Not All Discipline Disparities May Be The Result Of Implicit Bias

Over the past few months, we have heard a lot about discipline disparities by race/ethnicity and gender -- disparities that begin in the earliest years of schooling. According to the Civil Rights Data Collection Project by the U.S. Department of Education's Office for Civil Rights, "black students represent 18% of preschool enrollment, but 42% of preschool students suspended once and 48% of students suspended more than once." It also found that "boys receive more than three out of four out-of-school preschool suspensions."

This focus on student discipline disparities has also drawn attention to the research on implicit bias -- the idea that we all harbor unconscious attitudes that tend to favor individuals from some groups (whites, males, those judged to be good looking, etc.), and that disadvantage people from other groups (people of color, women, ethnic minorities, etc.). The concept of implicit bias suggests that good or bad behavior is often in the eye of the beholder, and disparities in disciplinary outcomes (e.g., suspensions and expulsions) may be influenced by unconscious stereotypes.

Part of me is very glad that we are finally having this conversation. Acknowledging the existence and consequences of subtle, implicit forms of prejudice is an important and necessary first step toward mitigating their effects and advancing toward fairness -- see my implicit bias series here. But it sometimes seems that the discipline and the implicit bias conversations are one and the same, and this concerns me for two reasons.

First, implicit attitudes of teachers and administrators are likely to affect school-related areas other than discipline. A case can be made for the contribution of implicit bias to many types of decisions about students, aside from the obvious, such as decisions about grading, tracking, evaluations and disciplinary actions. Indeed, all kinds of student behaviors have the potential of being misinterpreted due to biases and stereotypes. For example, a student who keeps to herself in the classroom could be perceived as shy or reserved, but also as sullen, uninterested or disengaged. This is often a subjective call and most likely mediated by the student's race, ethnicity and gender (among other factors, including context).

Limiting our understanding and application of implicit bias to discipline-related actions and outcomes won't be enough to create equity in our classrooms. Implicit biases can also be responsible for more subtle, unconscious actions and attitudes, such as paying more attention and giving more praise to white kids, while inadvertently dismissing the contributions of students of color -- I discuss some of these issues here.

The second reason for my concern is that the existence of implicit biases does not mean that there do not also exist actual differences in behavior. For example, it is possible that teachers and administrators are sometimes too quick to see the behavior of black males as problematic. But it is also possible that students from different cultural backgrounds are more likely to act according to non-mainstream cultural norms. In fact these two explanations can co-exist. Moreover, it is impossible to determine empirically the extent to which biases versus actual differences in behavior are responsible for the outcomes we observe. What's important is that both are factors to consider if we are serious about addressing the issue of discipline disparities by subgroup.

Consider a different issue to which the above also applies: Women's underrepresentation in science and engineering jobs. Does it reflect women's choices (based on their preferences and skills), employers' conscious and unconscious preferences, or a combination of both? The answer is probably all of the above; indeed, things are actually more complex than the questions suggest and I am actually much more interested in understanding is the various mechanisms at play and then devising strategies that target any (or ideally all) of the identified factors.

I would be equally supportive of an intervention that consists of offering training for employers on the mechanisms of unconscious biases and, say, a program that encourages young women to pursue careers in STEM by connecting them to women mentors.  I feel the exact same way about discipline.

Do kids from different socio-demographic backgrounds behave differently or are teachers and administrators' unconsciously interpreting kids' behavior in a biased manner -- or both? It's hard to know the exact extent to which each factor contributes to the problem. But the bottom line is that the end result is indeed problematic. There is no (moral) justification for why children of color are overrepresented in suspensions and expulsions. In other words, there is something wrong with the outcome regardless of the exact combination of factors that are causing it.

This means that, if we care about discipline disparities, we must worry about the possibility that teachers' and administrators' harbor unconscious biases. But we must also worry about the possibilities that teachers and administrators need additional training in culturally responsive pedagogy and that some kids from difficult backgrounds will need additional help and support to improve their behavior.

I would like to conclude by highlighting one strategy that can potentially support the latter in the context of preschool education.

A few weeks ago, I wrote about a paper that looks at the link between language and behavior in early childhood. We know that oral language is the foundation for reading and learning -- but could oral language development also reduce young children's behavioral problems? Before this study, most research on the subject had only identified a co-occurrence of language delays and behavioral maladjustment. However, this paper, written by Marc Bornstein, Chun-Shin Hahn, and Joan Suwalsky (2013), was able to establish that “language competencies in early childhood keep behavioral adjustment problems at bay" -- and not the other way around. Thus, one way to 'protect' young children from developing behavioral problems later on is supporting their oral language development early.

Although this sort of strategy may not appear to address early discipline disparities directly, it is likely to help mitigate the problem. We know that poor children (who are also disproportionately black and brown) start kindergarten already behind; Bornstein and colleagues' work suggests that by providing extra support for these kids' oral language development, we would not only be helping them academically, but also we would be "inoculating" them from developing behavioral problems down the road.

We now know that 'zero tolerance' policies are not a viable option, so then what? If we are serious about addressing discipline issues, particularly racial disparities, we must provide adequate support both to children and to the adults who work with them. Ultimately, it's not easy to tell what exactly is causing the patterns we observe -- thus, interventions and strategies that address any or all of the contributing factors are a step in the right direction.

- Esther Quintero