I have always been suspicious of consensus documents with multiple signatures. They have the patina of authority but usually produce pabulum. So it was a pleasant surprise to read the latest consensus document from the Brookings Institution on “the important role of value added” when assessing teacher performance.
The merit pay movement depends on value-added testing to make its case that teacher salaries need to be based on performance, not experience. But the opposition to value-added testing is potent, as teacher unions have been fighting virtually any kind of performance-based evaluations of their members. Testing has also been attacked relentlessly in education schools across the country. Just last Thursday, I attended a seminar at Boston College where David Kirp championed his forthcoming book, Kids First. Not a person in the room could find anything good to say about using tests to hold schools or teachers accountable. The well-known Princeton economist Jesse Rothstein has done much to advance his career by purporting to show that value-added measures get it wrong.
Given this kind of organized opposition to testing, one would expect Brookings, the very definition of a middle-of-the-road think tank, to produce a consensus document that balances the arguments carefully, saying “yes, testing has some value,” but “no, there are a lot of problems with it.” But, on the contrary, Grover Whitehurst, director of the Brown Center at Brookings, has put a group together that not only makes a clear case for value-added measurement of teacher performance but states that case simply and persuasively.
The group admits that test-based measures of teacher effectiveness correlate, on average, for any given teacher, from one year to the next, at no better than 0.35, well below the 0.90 correlation one would in principle like to have. But the report then points out that the “between-season correlation in batting averages for professional baseball players is 0.36.” In the best line of the report, they say: “Ask any manager of a baseball team whether he considers a player’s batting average from the previous year in decisions about the present year.” If general managers, in making salary decisions, look beyond years of experience, so should school districts. Thanks, Brookings.
-Paul E. Peterson