Updated 11 months ago
Trying to come up with a formal system that identifies and rewards good teachers, and either improves or fires the bad ones, is a challenge that nobody has mastered, however. A host of states has experimented with various merit pay systems over the years. Among others, Tennessee — I worked briefly in that state's Department of Education about 30 years ago — enacted what it called a Career Ladder program in 1984. The program used master teachers to observe and evaluate instructors who applied for higher-level certification and the higher pay that came with it. The program was completely voluntary — if a teacher didn't want to try to move up the ladder, she didn't have to.
The program didn't address firing bad teachers but did offer a professionally sound basis on which to pay the best teachers more. It was always a political football, and after a promising start the teachers union succeeded in watering down the standards so that everybody qualified for the upper levels — in the union's eyes, everybody's a star. The Tennessee Legislature closed the program to new teachers in 1997.
Florida now has adopted an evaluation and merit pay system, creating a way for teachers to be paid more based on their students' performance on standardized tests. It's not a horrible idea in presuming a connection between effective teaching and learning. (Teachers unions see no such link.) But even if it survives a court challenge, it will fail. Not because there's no funding source for the merit pay but because it's a ham-handed, top-down and unsophisticated approach to a very sophisticated set of issues.
How will standardized test performance be used to evaluate teachers in subjects for which there's no standardized test, for instance — foreign languages, for example? Will students be assessed on pure performance or on gains? If a student in an honors program scores well early in the year and then scores at more or less the same level late in the year, did the teacher do a good job? If a teacher in a tough school makes even a little progress with a group of disadvantaged kids, how is that added-value calculated vs. an easier-to-teach group that shows better progress on the tests? How do you account for cumulative progress over several years?
The new Florida law will also fail because the educational establishment — the unions and district leadership — will game it to death. For the union, the very idea of merit pay will always be a non-starter. It's simply part of the union's DNA that a teacher with eight years' experience is worth more than one with five years' experience, regardless who does the better job. And that an elementary school teacher is worth just as much as a high school physics teacher, supply and demand be damned.
Some gamesmanship is already occurring. At a local middle school, I'm told, the principal already has decided that each teacher will teach classes of honors students from the school's magnet program and an equal number of classes of students from the "traditional" program, many of whom tend to be tough-to-teach kids passed up the grade levels without learning much. So no matter how poorly the traditional students perform, the magnet kids will likely do well enough so that each teacher, on average, can show some student "gains." At the least, a teacher can balance out the overall results. Merit pay for everybody.
Nobody, of course, will be held accountable for failing to really teach the traditional kids anything. And nobody will be accountable for actually challenging the magnet kids.
Florida's merit pay law is a nice piece of chest-thumping by the Legislature, but real, substantive educational change will take a lot more. Pay differentials for subject-area teachers in short supply — math, science and technical education, for example. More training, empowerment and accountability for principals, who ought to be able to manage their school's budget and hire, fire and reward personnel the same way branch managers of stores can. More competition for the public system, via charter schools and, yes, vouchers.
There's nothing wrong with tests as one gauge of performance — teachers have no problem basing students' grades on the tests they give. But the one-gauge-fits-all approach to evaluating teachers is as dumb as the school districts' one-approach-fits-everybody approach to educating kids.
|More columns by Executive Editor Mark R. Howard are here. Note: Articles older than 30 days require registration (it's quick and free).|