Monday, October 11, 2010

The Limits of Value-Added Models of Assessment

Diane Ravitch (and others) have a lot to say about the limits of the value-added model (VAM).  For those in need of a tune-up on VAM, here is a nice synopsis from the Value Added Research Center.

One of the strengths of the VAM is that is suggests that there are number of variables that influence how "learning" takes place: schools, teacher prep programs, families, communities, teacher:student ratios.  What has to happen with these varables so that children are ready to learn and teachers are supported?

Taken in isolation, VAM has limitations.   For example, effective teachers may be miscast as ineffective, and therefore unworthy of employment, recognition and/or bonuses.*  Perhaps worse -- ineffective teachers could be identified as effective due to stand-alone standardized test scores.

And what if Oak Tree A is transplanted to Garden B?  Or how much time would Gardener B need to be effective in Garden A?  Think about switching from teaching 11th grade physics to 8th grade general science, for example.  Or switching from 5th graders to 3rd graders.  How much time, planning and coaching would you need to get to know the students, the 3rd grade teaching team, the curricula, and the families?  How long would it take for you to be "effective?"

We can also ask whether standardized tests scores are an adequate measure of learning, no matter how you adjust the numbers.

* For more on teacher bonuses, "Teacher Bonuses Don't Improve Test Scores."

No comments:

Post a Comment