Research

Across the United States, suspension bans have become a popular policy response to address excessive and inequitable use of suspension in schools. However, there is little research that examines what strategies school staff employ when suspension is no longer permitted. I examine the effect of New York City's suspension ban on the use of a potential unintended substitute for suspension: special education classification. Using a dosage difference-in-differences strategy, I find that the ban induced an increase in disability classifications at high risk for classroom exclusion. I show that, on average, students with these classifications in schools with high pre-policy reliance on suspension experienced large declines in test scores, whereas general education students experienced slight test score improvements. Notably, I show that these declines are not due to new, ban-induced classifications actively harming student achievement. These results underscore the importance of considering unintended consequences and vulnerable groups when employing a seemingly "costless" and popular policy lever to reduce schools' reliance on suspension.

We study the effects of local labor market shocks and state unemployment insurance (UI) policies on student discipline outcomes in U.S. public schools. By leveraging data on both school-level disciplinary incidence and local, firm-level layoffs across 23 states, we find that ---on average--- exposure to a layoff event has a limited impact on discipline, but this average effect masks important heterogeneity across states with varying levels of maximum UI benefits. At the lowest level of UI benefits ($265), out-of-school suspensions increase by 4.5% from its mean. However, as UI benefits increase, these negative effects are mitigated. Specifically, we estimate that on average, between $480 and $600 in maximum weekly benefits is able to effectively nullify the impact of layoffs on suspensions and expulsions. These effects of layoffs on discipline are driven by large impacts on Black and male students, and we further document that layoff shocks can increase the Black-White gap in out-of-school suspensions when UI benefit levels are low, particularly in predominantly White schools. 

This study evaluates the 2015 suspension policy reform in New York City that co-occurred with a pilot program for restorative practice funding. In analyzing the joint and independent impacts of these programs on suspension rates, on average and for historically marginalized groups, the data show the suspension restrictions reduced suspension rates, leaving little room for RPs to further reduce suspension rates. Additionally, the restrictions reduced suspensions for low-level offenses and appeared to have spillover effects onto suspensions for more severe behaviors. RP funding did not provide differential spillover effects in this context. The suspension restrictions reduced racial and ethnic disparities in suspension rates, but RPs did not appear to provide any further reductions for Black or Hispanic students. Although there might not be an observable direct reduction in suspension rates attributable solely to RP funding, the provision of and training in a direct alternative to suspension in the wake of restrictions likely reduced any unintended consequences that may ensue when only restrictions are present.

Policy regarding clock and solar time remains unresolved. Within the past decade, policymakers have pushed to implement permanent daylight savings time (DST) at the state and federal levels, despite evidence showing its negative effects on health outcomes. We study the effects of reduced sleep induced by late sunset time on student discipline. We use a boundary regression discontinuity design and national school-level data, leveraging variation created in sunset time generated by time zone boundaries. We find that one-hour-later sunset results in increases in student disciplinary incidence. These impacts are driven by male students and students in early and middle grades. These results add to the body of evidence that documents the detriments of time policies like permanent DST and policies that negatively influence adolescent sleep patterns, including early school start times.

This study provides a descriptive analysis of police intervention as a response to student behavior in New York City public schools. We find that between the 2016/17 and 2021/22 academic years, arrests and juvenile referrals decreased while non-detainment-based and psychiatric police interventions increased. However, Black students, especially those enrolled in schools located in predominantly White police precincts experiencing a shrinking White student population, experienced disproportionate rates of arrests, juvenile referrals, and police-involved psychiatric interventions. Schools serving more Black students experienced higher rates of interventions relative to schools with fewer Black students, but these higher rates of intervention are not explained by differences in observable student behavior and characteristics. Instead, differences in teacher characteristics and resources contribute to the excess use of police interventions in predominantly Black schools. 

This study estimates the impact of a common informal diversion strategy: warn and release interventions. Despite their use, there is limited evidence regarding these interventions’ impact on youth outcomes, especially within the school context. We use unique school-level data on school-based arrests and warn and release interventions from the New York Police Department. We use a difference-in-differences instrumental variables (IV-DiD) strategy, relying on policy-induced variation in rates of warn and release, to derive the causal effect of warn and release interventions. Preliminary results suggest that increasing school-level rates of warn and release interventions increases a student’s likelihood of receiving a suspension or office discipline referral. Additional estimates will provide evidence regarding warn and release interventions’ impact on juvenile felony, misdemeanor, and violation offenses surrounding a school, as well as impacts on student attendance and test scores. Estimates will also consider heterogeneity by race, class, and school context.

Men considerably outnumber women in physics, engineering, and computer science (PECS) majors, with a recent male-to-female ratio of ~4:1, a stark contrast to the near parity in other science, technology, engineering, and mathematics (STEM) disciplines. This gender disparity in PECS carries wide-reaching implications for equity, innovation, and scientific advancement. Analyzing a near-census of >34 million bachelor’s degrees awarded in the US from 2002 to 2022, supplemented with two nationally representative datasets, we provide the first comprehensive study of the gender gap in PECS across higher-education institutions. Institutions serving students with higher math SAT scores have made greater strides in closing PECS gender gaps—both in initial recruitment and retention—whereas those serving students with lower math SAT scores are increasingly struggling. Our findings highlight institutions as strategic leverage points for targeted interventions, particularly to benefit women of color who remain underrepresented even where gender balance is improving.

City of Marysville Residential Analysis (at the Western Washington University Center for Economic and Business Research, with Samantha Adams)

2020 Peer Cities Report (at the Western Washington University Center for Economic and Business Research, with Audrey Barber)

Point Roberts Economic Profile (at the Western Washington University Center for Economic and Business Research, with Allison Rucker)