The Adoption of Statewide Corporal Punishment Bans in U.S. Public Schools, 1970-2023

Background: Despite efforts at the federal level to end corporal punishment in schools across the United States, the question of whether to use corporal punishment (e.g., paddling) as a disciplinary strategy in schools remains a state and local decision. Approximately one-third of states either allow or do not universally prohibit school corporal punishment (Allison et al., 2023). Consequently, during the 2017-2018 school year, nearly 70,000 students experienced corporal punishment in public schools (U.S. Department of Education, 2022). School corporal punishment is associated with lower academic achievement, feeling isolated during high school, and depressive symptoms during young adulthood (Gershoff et al., 2019). In order to advance the understanding of the contexts in which states have banned corporal punishment in public schools, this study examined how state- and regional-level contextual factors predict statewide ban timing.

Method: A novel panel dataset was constructed in which state-year was the unit of analysis (N = 1,649). Each year was included from 1970 (when only one state had a ban) to 2023 (a year in which three states achieved a ban). States exited the panel the year after ban adoption. Statewide ban timing was the year in which a state banned school corporal punishment either through legislation, through regulation, or by all districts separately adopting a ban. Statewide ban timing was verified using state government websites and news archives. Focusing on historical, political, sociodemographic, and religious contexts, state- and regional-level contextual predictor variables were drawn from sources such as the U.S. Census Bureau, the Association of Religion Data Archives, and the General Social Survey. Event history analysis was used to predict statewide ban timing. Coefficients were converted to average marginal effects (AMEs) and multiplied by 100 to the percentage point scale.

Results: States with Republican-controlled legislatures had a 5 percentage point lower probability than states with Democrat-controlled legislatures to ban corporal punishment in a given year (p < .01); however, the political party of the governor was not associated with ban adoption. States with a higher percentage of school-aged children enrolled in private school (AME = -.19, p < .05) or not enrolled in regular school (e.g., homeschooled, AME = -1.04, p < .01) also had a lower probability of a ban. Each percentage point increase in the total population that was evangelical was associated with a .34 lower probability (p < .001) of ban adoption. Regional-level attitudes about spanking were not associated with ban adoption. During state-years with split control between Republicans and Democrats in the legislature, having a larger percentage of legislators who were women was associated with a higher probability of a ban (AME = .52, p < .01).

Conclusion and Implications: The probability of banning corporal punishment in schools varies by political, sociodemographic, and religious contexts across states and time. Nevertheless, bans are possible in all states. Policy practitioners, educators, and advocates who wish to achieve school corporal punishment bans in all states can benefit from an understanding of the contexts that predict ban adoption at the state level.