The evolution of “grading” teachers

Why we need to reconsider the “value-added” that value-added measures supposedly lend to teacher evaluation.

Joe Nocera’s New York Times op-ed, “How to Grade a Teacher” (June 16, 2015), reflects the growing recognition that our nation’s current teacher evaluation systems are not working. He rightfully acknowledges that the vast majority of teachers want high-quality feedback, but the system imposed by Governor Cuomo of New York, fueled by the federal government’s Race to the Top initiative, is actually failing students and those who teach them.

However, Mr. Nocera’s piece could have gone a bit deeper. He notes that it’s more appropriate to judge teachers based on the growth in their students’ test scores—not just one or two absolute measures. Mr. Nocera points out that teacher unions believe that value-added (or growth) measures are “simplistic” and “unfair.” But serious academics recognize these statistical facts of teacher evaluation “life.” The work of Jesse Rothstein is very instructive—surfacing the reality that value-added results for state tests are poorly correlated (0.5 or less) with alternative assessments, meaning that many teachers whose value-added for one test is low are in fact quite effective when judged by other factors.

When asked why he believes politicians are so focused on value-added, Mr. Rothstein noted:

There’s been a tendency to prioritize statistical computations based on student test scores, because all you need is one statistician and the test score data. Classroom observation requires having lots of people to sit in the back of lots and lots of classrooms and make judgments.

However, top-performing nations like Singapore know how to assess teachers without rigidly using test scores. Master teachers, working closely with principals, lead the way. They use multiple measures—none of which include standardized test scores. They create time for the most accomplished teachers to mentor and assess their colleagues. And principals insist that teachers assist them in conducting evaluations.

Singaporean principals and teachers use professional judgment in determining who is effective and who is not. Perhaps President Obama, Secretary of Education Arne Duncan, and Governor Cuomo might take a cue from these professionals. They might also consider reading Mr. Nocera’s op-ed and then exploring Mr. Rothstein’s work. But most importantly, they should pay attention to the wisdom of our nation’s three million teachers to determine if our approach to grading teachers is evolving in ways that make sense for improving teaching and learning.

Related categories:
  • TriciaEbner

    Simple vs. Meaningful

    The value-added evaluation system is problematic in so many ways. It gets even more complicated when student performance on standardized testing becomes a key component of the “report cards” the state issues for districts and individual school buildings. A year ago I was sitting with several other parents as we watched our children playing coach-pitch baseball, and one of the moms commented that she was frustrated her child, a third-grader, was having to take the state’s reading assessment again in the spring, even though her child had a good score on the fall administration of the same test. She didn’t understand it. I did: the building and district were aiming for an improved report card because the higher every third grader’s reading score on that assessment, whether taken once or taken twice, the better the report card indicator. 

    So we’re testing kids more, and we teachers aren’t getting the kind of feedback that could help us really stretch and improve our practice. 

    It’s very frustrating. And we teachers need to speak up about it.

  • BarnettBerry

    Keep speaking out!

    Tricia. Yet another example of the deepening failure of how we are using test scores and VAM. I will be the first to agree with the so-called “reformers” that we need good evidence on which kids are doing well or not.  (I am not labeled as a reformer, however, because I have critiqued the use of VAM as currently used.) But the research from scholars in the field is clear. Check out YET another huge statistical problem with VAM. We keep plugging away at supporting teachers who can and will speak out with their solutions in seeking accuracy and accountability as well as sanity and significance for the kids and their parents as well as those who know them best (i.e., you and millions of other teachers)

    –bb

  • Douglas W. Green, EdD

    VALUE ADDED

    Sorry to say that the value added  system  of evaluating teachers is nowhere near valid. For extensive support see my summary of Rethinking Value-Added Models in Education: Critical Perspective on Tests and Assessment-Based Accountability by Audrey Amrein-Beardsley http://bit.ly/1zaiI6Y

  • Wanda Porter, Ph. D.

    Teacher Evaluations

    This new information on evaluating teachers revealed important changes.  It leads to more questions as to how and why certain entities determine criteria.  The standard measures to evaluate teachers show continuaton but paradoxes. Yet this new research demonstrates why alterations are necessary.

  • Kathleen L. Gallagher

    Teacher Evaluation

    I think that value-added models are important, but they don't tell the whole story. First, they are dependent on the tests actually measuring knowledge our profession agrees is important for children to attain. If standards are our agreed-upon target, then the validty of the tests are dependent on all teachers being equally committed to ensuring students have access to standards-based instruction. Then, you have to factor in the quality of that instruction, which I believe is the intent of teacher evaluation. I think that if we agree that standards-based instruction has value, then value-added models only make sense if we add a level to the analysis that includes an instructional-quality measure for each classroom. This would be dependent on a robust observation protocol that teachers undersand and value. If we can correlate growth in student achievement (The difference between students' achievement last year and student achievement this year) with meausres of high quality instruction that teachers understand and value, then we might be able to develop teacher evaluation systems that acutally have the effect of improving teaching and learning.