11-14-06 Teachers II

This week’s topic is Teachers, Part II. Last week, I began with the observation that any parent will tell you that the quality of teaching is the most important part of any child’s school experience. Interestingly, we’ve only recently developed the research tools in education to support this common sense notion.

For a long time, statistical studies in education found that the effects of socio-economic status were so strong a determinant of educational achievement that it was easy to conclude that schools really couldn’t make much of a difference. The famous Coleman report from 1966, for example, concluded that: “only a small part of [student achievement] is the result of school factors, in contrast to family background differences between communities.”

In the past 10 years, however, these results have been called into question. For example, statistical researchers have developed much more sophisticated techniques for identifying factors that lead to academic achievement, irrespective of socio-economic status. These techniques allow us to look past socio-economic status and begin to isolate what matters in schools. And guess what? It turns out that teachers matter a great deal. The primary vehicle for making this determination has been a research technique that has come to be known as value-added analysis. Rather than measuring students’ absolute achievement levels, value-added analysis measures how much each student has grown academically over the course of a school year. Value-added analysis allows you judge students’ actual growth compared to the growth that would be expected based on similar students, irrespective of the students’ starting point. This analysis can be done at the classroom level, which reveals the learning growth caused by individual teachers. Using value-added analysis, a number of large scale recent studies have shown that the quality of the individual teacher matters a lot. (By the way, the DOE’s accountability initiative will use a type of value-added analysis as one component of determining school quality.)

Value-added analysis shows statistically what parents have known all along. For example, a well-known study in Tennessee found that for low-achieving students, teachers in the bottom quintile produced academic gains of about 14 percentile points during the school year. Teachers in the top quintile, on the other hand, produced gains that averaged 53 percentile points. This same study showed that elementary students who had 3 bottom quintile teachers in a row grew by only 29% over the three years; those who had 3 top quintile teachers in a row grew by about 83%. Even more striking, the effect of three bottom quintile teachers in a row seems to persist well after the student is assigned to more effective teachers. It is very hard for a student to recover from three ineffective teachers in a row, no matter how effective the teacher to whom they may subsequently be assigned. Numerous other studies, using varying types of value-added methodologies, show similar results.

If you want to know more about studies of teacher effects, including the one above, here’s a nice, if a little dated, report from the Education Trust.

K:\Pool\Leigh's Weekly Reader\Haycock Good Teaching Matters.pdf


This is pretty geeky stuff, but it’s very important. Once we can reliably identify teachers who produce greater than average growth over time, we can begin to discover what they do differently from other teachers. Indeed, just such a project has been underway in Chattanooga, Tennessee. Based on multiple years of value-added data, a local foundation identified teachers who produced top quartile growth. They then did an extensive study of those teachers, resulting, among other things, in a series of tapes that captured their teaching techniques. You can see the results of this research and a description of the tapes here:

http://www.pefchattanooga.org/www/docs/5.261.html

Good teachers matter. Next week, a little more about the daily lives of teachers, and a few perspectives on teacher education.