In 2010, I was in Denver the day that the Legislature was debating S. 10-191, a bill sponsored by young Senator Michael Johnston. It was a bill to base 50% of teachers’ evaluation on test scores, a new, untried, and very controversial idea. Teachers were strongly opposed, and the Legislature was deeply divided but the bill passed. I was supposed to debate Johnston at a lunch in downtown Denver, but the debate didn’t work as planned. There were about 60 civic leaders in the room, and we waited patiently for Johnston. We finished lunch and still no Johnston. So I got up and gave my talk and explained why it was wrong to evaluate teachers and principals by test scores (at that time, I was working with Richard Rothstein on a statement against test-based evaluation that was signed by a bevy of testing experts). No sooner did I finish, then presto-change-o, young Senator Johnston strides through the doors in the back of the room. He had carefully managed not to hear anything I said.
He then proceeded to talk for 20 minutes or more about the glories of using test scores to judge teachers, principals, and schools. He predicted that the passage of his bill would bring about miraculous improvements in education across the state of Colorado. He praised his legislation as the dawn of a new day. Michael Johnston is an alum of Teach for America (were you surprised to hear that?). The title of his bill was something grandiose and completely fraudulent, something like “Great Schools and Great Teachers Act of 2010.” Gosh, it is six years later, and almost everyone except Michael Johnston knows that test-based accountability flopped. It flopped in Colorado and it flopped everywhere else, despite the billions pumped into by the federal government, the Gates Foundation, states and local districts.
Just in the past few days, both John Merrow and the team of Checker Finn and Michael Petrilli independently agreed that teacher evaluation by test scores was Arne Duncan’s worst mistake. John Merrow said, “Tying teacher evaluations to testing was a mistake, probably Arne Duncan’s biggest mistake.” Petrilli and Finn said that the federal mandate for teacher evaluations was “politically poisonous.” But not in Colorado, it seems.
A group of legislators proposed revising his bill to eliminate evaluation by test scores, and it appeared to have the support it needed. But at the last minute, two of the Republicans changed their minds about dropping the teacher evaluation by test scores, and Michael Johnston’s failed idea survived by a vote of 6-3. So Johnston and five Republican Senators managed to preserve this program, which has not worked in Colorado nor anywhere else in America. Six years after passage, there is not a whit of evidence that it improves teaching and learning.
Do you think Michael Johnston read the statement by the American Statistical Association in 2014 warning that using test scores to evaluate individual teachers is not a reasonable idea, because teachers influence between 1-14% of the variation in student test scores? I don’t think so. Do you think he saw the statement by the American Educational Research Association last fall against the use of this method? I don’t think so. Do you think he read the statement by Edward Haertel, the Stanford University testing expert, on the flaws of value-added assessment? Do you think he knows that it has been dropped by district after district because it costs millions and it has failed everywhere to identify the best or the worst teachers? Apparently not.
Michael Johnston doesn’t know what he doesn’t know. With this last-ditch effort to preserve the bad idea he sponsored, he has proved that he neither reads nor thinks.
Message to Colorado parents: Opt out. Resist. Do not let the state impose bad policies on your children or their teachers.
Diane Ravitch's blog 