School Leadership for Student Achievement: A Survey and Quasi-Experimental Analysis of Leadership in Florida
Purpose: Using surveys and an existing state longitudinal database, this project will estimate the effects of school leadership on student achievement in a large sample of schools in the state of Florida.
Project Activities: Researchers will identify Florida schools that have recently changed principals through telephone interviews with school district staff. Surveys will be sent to teachers, principals, and other school leaders at 300 schools. Survey data and data from the Florida Education Data Warehouse (FEDW) will be collected and used to establish baseline estimates of the degree to which principals and schools influence student achievement net of student factors. Researchers will examine associations between reported leadership practices and principal value-added estimates, and will also study mediating variables such as school climate.
Products: This project will identify avenues through which school leadership influences student achievement, and will also provide data on mediating factors that influence the efficacy of leadership.
Setting: The study is set in Florida due to the availability of a statewide education data warehouse. Surveys will be sent to 300 Florida schools that have recently changed principals.
Population: The population of interest is school leaders, including principals and other school staff. The sample will include 300 principals, 3,000 other school leaders (assistant principal, instructional coach, mentor teacher, department chair, etc) and 9,750 teachers.
Research Design and Methods: The project will use data from both the Florida database and the survey. First, the data will be used to estimate the contribution of schools to student achievement (school value added) and the portion of that contribution that is attributable to principals (principal value added). Second, researchers will study the degree to which differences in principal and school value added are due to specific leadership practices. These first two components will identify potentially causal effects of principals and their leadership practices. Third, researchers will study the avenues through which school leadership affects student achievement, especially how the results of principals' personnel practices—that is, which specific teachers they hire and which teachers choose to stay—affects achievement outcomes. The final component of the study will examine how school context influences the efficacy of leadership and the roles of the mediating factors.
Control Condition: There is no control condition.
Key Measures: Florida Education Data Warehouse (FEDW) will provide a rich set of information on nearly all individual students and individual school principals in Florida. The surveys of teachers, principals, and other school leaders will measure leadership strategies in four domains including: communicating goals, impacting social structures, connecting teachers to knowledge about instruction, and human resource practices. The surveys will also measure key mediating factors in two broad domains: faculty capacity for instruction, and school culture and climate. Principals will be surveyed twice, yielding longitudinal measures.
Data Analytic Strategy: The first analytical goal of the study will be to establish baseline estimates of the degree to which principals and schools influence student achievement net of student factors, by estimating value-added measures for each school in Florida, studying the explained variation and the distribution of school value added, and producing estimates of principal value added. Researchers can distinguish between school and principal value added because administrative data from the state of Florida documents the movement of principals in and out of, and between Florida schools. Thus principal value-added measures are based on estimates of the degree to which school value-added estimates change after a school changes principals. Models will include various combinations of fixed and random effects to address sensitivity. Researchers will then analyze the contribution of principal leadership behaviors reported in the surveys to principal value-added measures. Structural equation models (SEM) will be used to examine the role of mediators such as hiring practices and school climate.
Journal article, monograph, or newsletter
Camburn, E.M., and Han, S.W. (2017). Teachers' Professional Learning Experiences and their Engagement in Reflective Practice: A Replication Study. School Effectiveness and School Improvement: 1–28.
Le, T, Bolt, D., Camburn, E.M., Goff, P., and Rohe, K. (2017). Latent Factors in Student–Teacher Interaction Factor Analysis. Journal of Educational and Behavioral Statistics, 42(2): 115–144.