Value-added estimates, as one component of teacher evaluations, offer important information for teachers and administrators, and their reliability is roughly on par with performance estimates used in other fields, a bunch of high-powered scholars assert in a report released today by the Washington-based Brookings Institution.
While an imperfect measure of teacher effectiveness, the correlation of year-to-year value-added estimates of teacher effectiveness is similar to predictive measures for informing high-stakes decisions in other fields, the report states. Examples include using SAT scores to determine college entrance, mortality rates and patient volume as quality measures for surgeons and hospitals, and batting averages as a gauge for selecting baseball talent.
Statistical predictions in those fields are imprecise, too, but they're able to predict larger differences across providers than other measures and so are used, the authors write.
The report is notable for standing in contrast with several others issued recently, such as this one from the Economic Policy Institute. The EPI report, along with another from the National Academy of Sciences, frowns on using value-added in decisions involving teachers, due to the imprecision of the measures.
The Brookings report was written by Susanna Loeb of Stanford, who's done a lot of work using value-added to analyze teacher mobility; Dan Goldhaber and Doug Staiger, experts on value-added; and Steven Glazerman, of Mathematica Policy Research, among others.
The authors contend that many discussions of value-added conflate the value of the data with potentially "objectionable" uses, such as printing individual ratings in the newspaper. But evaluations that include a variety of measures benefit from incorporating all the best sources of information. Teachers, their mentors, and principals should have access to them, the authors write.
They go on to note that value-added is criticized by scholars, but often without consideration of the fact that there aren't alternatives that produce better estimates. And finally, much of the discussion is concerned with value-added estimates that might misidentify effective teachers as ineffective. But they don't often address the inverse problem—teachers identified as effective who in reality are not.
Bottom line from the authors:
"When teacher evaluation that incorporates value-added is compared against an abstract ideal, it can easily be found wanting in that it provides only a fuzzy signal. But when it is compared to performance information in other fields in other fields or to evaluations of teachers based on other sources of information, it looks respectable and appears to provide the best signal we've got."
The debate reminds me of a question I posed on this blog quite a while back, when teacher evaluation was first starting to come to the forefront in policy discussions: What degree of error in a teacher-evaluation system should be deemed acceptable?
Like all the analyses on the controversial topic of value-added, this one opens itself up to some criticism. For instance, it doesn't confront the very real concern teachers have about whether the use of value-added will increase the pressure on them to do test prep, or otherwise negatively inform curricular decisions.
Still, this is an interesting and important addition to the debate about value-added,especially now that there are academics and scholars lined up on both sides of the issue.
Now, let's hear your response.