PBS’s News Hour had a report yesterday about a Vanderbilt university study focused on whether tying teacher bonuses to student scores on standardized tests had any effect. According to the report, the researcher, Matt Springer, “offered 143 Nashville teachers bonuses of up to $15,000 if they could substantially raise test scores. Then he compared those teachers’ performance to a group of teachers offered nothing.” The result?
The conclusion of the report is that opportunities to earn a large financial incentive didn’t increase student performance. It didn’t change teacher behavior overwhelmingly.
In other words, the students of teachers offered this incentive didn’t do any better on standardized tests than students of other teachers. It seems you can’t buy better test scores.
Why does this matter? There are profound implications for the kinds of national education policies being pursued right now by the Obama administration. The President’s “Race to the Top” program, which basically has states competing for stimulus funding for education reform, has shown a clear preference for states that are willing to implement some form of “pay-for-performance.” According to the News Hour report, 11 out of the 12 states that have “won” this funding competition have agreed to tie teacher pay to things like student test scores. The idea here (besides busting teacher unions, which is, of course, also part of the point) is to create economic incentives for teachers to increase student test scores.
As incentives go, at least this one is a carrot instead of a stick (as NCLB basically is). But if Springer’s research is up to snuff, then the major premise behind the administration’s entire approach is terribly misguided. It doesn’t look as if tying teacher pay to student test scores will make a bit of difference, because there is no causal relationship (or apparently even a correlation) between those two things. It’s the wrong leverage, like trying to use a wet noodle to lift a tractor.
For the record, I have nothing against the idea of merit pay. I think better teachers should earn more than mediocre ones, and it is unfortunate that teacher unions and state governments seem to privilege seniority over capability. The problem is that student scores on standardized tests aren’t a decent measure of teaching effectiveness. What is? Everything else. If you have a child in school, you probably know exactly who the good teachers are and who the bad teachers are. Everyone knows. It’s not rocket science, but it’s also not quantifiable the way student scores are. Some folks just find numbers more convincing than what’s obvious.
We should spend more money on schools, and we should pay all teachers more. We get the quality of teachers we, as taxpayers, are willing to pay for. If teachers’ salaries were substantially higher, it wouldn’t necessarily raise test scores, but it would attract more people to the profession, thereby creating more competition for available teaching slots. Under the current “Race to the Top” system, we’ll have all the same teachers we’ve always had, but the (ostensibly) better ones will earn more, and the crappy ones will earn less. If teaching was a better-paying profession, though, then schools could be more selective about who they hire, more-effective teachers would get jobs, and less-qualified teachers would have to find other careers. Isn’t that what we want?