Dana Goldstein and Ezra Klein are two smart, capable writers who know plenty about education policy. And yet, when they get together to discuss teacher evaluation, they completely - COMPLETELY - miss the most important point to be made in the entire debate:
I've watched this twice: there is a lot of talk about using student test scores to evaluate teachers, but not once do either mention that it doesn't work.
There is no debate on this - even the corporate reformers admit that value-added modeling based on standardized test scores is hugely error-prone. Every major research group that has studied this says not to do it, including the National Research Council (p. 8), the "gold standard"of science research in the US. The error rates for misidentifying a teacher are 35% - and that's only the 10%-20% of teachers you could judge with test-based evaluation anyway.
Isn't this the key part of the entire debate? Isn't this the most relevant point to be made in all this?
How in the hell can you have a serious talk about using test scores to evaluate teachers without mentioning that all the research says NOT TO DO IT?!?!
I'm sorry, Ezra and Dana, but this is just not acceptable. You MUST acknowledge this every time you discuss the topic. Omitting the fact that evaluation of teachers by tests scores has huge error rates is like omitting the fact that the Earth is warming from a climate change debate.