Winters' Work on VAM Adds Value to Colorado Educator Effectiveness Policy

I’m guessing that 2012 has been fairly busy for education policy researcher Dr. Marcus Winters. He started with the launch of his book Teachers Matter, which included visiting Denver as the first-ever speaker in the Independence Institute’s Brown Bag Lunch series. And he since has published work on school innovation and productivity, and the effects of Florida’s reading retention policy.

This week he has released a Manhattan Institute report that should help inform Colorado’s ongoing implementation of Senate Bill 191, Transforming Tenure: Using Value-Added Modeling to Identify Effective Teachers. Winters looked at key Florida teacher data to help determine the effectiveness of value-added measures (VAM) in educator evaluations.

For anyone who follows education policy, the first significant finding of his report should come as no surprise at all. Some critics complain that VAM is too unreliable because they say achievement scores are determined far more by the circumstances of students than the teacher’s impact. Well, value-added certainly isn’t perfect, but Winters shows that a new teacher’s VAM score is far more predictive than whether or not they have a master’s degree of their ability to raise student test scores during their fifth year in the classroom.

(Of course, cynics might say it’s not a fair comparison — given the unanimous findings that virtually no connection exists between teacher master’s degrees and improved student learning. But then again, the comparison in Winters’ analysis did result in a blowout.)

What about the policy implications? The Manhattan Institute analysis tends to confirm some of the key features in Colorado’s 2010 educator effectiveness law: in particular, tying tenure to regular evaluations even after the job protections have been achieved, and using multiple years of VAM data as a major (but not the only) piece in rating teacher effectiveness.

Given the limitations of existing tests, VAM certainly isn’t the be-all and end-all of teacher evaluations — though it holds far more weight in my book than vague admonitions to push students into social activism, a la Denver Public Schools. But VAM is a much more reliable indicator of quality instruction than the measures commonly used now.

If he keeps contributing research with valuable policy insights, I guess I’m not too sad that Marcus Winters seems to have been so busy.