Clare and I (Clive) have often argued against the use of “value-added measures” (VAMs) to assess teachers and teacher educators, measures that rely exclusively on standardized test scores. Others (e.g., David Berliner, Diane Ravitch) have taken a similar stand.
In the May 2016 issue of the Educational Researcher, opposition to VAMs receives dramatic support from Steven Klees of the University of Maryland. In a letter to ER, Klees welcomes a recent AERA Statement about how difficult it is to assess teachers using VAMs. However, he goes on to say that it’s not only difficult, it’s impossible! He notes that “dozens, perhaps hundreds, of variables” influence test scores, and hence misattributing cause is not only a “significant risk,” it is “rampant and inherent” in the use of VAMs. He concludes:
“The bottom line is that regardless of technical sophistication, the use of VAM is never ‘accurate, reliable, and valid’ and will never yield ‘rigorously supported inferences’” (p. 267).
In my view, even if we give some weight to test scores, it is imperative to supplement them with other considerations: e.g., the judgment of teachers and their colleagues about good teaching, opinions of students about their teachers’ effectiveness, theories about effective pedagogy. Effective teaching is so complex there can be no quick fix in assessing it.
It will be interesting to see how the education research and policy communities respond to Klees’s extraordinary claim, given that VAMs are the latest great hope for the reform of teaching and teacher education.