On December 4, Dr. Dale Ballou spoke to our Hierarchical Linear Modeling class about value-added models, which are statistical methods for quantifying teacher effectiveness. There has been a long-standing debate in education about pay-for-performance for teachers: proponents argue that we need a quantitative, data-based strategy for evaluating teachers, while opponents believe there are so many factors that go into good teaching that it would be futile to try to quantify it.
I was impressed by Dr. Ballou’s candor in discussing the issue. He described a 3-year, $9 million experimental study on pay for performance that’s currently in the works. In the study, Nashville mathematics teachers will vie for bonuses of up to $15,000 per year. Their performance will be compared to a control group of teachers who are not competing for bonuses.
One of my classmates, Chuck Munser from LPO, asked about whether there would be direct observations of teachers during the experiment so that changes in teachers’ instructional practices during treatment could be analyzed and reported. Dr. Ballou’s response was interesting. In essence, he explained that for the purposes of this experiment, it didn’t much matter how teachers’ practices changed as a function of the incentive - only that they did. In other words, the rationale for providing financial incentives to teachers based on their performance is for teachers to improve their effectiveness, and the question of the study is whether the incentives actually help them do it. Dr. Ballou didn’t rule out the possibility that future research may investigate the hows and whys, but at this time, the primary concern appears to be if -
At any rate, my interest is piqued. I’m sure I’m not alone in my curiosity to read the publications that come out of this study. With the current focus on improving the methods by which we assess student achievement to meet the requirements of the No Child Left Behind Act, it makes sense that we would want to figure out how to assess our teachers as well.
-posted by Peter Beddow

Recent Comments