Disparities in charter school resources—the influence of state policy and community

Abstract

Recent findings show that students attending charter schools in the United States achieve at comparable or lower levels to those enrolled in regular public schools, perhaps due to uneven quality and disparities in the levels of resources acquired by charter schools. But little is known as to what state and local factors contribute to disparate levels of resources in the charter school sector. This article examines how local context, the charter school's organizational form, and state policies may influence material and human resources obtained by charter schools and their capacity to innovate. We find marked differences among charter schools situated in different US states in terms of teacher qualities, student-staff ratios, length of the school day, and the propensity to unionize, drawing on data from the US Schools and Staffing Survey for the 1999/2000 school year. Charter schools rely less on uncredentialed teachers in states that more tightly regulate the sector, and state spending is associated with more equal teacher salaries among charter schools within states. But the lion's share of variance in charter school resources is attributable to highly variable local contexts, not to state-level factors, especially the kinds of students served and the school's organizational form. Charter schools serving predominately black students rely on less experienced teachers who are more likely to be uncredentialed; their teachers also report more demanding working conditions and lower levels of efficacy, compared with charter teachers working in white schools. Conversion charter schools pay staff over $5100 more annually and rely much less on uncredentialed and part-time teachers than do start-up schools. We examine implications for the reproduction of unequal student achievement within the charter school sector.

Paying for School Choice: Availability Differences among Local Education Markets

“…Wilson, 2011). If school zone discontinuity according to parents' educational level, housing price, and household income has restricted many students' access to better-resourced and more diverse schools (Bayer & McMillan, 2012;Black, 1999;Bodine et al, 2008;Gibbons & Machin, 2006;Kenn, 2001;D.E. Mitchell et al, 2010;Reardon & Yun, 2001), then empowering parents to choose children' s schools with their own hands has the potential to improve overall access to education by weakening geographical advantages or disadvantages and opening up invisible boundaries between communities.…”

“…Wilson, 2012). Others have indicated irreconcilable disparities in school funding and resources between school districts, stemming from the financial structure of traditional public schools that rely on property and income taxes from the area in which they are located (Bodine et al, 2008;Kenn, 2001;D.E. Mitchell, Batie, & Mitchell, 2010;Reardon & Yun, 2001).…”

Books or Guards? Charter School Security Costs

“…Differences in school level, student characteristics (e.g., percent special needs), and purpose (e.g., alternative instruction) also confound our ability to compare charter and traditional school finances (Bodine et al, 2008). Consider that charters are disproportionally elementary schools (U.S. Department of Education, 2011), which tend to be less costly to operate than secondary schools (Miron & Urschel, 2010).…”

Sorting schools: A computational analysis of charter school identities and stratification

“…I next estimate mixed-effects linear models (which have both fixed and random effects) in STATA 15 for dependent variables at the school level. Given that state policy (e.g., Bodine et al 2008;Finnigan 2007), school district practices (e.g., Paino 2018), and CMO structure (e.g., Furgeson et al 2012) have significant, convergent impacts on charter school practices and outcomes, ignoring the nesting of charter schools (level-1) in states, school districts, and CMOs (level-2) by pooling all observations would violate the assumption of independence required for ordinary least squares (OLS) regression and thus bias the results. 7 Furthermore, these are cross-classified data in that each CMO could have schools in multiple states, and each state could be home to multiple CMOs.…”