Association for Public Policy Analysis & Management

Panel: Productivity Spillover and Knowledge Diffusion Among Employees: Evidence from School Teachers

(Education)

As in many other workplaces like schools where employees need to work in teams, it is critical to consider employee spillovers in program design, implementation and evaluation. This panel comprises four papers that examine spillover effects at different levels of an organization: formally and informally structured peer groups within schools, and schoolwide. The findings shed light on critical policy issues related to teacher placement strategies, performance assessment, and their adoption of a new policy. The first paper examines whether Teach For America (TFA)’s new strategy of placing clusters of TFA corps members in a targeted set of disadvantaged schools led to spillover effect on other teachers’ performance. The second paper specifically tests four models of peer structures to seek for optimal grouping of teachers to maximize student learning in aggregate. The findings from these two papers shed light on teacher placement strategies at school and district levels, “Talent Transfer Initiatives”, and the 50-state teacher equity strategy initiated by the U.S. Department of Education. These efforts aim to increase the equitable distribution of teacher quality across schools, particularly to ensure that low-performing schools that disproportionately serve disadvantaged students are able to recruit and retain effective teachers. Unfortunately, prior studies of these policy strategies are mainly based on a human capital explanation that effective teachers would benefit their own students’ learning, yet rarely from the perspective of re-arranging teachers to augment peer learning in schools. These two papers underscore the discussions about how to distribute teacher quality and what the impacts of different ways of re-arranging teachers are. Different from the previous two papers of estimating spillovers on students that teachers may not directly work with, the third paper studies teacher spillover effects on own students’ achievement on another subject taught by another teacher in secondary schools. This study finds consistent evidence that ignoring spillover effects in value-added models may lead to biased estimates of teacher performance, as it ignores the contribution from cross-subject teachers who teach the same group of students. This bias would threaten the fairness of any subsequent personnel management decisions based on these value-added performance measures. While the previous three papers infer spillover mechanisms without direct observations of teacher interactions, the fourth paper uses unique longitudinal network data to examine the role of the informal subgroup or clique in cultivating and distributing locally adapted and integrated knowledge, or know-how. Results indicate that the more knowledge is restricted to flow from subgroups with high know-how, the greater the systemic implementation of practices is dependent on that knowledge. This work has significant implications for building intra-organizational networks for generating and distributing knowledge to support policy implementation. Together, these four papers represent the latest conceptual and methodological development from economics, sociology and education to study productivity spillover and knowledge diffusion among employees in a high-skilled and knowledge intense workplace. Panel participants include both leading researchers and junior scholars from multiple institutions (AIR, University of Washington, Stanford, Vanderbilt, RAND, Michigan State, University of Colorado Boulder, and University of Toledo).