in increased, rather than decreased dropout rates.

What happens when all those kids who have sat in that sixth grade classroom preparing over and over for the test are asked - sometimes a year or two or four later - to link the concepts together? For those who actually learned it, it's there. But most never learned it...quick, how many of you once passed a test on the quadratic formula...how many of you still know it...how many of you can cite an example where a person not working as a teacher would use it in their work, and for what? And what happens when, after three, or five, or eight years of that in all their classes - they decide that something - maybe sports, maybe video games, maybe sex, drugs and rock and roll - are far more interesting, and immediately relevant, than helping their teacher become the Tiger Woods of "merit pay?" And though you may be tempted to say we know the answer, I'm not sure we know what it looks like on the much more massive scale it may be implemented on in the states that "won" the race to the Top.

We do know, though, of one, and only one, demonstrable connection between incentive pay and improved student learning. "New York lawmakers never considered some of Mr. Klein’s more far-reaching demands, including the discretion to pay more to teachers who work in poor neighborhoods and in certain subjects, like math and science." That's right, the lawmakers never considered the only thing that does work - pay people more to do the hardest work.

And here, I hate to say it, is where the whole "merit pay" agenda of the Obama administration is built on ether. Arne Duncan is wicked smart and has the right idea - he needed strong, unambiguous data to undergird his efforts to rid Chicago of the worst teachers. But the emphasis on linking individual teachers to test scores will only create a new version of the kind of "merit" that has had generations of graduates of elite schools creating systems that worked for them - such as our health care, political, and oh, say, mortgage-backed securities systems - that may not work so much for the rest of us. These are bright, hardworking kids who always did the right thing, and then got into the best schools, and then got the best jobs, and then got a chance to work in the best administration, and it is no surprise that their instincts run towards a metric that can tell which of them is the "best" teacher.

Unfortunately, we know exactly what this produces, because we have that system already...the newest teachers teach the most difficult kids, the "best" teachers teach the AP classes...and then, if they're in a school with too many "hicks," as one Maine teacher recently told me, they take their superb resume to a district where they can really "teach." Oh, did I mention we pay those teachers MORE to move to those fancy suburban schools?

In fact, the best teachers know that teaching kids who come to school without every advantage is FAR more difficult. They know that motivating a young person to care about learning, not just passing a test, is a necessary precondition to any result worth getting. And they know that - with deep apologies to all for whom the shoe fits - that teaching the next generation of cabinet members is far less meritorious, but far more likely to win you the label of "master teacher," than teaching hicks. While some might call that merit pay, it's more like "born on third base and think you hit a triple" pay.

So what will work, then? As usual, we know that already, and we won't do it.
1. Pay the best teachers to teach in the places where they are most needed.
2. Professionalize teaching. Built into Maine Enterprise Schools' model contract are four levels that comprise a realistic and rigorous pathway: an intern or practice stage that exposes the prospective teacher to the discipline and ascertains some baseline level of likely competence; a novice stage where techniques and talents are honed for several years before certification - via portfolio or performance assessment, not just test scores; a "professional" teacher stage, where most teachers would spend most of their careers, engaging in regular professional development and demonstrating the connection between that development and outcomes for their students. Some may then choose to become a master teacher, or teacher of teachers, adding the significant responsibility (and significant pay) of someone who is at the top of their craft and chooses the added work and challenge of passing that on.
3. Set clear standards for teacher performance (which means measuring student performance across many values), assess them fairly, and create systems where the master teachers mentioned above decide who moves to the next stage, who stays at the stage they are at, and who, no matter how long they've been teaching, how nice they are, or how many football games their team has won, gets the boot.
4. More important than anything else: Dismantle, boldly and swiftly, the factory model schools that we continue to poor money into. Maine has a school construction formula that makes it impossible to build any new public school that will enable teachers to work in ways that are likely to truly produce results for each learner. Instead of putting that taxpayer money into a flexible, multi-discipline, student-centered approach, Brunswick is building a $23 million, 750-student school to serve grades 3-5. That's right - 250 kids per grade, to be shepherded through teaching systems designed to create the best results on those standardized tests. We can claim, as those who support this approach in Brunswick do, that we're not just fattening the kids up for slaughter like so many Perdue chickens, but good intentions pave a certain road. As some you may have heard me say, I can use that $23 million to build 23 schools for a million bucks each, each serving 100 kids - that's 2300 kids - with teaching techniques that are made for kids, not chickens...and do it a lower operating cost per student, with happier, more effective teachers.

Anyone know a candidate in search of a platform?