On Brittanica Blog, Dan Willingham takes a look at teachers’ wish for greater respect and the role of unions in winning it. He observes that unions perform two important functions that are fundamentally at odds with each other: they protect the rights of individual teachers in personnel matters, and they undertake public relations and other activities in an effort to promote the profession.
On the one hand, if your mission is to protect the members of the profession from unfair termination, you will insist on a rigorous process by which their incompetence must be demonstrated. On the other hand it must be admitted that in any profession employing several million people some are incompetent, and if your job is to protect the reputation and integrity of the profession, you should want those people to leave.
Since the process of determining who is or isn’t a good teacher is far from foolproof, mistakes will be made, Dan notes. So the question becomes what kind of mistake do you prefer: firing someone who is actually a good teacher? Or failing to fire an incompetent teacher? If you’re cautious about not allowing good teachers to be fired, you’ll inevitably allow more poor teachers to remain. If protecting the reputation of the profession is your main concern some good teachers will end up being drummed out of the corps unfairly.
“If your diagnostic is imperfect, you’re going to make errors,” Willingham writes. “All you can do is choose the proportion of error types.” He argues that teachers unions have handled this tradeoff badly, harming the reputation of teaching as a profession.
While Dan’s post is at Brittanica Blog, the debate over it is at Eduwonk. Teachers’ unions “are in a purgatory of their own creation,” opines Andy Rotherham. ”They don’t want to use data to evaluate teachers and they don’t want to use managerial discretion. I guess that leaves the Magic 8-Ball?” After much back and forth about the union’s preferred role Willingham makes an observation that seems unassailable: “The President is talking about getting rid of poor teachers,” he writes. ”It appears likely that something is going to be done, so you may as well try to take control of the situation so it’s something you are doing, rather than something that is done to you.”
Lead, follow, or get out of the way, in other words.
N.B. Dan has a brilliant new book out called Why Students Don’t Like School, which if I had a magic wand would appear on the desk of every teacher in America. Absent that, I’m thrilled to report that Professor Willingham will be taking over the Core Knowledge Blog all of next week to talk about some of the insights from his work and his new book while I take a week off from blogging. Don’t miss it.
I second the recommendation of Willingham’s book. It’s very good. I got it just in time for Spring Break.
Comment by Redkudu — March 18, 2009 @ 3:48 pm
I refuse to believe that all sides could not come together and find SOME mutually satisfactory plan for making teacher evaluation more rigorous while at the same time avoiding the folly of basing it only on test scores. Surely some kind of multifaceted gauge that would include “service to the school” (I believe many college professors are evaluated on this basis, looking for things like committee memberships, sponsoring student activities, etc.), engagement with families and the community, lessons that are dynamic and rigorous, good attendance and punctuality, collegiality, etc. would be fair, comprehensive, and actually helpful in pinpointing good teachers and bad ones…?
I mean, I’d like to have some kind of validation that I am a “good teacher.” Right now, I might be excellent, I might be mediocre, and I might be terrible but just not quite egregious enough to merit my principal bothering with firing me. I’d like to be judged by my colleagues, my superiors, my students, and their families, and get some more strong sense of what I do well and where I need to improve. Some kind of weighted formula could be introduced, and if a teacher didn’t aggregate a certain number of points or a certain percentage, say, two years running, he or she would have to go.
This seems like a heady task to be sure, but not impossible–maybe a nice job for end-of-career teachers and administrators who want to stay active in the profession but transition out of the classroom/school? It should happen throughout the school year and not be final until the end of the school year (avoiding the problem of judging on one standardized test that might happen anytime between October and May), and include multiple opportunities for growth and rewards for teachers.
This isn’t crazy, right?
Comment by Miss Eyre — March 18, 2009 @ 7:04 pm
Ms. Eyre:
Is there anyone on God’s green earth seriously considering the evaluation of teachers based “on one standardized test?” We keep talking as if there are. It is also very difficult to discern if teachers currently receive no evaluation, or just stupid evaluations (by incompetent principals), and whether or not teachers improve, stay the same or go away based on current measures.
Comment by Margo/Mom — March 19, 2009 @ 10:48 am
I agree with you on the latter part. As I said in my comment, even as a teacher I feel like it’s difficult to measure myself against my colleagues. But unfortunately there are indeed people who want to base merit pay decisions on standardized test scores, and nothing else. Anyone who proposes that achievement–teacher or student–can be measured by standardized test scores is looking for a way to gauge achievement that may be easy and fast, but is not accurate or fair.
No teacher worth anything is afraid of being evaluated comprehensively. I’m not. I just don’t want to be judged on one day, one score, that’s all.
Comment by Miss Eyre — March 19, 2009 @ 7:26 pm
“But unfortunately there are indeed people who want to base merit pay decisions on standardized test scores, and nothing else.”
Ms Eyre:
Can you name three? And even better, point to a statement in which they say so?
Comment by Margo/Mom — March 20, 2009 @ 5:11 pm
Certainly.
1.) http://www.nytimes.com/2006/01/13/national/13houston.html
(In 2006, the schools of Houston implemented a merit pay plan centered around standardized test scores.)
2.) http://www.heraldtribune.com/article/20070916/NEWS/709160385
(The STAR program in Florida sought to tie merit pay to standardized test scores, although the program was ultimately rejected.)
3.) http://www.rethinkingschools.org/archive/22_04/meri224.shtml
(Right here on my home turf of NYC, a pilot merit-pay plan was based almost entirely [85%] on standardized test scores.)
4.) http://www.imdiversity.com/villages/Careers/articles/pns_calif_teachers_0105.asp
(This commentary from California examines a proposal by Gov. Schwarzenegger to base merit pay on–you guessed it–standardized test scores.)
These are but four (do I get merit pay for finding four?) examples of proposed merit pay plans based entirely or very heavily on standardized test scores. Not cool with me, with most teachers, or, I would imagine, with most parents.
Comment by Miss Eyre — March 20, 2009 @ 5:50 pm
Actually, allow me to clarify #1 above: The Houston plan was to be based ENTIRELY on standardized test scores.
Comment by Anonymous — March 20, 2009 @ 5:51 pm
Crap, my response got eaten! With all these nice links and everything. Boo. All right, I will re-create.
1.) http://www.nytimes.com/2006/01/13/national/13houston.html
(Houston plan for merit pay is based ENTIRELY on standardized test scores.)
2.) http://www.heraldtribune.com/article/20070916/NEWS/709160385
(Florida’s STAR program, though it was ultimately rejected, would have tied merit pay ONLY to standardized test scores.)
3.) http://www.rethinkingschools.org/archive/22_04/meri224.shtml
(Right here in NYC, a pilot merit-pay plan was based almost entirely [85%] on standardized test scores.)
4.) http://www.imdiversity.com/villages/Careers/articles/pns_calif_teachers_0105.asp
(A proposed merit pay plan in California would have been based on–you guessed it!–standardized test scores.)
There are but four examples of proposed merit pay plans that would have looked only, or very heavily, at standardized test scores.
Comment by Miss Eyre — March 20, 2009 @ 5:58 pm
The subjective system of teacher evaluations in our schools has been and continues to be nothing short of a farce. The principal spends one 30-45 minute visit (usually announced) in a classroom observation which often turns out to be the classic dog and pony show. The teacher runs out, gets her hair done, buys a new outfit, and prepares one exemplary lesson all for the obvious purpose of impressing the evaluator. What a joke!
The same teacher shows up the next day in a sweat suit, unprepared, and proceeds to hand out one set of worksheets after another for the sole purpose of keeping the kids busy.
I saw it happen over and over in three and a half decades in a good middle class Massachusetts district. It was borderline embarrassing. It satisfied the teacher’s and the principal’s obligations but was an outrageous diservice to the students.
The public knows all about these practices. The folks responsible for education reform since 1983 (NAR) are also very aware of this travesty of practice. They’re the ones pushing for a change in the existing practice looking to inject some degreee of objectivity into the process and I cannot fault them one bit.
The current system of teacher evealuation in US schools today is nothing short of a joke. It has to change.
Comment by Paul Hoss — March 21, 2009 @ 7:35 am
Good point, Paul. Observations as evaluations are a bit of a farce. In New York, you do a pre-observation meeting with the principal to go over the lesson he or she will observe. It’s a bit like inviting the batter to go over pitch selection with the other team before the at bat. Not that an evaluation should be adversarial, but it really is a show and nothing more.
Comment by Robert Pondiscio — March 21, 2009 @ 7:47 am
And then your evaluation for the year is based on these “observations,” which take place more or less exactly as Robert described, and most people get less than three of them. Even as an early-career teacher, I’ve only ever had two in a year. And people don’t slip through the cracks? Of course they do.
Comment by Miss Eyre — March 21, 2009 @ 10:48 am
Gentlemen:
While I certainly agree that teacher evaluation and development is in sorry shape in this country, in general (and there are a few notable exceptions)–neither of your descriptors really leads to an understanding of what a useful evaluation of teaching might look like. A valid, reliable evaluation of teacher effectiveness would be a great asset–and might allow us some confidence that we have hired and fired the right teachers (which was the point of Dan Willingham’s original blog).
That evaluation would certainly include a pre-conference. In an ideal conception of teacher evaluation, the teacher would tell the evaluator (who might be a principal, but also might be a demonstrably accomplished colleague) what they were aiming for in the lesson to be evaluated. Learning goals, planned activities, how learning will be assessed, and how success will be defined–all things that a good teacher should be able to articulate. Then, the evaluator watches the teacher not only to see how the plans are carried out, but also how the teacher deals with unanticipated problems. Because good teaching is all about decision-making, often in the moment.
That’s not to say that a supervisor (or, again, supportive colleague) should not regularly walk through any/all classrooms, observing the nature of interactions and content presented, and keeping notes. Or that a teacher should not feel free to ask a principal or colleagues for assistance (without being judged a failure). The movement to flesh out teacher evaluation processes, with pre- and post-conferences, annual self-set teacher improvement goals and peer collaboration to improve practice (for example–Japanese lesson study models) are coming from practitioners. It’s the policymakers who want to know why we can’t just use test scores.
There was a piece in Education NEXT a couple of years ago around research showing that principals, asked to rank order their teachers’ effectiveness (based on test scores) could reliably name the top 10% and the bottom 10%. They were not good at figuring out how effective 3/4 of their teachers were. And it’s that same 3/4 who could most benefit from an evaluation process that not only weeded out the lousy teachers, but could help mediocre teachers get considerably better.
Painting a picture of teacher evaluation as a universally useless sham or a game of gotcha doesn’t move us toward a better teaching force. Teachers should be telling the principal exactly which “pitches” they plan to use, and why they’ve chosen them. You’ve both described adversarial teacher-principal relationships as an essential part of the evaluation equation.
Comment by Nancy Flanagan — March 21, 2009 @ 3:04 pm
Ms. Eyre:
OK, we’ve drifted a bit, equating bonuses and merit pay with teacher evaluation–two completely different things, but I didn’t catch it when I asked my question, so here we are. The Houston plan that you cited actually relies on three different tests, or sets of scores (as well as using comparative, rather than absolute, or improved scores–which I found a bit puzzling). It’s hard to tell exactly what Florida is doing, or intends to do–it sounds like a grand free-for-all between the legislature, local school boards, and of course, the unions. In NYC, there was a trade-off in the inclusion of test scores in the bonus system (full retirement at age 55–and don’t I wish!). As near as I can tell, while the bonus is based on scores and numbers of teachers in the building, it is distributed based on decisions made at the building level–which I assume would play out as an equal sharing, or distribution based on seniority or something.
So, what we have is salaries based on canned criteria (degree plus years of service), evaluations based on who knows what (and in all likelihood the kind of dog and pony show that Paul describes–pretty much all that the contract will allow in my disrict)–that affect nothing at all in the life of the teacher (advancement, continued employment, certainly not salary) and a possible bonus based on the overall scores of all students granted to teachers in a building across the board. Wow. Good thing you can retire with full benefits at age 55. Sounds unbearable.
Comment by Margo/Mom — March 21, 2009 @ 4:48 pm
On a tangent, I am currently reading Willingham’s book, and while I agree with some of his initial ideas – at least once he elaborates on them rather than the pithy and provoking statements he makes – I think he assumes too much in the latter half of the book. His arguments are loosely based on research, and it is this that gives it the sense of being sound, but I can’t help but feel some of his conclusions fly directly in the face of things I have observed in 18 years of teaching. I think he is a valuable voice in the conversation, but his conclusions are far from certain, IMHO.
Comment by Jim Wysocki — September 13, 2009 @ 7:09 pm