Association for Public Policy Analysis & Management

State education officials play a crucial role as the primary implementers of education policy. Officials including the Chief State School Officer (CSSO) and State Board of Education (SBE) members are responsible for education standards, school accountability, curricula, and other policies. States select these education officials using a variety of methods (e.g. gubernatorial appointment, SBE nomination, legislative nomination, or election). States where the governor has the power to appoint both the CSSO and all of the SBE members have centralized governance structures, whereas states where the governor has neither power have decentralized governance structures.

Governance structures influence student outcomes through shaping the context in which state education officials operate (Lynn, Hill & Heinrich, 2002). Officials in states with centralized governance structures are accountable to the governor. These officials face political dynamics that incentivize the investment of resources into large changes to policy. Conversely, education officials in states with decentralized governance structures are responsive to the broader and more diverse interests of voters or legislators. This creates an incentive for small and iterative reforms to education policies. Prior research suggests that the centralization of governance structures will influence education policymaking processes (Skocpol & Amenta, 1986; Stevenson & Baker, 1991). As a result, I would expect Gubernatorial appointees to pursue different education policies with a higher level of implementation fidelity than education officials selected by other mechanisms.

The recent devolution of education policy calls attention to the importance of state education governance structures. Previous research suggests a possible link between state centralization and student outcomes (Manna, 2013). The study asks the question: to what extent are standardized test scores higher for students in states with gubernatorially appointed education officials, taking into account other relevant features of the state education system?

This study uses student level data from the National Assessment of Educational Progress (NAEP) Long-Term Trend Assessment (LTT). The data includes 12 years of data from 1971 to 2008. Analyses were conducted using 6 datasets (Reading and Math; 4^th, 8^th, and 12^th grade). The LTT NAEP data include a rich set of student level covariates. I also control for a variety of state characteristics (political and economic) in addition to state education policies.

I exploit state changes in education governance structures to estimate their relationship with student outcomes. This approach allows me to control for time invariant characteristics of states including education policies. I also control for time variant state education policies that were adopted at the same time as changes to governance structures. Descriptively, states where the governor can appoint both the CSSO and the SBE members have higher average test scores than states in which the governor has neither power. Regression results suggest a positive and significant relationship between gubernatorially appointed officials and student outcomes. The effect of centralized state governance structures (gubernatorial appointment) on math test scores is small (about a tenth of a standard deviation) but significant. This suggests that states with centralized governance structures influence the policymaking process in a manner that is positively related to student achievement.