Persistent racial and disability disparities in school discipline remain a critical concern in U.S. education, yet limited research has examined how these patterns manifest differently across charter and traditional public schools. Specifically in the state of Texas, state policies have historically strongly influenced the use of exclusionary practices in both public traditional and charter schools for several years. This study analyzes eight years (2016–2023) of Texas district-level data on exclusionary disciplinary actions, including in-school suspension (ISS), out-of-school suspension (OSS), expulsions, and alternative education placements, to compare disciplinary practices between charter and non-charter schools. Using two-way fixed-effects negative binomial regression on district-year panel data (2015–2022), we analyzed exclusionary disciplinary disparities. We examined the effects of race, socioeconomic status, and special education status on discipline rates and explored interactions between school type and student demographics. Findings revealed charter schools imposed substantially higher expulsion rates than traditional schools, regardless of student race, and showed unexpected “charter penalties” for Asian students, who faced significantly elevated OSS and ISS rates compared to peers in traditional schools. Conversely, charter schools generally reported lower rates of ISS and OSS for White students, suggesting selective protective effects. Patterns for alternative placements, such as DAEP and JJAEP, varied and were sometimes contrary to trends observed in suspensions. Results underscore the complexity of discipline disparities and highlight the need for policy reforms to ensure equitable disciplinary practices across school types. Implications for policy, practice, and future research are discussed.
No takes yet. Share an insight, caveat, or question.
McCree et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: