Association for Public Policy Analysis & Management

Race-based disparities in student discipline has been documented across multiple school settings, with black students more likely to be perceived as exhibiting problem behaviors (Bates & Glick, 2013) and to face exclusionary discipline (Rocque, 2010; Skiba et al., 2014) than their white peers. More recent studies of school discipline have begun to investigate the mechanisms behind those disparities, finding evidence that student-teacher and student-administrator race congruence affects disciplinary citations (Skiba et al., 2002) and consequences (Lindsay & Hart, 2016). However, findings on the role of race congruence in discipline have not been as consistent as those around racial disparities: for example, using North Carolina data on disciplinary infractions resulting in suspensions or expulsions, Kinsler (2011) found race-based disparities existed but did not find evidence that those disparities were moderated by student-teacher or principal-teacher race congruence.

This study advances the research on racial disparities in disciplinary citations and the role of student-teacher race congruence in those disparities by exploring the probability of disciplinary infractions at the student level. Specifically, we ask:

(1) To what extent are black and white students differentially likely to receive disciplinary infractions?

1. To what extent does having more race-congruent teachers affect the likelihood and number of disciplinary infractions a student receives?

(2) To what extent are black and white students differentially likely to receive infractions for more subjective misconducts, such as disrespect or insubordination?

1. To what extent does having more race-congruent teachers affect the likelihood and number of these more subjective disciplinary infractions a student receives?

We estimate the effect of race and race congruence on number of infractions using negative binomial regression to account for the over-dispersion of the infraction count data. We also model the probability of receiving at least one infraction, conditional on teacher race, using student fixed effects linear probability models, exploiting the year-to-year variation in the racial makeup of a student’s teachers.

This study advances the research on racial disparities in disciplinary citations and the role of student-teacher race congruence in those disparities in three key ways: first, we draw on a robust dataset of infraction-level data linked to course rosters to construct a novel measure of race congruence that comprises all teachers a student works with on an average school day. We test the effect of student-teacher race congruence on disciplinary citations using this more fine-grained measure. Second, our analysis extends the conversation about student-teacher race congruence to all grade levels, where others that have linked students to teachers focus on elementary schools. Finally, we have access to all infraction-level data regardless of consequence assigned, allowing us to explore the probability of being cited for lower-level offenses that would not be included in typically available discipline data, which is limited to infractions resulting in exclusionary discipline consequences.