Date of Graduation


Document Type


Degree Name

Doctor of Philosophy in Education Policy (PhD)

Degree Level



Education Reform


Gema Zamarro

Committee Member

Gary Ritter

Second Committee Member

Patrick Wolf


Implementation Fidelity, Policy Implementation, Racial Disproportionalities, Student Discipline


In the United States, exclusionary discipline, including out-of-school suspension (OSS) and expulsion, is disproportionately administered to students of color and special education students. Exclusionary discipline is associated with lower academic achievement and higher risk of dropout, grade retention, and involvement in the juvenile justice system, but there is little causal evidence on this topic.

This dissertation reports on three analyses on school discipline, using administrative data from Arkansas public schools. The first study estimates racial disproportionalities in the use of exclusionary discipline. Controlling for reported behavior and student characteristics, my co-author and I find that Black students are 2.4 times as likely as White students to receive exclusionary discipline. Within schools, this race-based gap is insignificant, suggesting the gap is driven by differences across schools.

In the second study, my co-authors and I use student fixed effects within dynamic panel data models to attempt to estimate a causal effect of exclusionary discipline on student test scores in the following year. Counterintuitively, we find almost zero evidence of negative effects, suggesting that reductions in OSS, without additional supports or interventions, will likely not improve student achievement.

The third study examines the implementation and outcomes of a statewide policy eliminating OSS as a consequence for truancy. That study tests which school-level factors predict policy compliance and whether there were any policy-related changes in test scores, attendance, chronic absenteeism, truancy rates, or other student disciplinary outcomes. I find that compliance was low in high-minority, high-discipline schools, and there was no policy-related change in school-level test scores, attendance, and chronic absenteeism. Reports of truancy and the use of OSS for truancy declined following the new policy, but part of this result may be due to changes in how schools report discipline.

In summary, my research indicates that real disparities in exclusionary discipline exist, but they are primarily between schools, and the negative impacts of the exclusionary discipline on its own may be minimal. What is likely more important is focusing on prevention and building positive school climate, rather than setting high-level policies and hoping schools will rise to the challenge.