Paying for high- and low-quality teaching

The extensive research on teacher quality has led to two conclusions. First, there are large and significant differences among teachers in terms of their capacity to improve student achievement. Second, these differences are not captured by common measures of teacher qualifications (E.A. Hanushek, Teacher quality, in: L.T. Izumi, W.M. Evers (Eds.), Teacher Quality, Hoover Press, Palo Alto, CA, 2002, pp. 1–12). We present the argument that in order to improve teacher quality, one must focus on teacher performance. By creating teaching standards and performance rubrics, rating teachers against those standards, and then employing a multivariate multilevel mixed statistical model to attribute students’ achievement gains to teachers, we both designed and provided preliminary validation evidence for a system to judge teachers based on performance. Our results demonstrate that teaching performance as defined by our standards and rubrics is highly predictive of student academic progress across the elementary grades. Implications for our research to improve teaching through implementing performance pay systems are discussed.  2003 Elsevier Ltd. All rights reserved. JEL classification: I20; I21; C10; C11; J24; J33; J38