"Twin" Studies Add More Pieces to Teacher Effectiveness Puzzle

Apparently, there has been some rampant speculation that little Eddie is actually little Eddies, that there is more than one of me. At least that’s what I’ve been told. Now I find that sort of talk a little disturbing. Who am I anyway?

Maybe someone has seen my doppelganger out there. I’d also given consideration to the possibility that my parents have locked an evil Eddie twin in a basement closet, only to be let out at inopportune times. Let me here and now assert my firm belief that such a notion was nothing more than the phantom of an overactive imagination.

Still, my curiosity is piqued at the potential boon to educational research that having a twin would provide. The National Council on Teacher Quality today brought my attention to a pair (!) of studies — one in the Netherlands, one in the United States. The idea? Take a set of twins and put them in different teachers’ classrooms to test the effect.

Let’s start in the Low Countries of western Europe, where respected adult sources have dispelled my concerns that research results might be tainted by the extensive wearing of wooden shoes. It seems there is no widespread wearing of wooden shoes, so some effort will be needed to help determine the applicability of this Dutch study’s findings to our schools on this side of the Big Pond.

Researchers Sander Gerritsen, Erik Plug, and Dinand Webbink studied nearly 500 sets of twins in different classrooms and found that students (especially in earlier grades) benefited significantly from having more experienced teachers. In the conclusion they note:

It remains unclear whether this effect is caused by training on the job or reflects the effects of unobserved teacher quality correlated with
attaining more experience in education. We have explored the plausibility of various mechanisms that might explain the robust finding that students in classrooms with more experienced teachers perform better. We do not find evidence consistent with mechanisms that stress the importance of changes over time such as changes in the quality of teacher education or changes in outside opportunities in the labor market. However, we find that teacher experience also matters for career stages with less labor market mobility.

Like any good research, one finding opens up other areas of meaningful inquiry. The Netherlands study’s American “twin” by Moiz Bhai and Irina Horoi yielded somewhat similar findings. Except that, following a corpus of research on American education, it finds the gains in teacher quality plateau after about five years.

The report also echoes an overwhelming body of evidence on teacher characteristics that don’t matter in improving student learning:

The effect of master’s degree on achievement remains statistically
indistinguishable from zero.

What I would love to see in a study someday is a distinction made between different types of master’s degrees — whether in relevant content areas or from schools of education. Anyway, in between the results of experience and master’s degrees, Bhai and Horoi find a small relationship between National Board certification and measurements of greater learning.

In the ongoing quest to explain just how Teachers Matter, what sets them apart based on effectiveness, and what policies make the best difference for kids, these two studies add just a little bit more to solving the puzzle.

And I don’t want any extra help working on puzzles. Being a twin may be fine for others, but not for me: I’m selfishly thankful to be the product of a single birth. You’re probably glad there’s only one of me, too.