Panel Paper:
Do School Finance Reforms Close Achievement Gaps? Evidence from the American States, 1990-2016
Saturday, November 10, 2018
Coolidge - Mezz Level (Marriott Wardman Park)
*Names in bold indicate Presenter
The debate over the effects of school funding on student achievement extends back at least to the issuing of the Coleman Report in 1966. Originally commissioned by the Civil Rights Act of 1964, the authors of the Coleman Report surveyed over 600,000 students, teachers, and principals. The report found large achievement gaps between racial subgroups yet did not find large gaps in funding (Coleman et al. 1966).
In response to the Coleman report, one of the centerpieces of President Lyndon Johnson’s War on Poverty was the passage of the Elementary and Secondary Act (ESEA) of 1965. Title I of the ESEA was designed to target resources to schools serving large populations of low-income students. The impacts of the ESEA have been studied extensively by educational researchers and have been found to be modest at best. There is very little evidence of achievement gaps closing significantly due to the enhanced funding provided by ESEA, either immediately following its passage (Stickney and Plunkett 1983; Mullin and Summers 1983) or in more recent years (Matsudaira, Hosek, and Walsh 2012; van der Klaauw 2008).
Despite these findings questioning the relationship between enhanced funding and student performance, many states began enacting school funding mechanisms to direct additional resources to high-poverty schools. By 2004, 22 states had adopted funding enhancements for poorer schools (Hanushek and Lindseth 2009).
One funding enhancement method commonly adopted by state legislatures is the use of education funding weights for socioeconomically disadvantaged students and English learners. For example, in 2013-14 California passed the Local Control Funding Formula (LCFF) that among other funding reforms, established a 20 percent supplement to the base per-pupil funding for English learners (ELs), socioeconomically disadvantaged students (SDs), and foster youth. The LCFF also established new concentration grants that provided a 50% enhancement of base-pupil funding if the percentage of ELs, SDs, and foster youth exceeds 55% of the school’s enrollment.
In this paper, we examine whether the implementation of funding weights and concentration grants by state legislatures have been successful at a) raising the achievement levels of socioeconomically disadvantaged students and English learners, and b) shrinking the achievement gap between ELs and non-ELs and between SDs and non-SDs.
Our research examines National Assessment of Educational Progress (NAEP) test scores in math and reading for 4th and 8th graders for each state since 1990. Our key independent variables come from EdBuild, a nonprofit organization that focuses on state funding of schools. We develop an index using the EdBuild data that captures the size of the funding weights and concentration grants employed by the various states. Additional control variables for each state include a variety of population and economic data, including population, education levels, proportions of racial minorities, levels of poverty, state GDP and unemployment levels, region, percent urban, and overall levels of education funding for each state.
To control for unmeasured and unidentified variables, we adopt an identification strategy using two-way fixed effects models to control for fixed effects at both the unit and time levels.
In response to the Coleman report, one of the centerpieces of President Lyndon Johnson’s War on Poverty was the passage of the Elementary and Secondary Act (ESEA) of 1965. Title I of the ESEA was designed to target resources to schools serving large populations of low-income students. The impacts of the ESEA have been studied extensively by educational researchers and have been found to be modest at best. There is very little evidence of achievement gaps closing significantly due to the enhanced funding provided by ESEA, either immediately following its passage (Stickney and Plunkett 1983; Mullin and Summers 1983) or in more recent years (Matsudaira, Hosek, and Walsh 2012; van der Klaauw 2008).
Despite these findings questioning the relationship between enhanced funding and student performance, many states began enacting school funding mechanisms to direct additional resources to high-poverty schools. By 2004, 22 states had adopted funding enhancements for poorer schools (Hanushek and Lindseth 2009).
One funding enhancement method commonly adopted by state legislatures is the use of education funding weights for socioeconomically disadvantaged students and English learners. For example, in 2013-14 California passed the Local Control Funding Formula (LCFF) that among other funding reforms, established a 20 percent supplement to the base per-pupil funding for English learners (ELs), socioeconomically disadvantaged students (SDs), and foster youth. The LCFF also established new concentration grants that provided a 50% enhancement of base-pupil funding if the percentage of ELs, SDs, and foster youth exceeds 55% of the school’s enrollment.
In this paper, we examine whether the implementation of funding weights and concentration grants by state legislatures have been successful at a) raising the achievement levels of socioeconomically disadvantaged students and English learners, and b) shrinking the achievement gap between ELs and non-ELs and between SDs and non-SDs.
Our research examines National Assessment of Educational Progress (NAEP) test scores in math and reading for 4th and 8th graders for each state since 1990. Our key independent variables come from EdBuild, a nonprofit organization that focuses on state funding of schools. We develop an index using the EdBuild data that captures the size of the funding weights and concentration grants employed by the various states. Additional control variables for each state include a variety of population and economic data, including population, education levels, proportions of racial minorities, levels of poverty, state GDP and unemployment levels, region, percent urban, and overall levels of education funding for each state.
To control for unmeasured and unidentified variables, we adopt an identification strategy using two-way fixed effects models to control for fixed effects at both the unit and time levels.
