Latest News

Studies Highlight Complexities of Using Value-Added Measures

As a number of states begin evaluating teachers’ effectiveness based on changes in their students’ test scores, academic research is raising more questions about such “value-added” models. …

[V]alue-added models can be messy. They can be “noisy.” They can be easily misconstrued. What are policymakers and education leaders to do? …

[S]ome districts and states already have policies in place that require them to use value-added models to help make high-stakes decisions about firing, tenure, and pay.

Asked what advice he would offer policymakers who have adopted or are considering adopting such measures, [Morgan] Polikoff of the University of Southern California replied: “I’d think about a way to design a system that takes each measure of effective teaching on its own terms, rather than a system that forces multiple, loosely-related measures into a single index for the purposes of making a decision,”

“Whatever system they choose, they should take it slow, thoughtfully study how this is all working (or not), and not be overly prescriptive,” he added.