Farina Casts Vote for 30% Teacher Eval Based on Test Scores

About evaluations: as an ambitious professional, I do not accept the principle that an inherently invalid procedure like misusing a diagnostic tool for children to assess me short form any part of my evaluation. Do we find it acceptable to eat a meal that includes 30%, 10% or 1% inedible ingredients?....Jeff Nichols, CTS

She's drunk.  What percentage should be applied to her job?...Khem Irby, Parent activist

I do believe we need to use test scores for a portion of the evaluations,” Fariña said. “The percentage we use is up for debate. In my opinion, 30% is acceptable.”
http://www.nydailynews.com/new-york/education/fari-takes-school-questions-packed-bronx-forum-article-1.2239812

Jeff Nichols, CUNY professor and parent activist in Change the Stakes, responded:

Here's how I understand this issue.

Standardized tests can be marginally useful to educators only in the following circumstances:
The educator administers the test to the child. The educator knows both the child and the circumstances under which the test was given. The educator then looks in detail at the test, and in the context of the totality of his understanding of the child, determines whether the test yields any insights into challenges the child might be facing. The educator then makes decisions about the child's needs based on the his overall understanding of the child, informed partly by test results if that seems appropriate.

No advance "percentage" can be assigned to the test score as part of that judgment, both because the score might be completely worthless and because no one can form a professional judgment parsing out the factors that go into it by exact percentages!

Therefore, it is not acceptable to attach ANY stakes to the tests. Test scores could be part of a formal recommendation by an educator that a child needs extra help. But even beneficial consequences should never flow from test scores alone.

About evaluations: as an ambitious professional, I do not accept the principle that an inherently invalid procedure like misusing a diagnostic tool for children to assess me short form any part of my evaluation. Do we find it acceptable to eat a meal that includes 30%, 10% or 1% inedible ingredients?

But even more importantly than the folly of using test scores in teacher evaluations, Farina seems oblivious to this fact that all of us I think are aware of: The moment you attach consequences of any kind to a standardized test, you destroy its original diagnostic function. You create an incentive to mask the problems the test is designed to reveal. Imagine there were a pill you could take before a blood test that would alter its reading of your cholesterol -- and then imagine you'd lose your job if the cholesterol number were the wrong one. Would you take the pill? How useful would the test result then be to your doctor, whose priority is to improve your health -- not tweak numbers to give a false picture of your body?

Farina says "in her opinion" 30% is not too high?

This isn't a matter of opinion. High, medium or low stakes attached to test scores are unacceptable, period.

What am I missing? Is there something Farina knows that I don't?

I think we cannot give any quarter to those who would lower the percentage of test scores in teacher evaluations and call it a day.

Jeff