“Merit pay” in public schools – what we mean by “junk science”
Think about scientific method. Start with questions and hypotheses instead of declaring the solution in advance without even thinking what the questions are.
What are the hypotheses?
Let's begin with facts we know: nobody’s complaining about the performance of teachers as a group in the top suburban public high schools. The problems we can identify are most prominently in inner city schools where the neighborhood is low-income, often below the poverty line.
So let's formulate the very first hypothesis implicit in support for "merit pay." Where is the slightest bit of empirical research establishing or even pretending to support the premise that the problem with performance of kids in inner city schools is that the teachers in the aggregate do not try hard enough? The whole premise has to be that merit pay will improve effort. But that first question has to be answered in a convincing fashion, or the whole premise falls apart. Sorry, saying a friend of yours told you a teacher he heard about who sat at her desk all day is slightly below the standard we usually require for data to count as empirical evidence. Neither does your experience with teachers you just knew were incompetent (even though you never actually set foot in their classrooms) during your Teach for America stint. In fact, even quantified empirical research evidence of more incompetent teachers in inner city schools, if it exists, has nothing whatsoever to say on whether they don’t try as hard as they know how. What, an extra $2 K will turn an incompetent teacher into a good one? Finally, saying everyone knows it’s true because (everyone knows) greedy unions keep the lazy teachers from losing their jobs is not evidence, either.
Without evidence on the question of whether the problems are due to teachers not trying as hard as possible, you don’t even get to the next question: whether it's the absence of the chance to win extra money that keeps teachers in inner city schools from trying to do the best job they know how to do. And, of course, without some bit of evidence suggesting that hypothesis could be true, you can’t even get to the next question you have to answer: whether there is the slightest bit of evidence that a merit pay system would fix that problem, by causing all teachers to try harder, and resulting in better test scores for inner city kids.
Simply to identify the underlying premises of so-called “merit pay” for public school teachers really reveals what a frivolous concept it is. Does anyone honestly believe that a merit pay system – forget thinking about the bureaucracy that will be needed to administer such a system – will be the measurable cause of improvements in student performance in a big-city public school system?
The concept of “junk science” has been formulated for the court system primarily by corporate defendants being sued for allegedly defective products. They use it trying to fend off expert witnesses with unconventional approaches to scientific disciplines. When you don’t even bother to formulate your hypotheses and jump straight to your solution without even trying to find any answers to those questions, as the merit pay proponents do, that is an iconic example of junk science.
This turkey should be beat into the ground until it's dead forever -- at least until those questions are answered in the affirmative. Calling it what it is -- junk science -- is a start.
What are the hypotheses?
Let's begin with facts we know: nobody’s complaining about the performance of teachers as a group in the top suburban public high schools. The problems we can identify are most prominently in inner city schools where the neighborhood is low-income, often below the poverty line.
So let's formulate the very first hypothesis implicit in support for "merit pay." Where is the slightest bit of empirical research establishing or even pretending to support the premise that the problem with performance of kids in inner city schools is that the teachers in the aggregate do not try hard enough? The whole premise has to be that merit pay will improve effort. But that first question has to be answered in a convincing fashion, or the whole premise falls apart. Sorry, saying a friend of yours told you a teacher he heard about who sat at her desk all day is slightly below the standard we usually require for data to count as empirical evidence. Neither does your experience with teachers you just knew were incompetent (even though you never actually set foot in their classrooms) during your Teach for America stint. In fact, even quantified empirical research evidence of more incompetent teachers in inner city schools, if it exists, has nothing whatsoever to say on whether they don’t try as hard as they know how. What, an extra $2 K will turn an incompetent teacher into a good one? Finally, saying everyone knows it’s true because (everyone knows) greedy unions keep the lazy teachers from losing their jobs is not evidence, either.
Without evidence on the question of whether the problems are due to teachers not trying as hard as possible, you don’t even get to the next question: whether it's the absence of the chance to win extra money that keeps teachers in inner city schools from trying to do the best job they know how to do. And, of course, without some bit of evidence suggesting that hypothesis could be true, you can’t even get to the next question you have to answer: whether there is the slightest bit of evidence that a merit pay system would fix that problem, by causing all teachers to try harder, and resulting in better test scores for inner city kids.
Simply to identify the underlying premises of so-called “merit pay” for public school teachers really reveals what a frivolous concept it is. Does anyone honestly believe that a merit pay system – forget thinking about the bureaucracy that will be needed to administer such a system – will be the measurable cause of improvements in student performance in a big-city public school system?
The concept of “junk science” has been formulated for the court system primarily by corporate defendants being sued for allegedly defective products. They use it trying to fend off expert witnesses with unconventional approaches to scientific disciplines. When you don’t even bother to formulate your hypotheses and jump straight to your solution without even trying to find any answers to those questions, as the merit pay proponents do, that is an iconic example of junk science.
This turkey should be beat into the ground until it's dead forever -- at least until those questions are answered in the affirmative. Calling it what it is -- junk science -- is a start.
posted by KISSWeb | 5:27 PM
2 Comments:
Absolutely brilliant,
There has never been one definition of teaching effectiveness accepted by all educators, students, parents, politicians, etc. It is impossible to measure what cannot be defined. Any bureaucracy developed to determine amounts of merit pay based on teaching effectiveness would do more harm than good. To get a glimpse of probable effects of merit pay, look at the narrowing of curriculum and the cheating on tests that have developed in response to the No Child Left Behind Act.
Post a Comment
<< Home