First of all, I am not a teacher or anything similar, and I am not a member of a teacher's union.

But this story has the same major mistake of almost all discussion of American education: it assumes that student output is largely the product of teacher input. And that just isn't true.

I mean, look at this quote: A good teacher this year will very likely be a good teacher next year. Unfortunately, when making hiring decisions, principals rarely have that information at their fingertips.

Except that this is not true. Anyone who's worked as a principal or administrator in a school knows the sad fact that test scores for individual teachers swing wildly from year to year. The author says that the change in students' basic aptitude is controlled by studying whether individual students improvement follows from year to year. But this makes no allowance at all for change in the student or their lives. Some students are ahead of the class in their early years and yet struggle later; they simply have a different aptitude relative to their position within grades. Why should a teacher be punished because a student has higher aptitude at the 3rd grade level than at the 5th grade level?

And this says nothing about a student's home life and changes within it-- probably the single most important factor for a student's success. I was orphaned in high school. As you can imagine, my test scores and grades plummetted. Were my teachers to blame? That is an extreme example. But every student has unique hardships and challenges that pop up, and those things definitely affect academic performance. How can you possibly find the signal among all that noise? These are still nothing like controlled test of teacher ability.

What I find particularly galling is that this principal both removes teachers based on test scores and yet thinks that he knows a good teacher when he sees one. But these are flatly contradictory claims. If you talk to public educators and principals, it is constantly the case that the teachers considered the best by their peers and principals-- the most dedicated, hardest worker, most involved teachers-- don't see the best test scores, or the best improvement in test scores. In fact, their test scores and change in test scores, like many teachers, often fluxuate wildly from year to year. Why? Do they suddenly become better or worse teachers? No, because their students have rapidly changing aptitude relative to their peers and have life situations that are anything but controlled and regular.

The idea that a student's output is roughly equivalent to a teacher's input is an article of faith in these discussions. But it is undercut again and again.