AEFP 45th Annual Conference

Toward a Meaningful Impact through Research, Policy & Practice

March 19-21, 2020

Disparate Implementation of School Discipline and Alternative Approaches for Equitable School Discipline

Habiba Ibrahim, Saint Louis University,

Over the past few decades, schools across the country have adopted zero tolerance policies mirroring the criminal justice system’s approaches in dealing with adolescents misbehaviors (Hirschfield, 2008; Kupchik, 2010; Welch & Payne, 2010). Most of the nation’s public schools today are characterized by the presence of uniformed police officers, surveillance cameras and policies that automatically suspend students for certain behaviors. As a result of these policies school suspensions have more than doubled since the 1980s (Losen, 2011). Just in 2015–16 school year, 2.7 million students received at least one out of school suspension (U.S. Department of Education, 2018).
In addition to the sharp increase in school suspensions overall, research has documented persistent disparities in school suspensions across race, social class and disability status. Studies have consistently shown that students of racial minorities, those from low-income households and students with disabilities are overrepresented in both in-school (ISS) and out-of-school-suspensions (OSS) (Gregory, Skibal & Noguera, 2010; Hoffman, 2012; Anderson & Ritter, 2017; Ibrahim & Johnson, 2019; Furtado, Duncan, Kocher, & Nandan, 2019). For example, while Black students represent about 15.5% of the student population in the country, they received about 39% of OSS (Furtado et al, 2019) and 25% of ISS (Ibrahim and Johnson, 2019). Similarly, students with disability represent 12% of the student enrollment but accounted for 28% of arrests and referrals to law enforcement within public schools (Furtado at al, 2019).
The proposed study uses data from the past 18 years (2001-2018) from the state of Missouri Department of Elementary and Secondary Education to examine whether there are similar trends of disparities in school discipline administration across schools and its impact on academic achievement. The study will specifically examine the following questions: i) Is there association between school racial composition and the use of school suspensions (both ISS and OSS)? ii) Is the school level of poverty (percent student receiving free/reduced lunch) associated with the use of school suspensions? iii) Is school rate of suspensions associated with students’ academic achievement? This study will look at changes over time and will also consider the possibility of different patterns in urban, suburban, and rural areas.
Preliminary results of this analysis suggest similar trends as the nation in the implementation of school discipline in the state of Missouri. The school discipline data from schools across the state of Missouri (N = 2703) reveal that school suspensions are systematically higher in schools that serve larger percentages of non-White students. The trend is the same for poverty, as represented by the fraction of students who receive free and reduced lunch. These results indicate that as the percentage of non-White students in a school increases, the number of students suspended significantly increase. Out-of-school suspensions were almost five times higher in schools where less than 25% of the student population were White compared to those with over 75% White population. Similar trends were found between the wealthiest and poorest schools measured by the percentage of student population who receive free or reduced lunch. The proposed study will extend these analyses over the previous two decades and will discuss implications for policy and practice.



I appreciate the attention to this topic. Do you have data on the severity of the offenses that cause disciplinary referrals? Some of your conclusions refer to incident severity implicitly (you talk about non-major violations)--it would be really interesting to see how changes in the severity of incidents have changed with OSS rates over time, and how the composition of schools relates to the intensity of the punishment for different levels of offense.

Habiba, Good work! This is an important topic. I appreciate you studying it. The longitudinal nature of your data may open some opportunities to explore how these relationships and trends vary in response to policies. For example, the 2014 federal guidance from the DOE and DOJ pushed many districts and states to put new measures in place to try to reduce suspensions. What sorts of policies and changes to practice occurred in Missouri over this time period? Your data might let you parse out the effects of some of these shifts in disciplinary practices. I look forward to seeing this work develop! Thanks for sharing. All the best, Chris Curran

Thank you!

On your OSS regression: Can you add any controls for the nature of the school staff (e.g., principal, teaching staff)? If yes, could you interact those variables with FRPL to see if, among schools with similar FRPL, there are characteristics of the staff that lead to less discipline? On your test performance regressions: Can you control for last year’s test performance of the same cohort? Right now, I’m concerned that part of what you are picking up with OSS is variation in student background that is not accounted for by the other controls. Adding in the lagged test score would take care of that problem and would give you a cleaner estimate of th3 link between OSS and test performance.

Thank you! That is a good observation, I will consider controlling for prior test scores and see if there are any changes to the results.

Add new comment