Does teacher certification matter? Evaluating the evidence

The authors respond to Dan Goldhaber and Dominic Brewer's article in the Summer 2000 issue of Educational Evaluation and Policy Analysis that claimed from an analysis of NELS teacher and student data that teacher certification has little bearing on student achievement. Goldhaber and Brewer found strong and consistent evidence that, as compared with students whose teachers are uncertified, students achieve at higher levels in mathematics when they have teachers who hold standard certification in mathematics. (The same was true to a somewhat lesser extent in science.) However they emphasized their finding that, "Contrary to conventional wisdom, mathematics and science [students] who have teachers with emergency credentials do no worse than students whose teachers have standard teaching credentials" and suggested that certification be abandoned. This article critiques the methodological grounding for this finding and presents additional data on the characteristics of the small sub-sample of teachers in NELS data base who held temporary and emergency credentials. It finds that most of these teachers have qualifications resembling those of teachers with standard certification, and that those who have more education training appear to do better in producing student achievement. It also reviews the literature on teacher education and certification as the basis for evaluating Goldhaber and Brewer's claim that states should eliminate certification requirements and proposes additional research that would illuminate how teacher education and certification operate-and could better operate-to enable teachers to succeed in their work.