Grading teachers on value-added measures falls short
Op-ed/UCLA Today Commentary by fifteen education scholars from UCLA, CSU Northridge, Stanford, Duke and the University of Wisconsin-Madison in response to the recent Los Angeles Times article, "Who's Teaching L.A.'s Kids?"
Research indicates that well-constructed VA estimates, especially those based on averages over at least three years, can provide valuable information about teachers’ contributions to student learning. Research also shows that teacher effectiveness varies considerably, even within schools. The public needs to know, however, that VA modeling is not an exact enough methodology to justify identifying individual teachers as “effective” or “ineffective” based solely on a particular VA score. The Times authors themselves cite a recent National Academy of Sciences report saying that value-added measures are promising but not ready for high stakes contexts. Providing a public database of teachers’ VA estimates constitutes a high stakes context both because it could affect teachers’ reputations in their school communities and because it could harm students’ educational outcomes if parents base decisions solely on this information. Thus, publishing such a database is scientifically premature. (more…)