@profpiro I think dropping student test scores from teacher evaluation makes sense. Just because we should doesn't mean we can. When you look at the "

@profpiro I think dropping student test scores from teacher evaluation makes sense. Just because we should doesn't mean we can. When you look at the "research base" ots the same five #edreform fols citing the same five economist who cite the same five #edreformers. Most of the studies published by think tanks and not peer reviewed.

In CT for example we have some districts with one school and 130 students. You really think we can parse teacher effect in that variance?

So the solution folks say is combine with class observation (r hopefully >60%) combining to unreliable measures do not make composite scores more reliable.