Interview: Taking a Look at Pre-K Teachers’ Implicit Biases

In September, the Yale Child Study Center released a study on whether early educators’ implicit biases impacted their behavioral expectations and recommendations for pre-K expulsions and suspensions. This interview with Dr. Walter Gilliam, lead study author and associate professor of child psychiatry and psychology at the Yale Child Study Center, helps to clarify the study’s results and implications.

In the study, researchers chose early care and education teachers at random at a national professional conference and asked them to watch a video clip of a typical classroom. The teachers were told to find the challenging behaviors in the video clip, although none were present. When challenging behaviors were expected, the teachers tended to observe the black children more closely, especially the black boys. The researchers were able to capture this finding by using eye tracking technology. 

The second part of the study asked teachers to read vignettes that described very challenging child behaviors. The vignettes were randomized and used child names to imply the identity of the child (a black boy, black girl, white boy, or white girl). Some of the vignettes included more information about the child’s family environment; these were also randomized. When family background was included and the teacher’s race matched that of the child, teachers tended to lower the severity rating of the child’s behavior.   

In the early childhood policy sphere, there is a growing call to end suspension and expulsion in the early years and grades because this harsh disciplinary practice has negative consequences for the children and families who experience it. Suspension and expulsion are implemented disparately, impacting far more young black children than their peers.

Dr. Gilliam, who directs The Edward Zigler Center in Child Development and Social Policy, explains that state and district policymakers should think about ending the ineffective practice of suspension and expulsion in the early years and grades, but he also emphasizes that it is essential that teachers are given support with an evidence-based behavior intervention, such as early childhood mental health consultation. He also highlights the importance of the school-family connection. Teachers must not only know the children and families they serve, but must understand and empathize with them in order to have a positive relationship that prioritizes a child’s well-being above all else.

Q: How did you define implicit bias in the study?

We defined implicit bias as the biases that we have toward the way we view things, and in this case, challenging behaviors from children that may be based in things that we aren’t even aware of as individuals. We assumed that many teachers who might be biased to believe that children who are black or children who are boys are more likely to have challenging behaviors or are more likely to require a suspension or expulsion may not necessarily be aware that they view it this way.

We knew that we needed to have a methodology that didn’t require people to be consciously aware of how they might be viewing challenging behaviors. We used eye tracking technology to be able to get at that. And to try to understand the nature of it a little bit better, we used deception studies in the form of a vignette study where we gave teachers a description of a challenging behavior, but we without their knowledge randomized them to different conditions to either receive that information to believe that that child was either a black boy or white boy or white girl or black girl.

Q: Can you go into more detail about how implicit bias among the preschool teachers manifested itself in both the eye tracking and vignette parts of the study?

Without any background information, you can think of the vignettes as a bit of intervention, some people got the intervention, in terms of knowing the background information and some people didn’t. When you take that background information out and just look at the challenging behaviors, the white teachers rated the black children’s severity level of this behavior as pretty low. As a matter of fact, we wrote the vignette (which was the description of the challenging behavior) to be a pretty significant behavior problem. This is a child who is hitting other children, scratching, and leaving marks on other kids, and is largely defiant of the teacher. So, it is a little surprising when the white teachers say that this is pretty normal. It is not normal behavior. Unless, of course, you feel that it is fairly normal for black kids. That’s what we were picking up on.

The same information given to the black teachers [resulted] in the total opposite. They rated the black children as being very, very, very high in severity. They rated the white children as being kind of high too, but nowhere near as high as they rated the black children.

Now, the thing to bear in mind is the same teachers that did the eye tracking were the same teachers that did the vignette study. When we put it together in that light, the eye tracking tells us that teachers, regardless of race, tend to expect black children to misbehave. If [teachers] are told that someone is going to misbehave, go find the misbehavior, then they are going to look more at the black child and particularly the black boy. We know that that’s true regardless of the race of the teacher.

The question then with the vignette study is what is the nature of that expectation. What seems to be the case is that white teachers expect that the black child is going to misbehave. Therefore, they hold them to a lower standard because that’s what black kids do. Whereas, the black teachers also expect that the child is going to misbehave. They hold them to a really high standard because not on my watch and I’m going to get you to quit doing that before it becomes too much of a problem later on in high school.

That doesn’t negate the notion that teachers regardless of their race tend to expect black children to misbehave more, but it does shed a little bit of light on how we respond to that. White teachers tend to respond to that expectation, that black children are going to misbehave, by forgiving them, maybe too much. And black teachers tend to respond to that by being fairly vigilant with them, maybe too much or maybe not.

If you expect a child to misbehave, how that impacts your behavior with that child may vary on the basis of whether you yourself are black or white.

Q: What was the most surprising finding for you? Or were the findings surprising at all?

What surprised us a little bit with the eye tracking was that after we finished we asked the teachers to take a look at these four kids who were in all of the videos. Which child do you think required the most of your attention? We really didn’t know whether the teachers would be aware of the fact that they were looking more at the black children, and more specifically at the black boy. We asked that question basically to see if the teachers were aware of that. They overwhelmingly said that they looked more at boys, and in particular the black boy. In terms of what they do, they look more at black kids, but especially that black boy. What they think that they are doing is looking more at boys, but in particular the black boy. Either way the black boy is coming on the short end of the stick.

What probably was the most surprising finding was from the second paragraph of the vignettes. In the study, we gave some of the teachers, but not all of them, a second paragraph that described the home life of the child. The purpose of that was to see that if we were to provide a description of this child’s home and community life whether or not people might think that this was explanatory of this child’s challenging behavior. Would the description elicit empathy? And, if it did elicit empathy, would it be possible that that empathy would mitigate or reduce the amount of bias that we found?

We kind of thought that it would be based on other research that’s been done on empathy. And it did if the teacher and the child were of the same race. And if the teacher and the child were not of the same race, it didn’t just not mitigate the bias, it made it stronger. And, that’s fascinating. Either it’s confirming an already existing bias or it overwhelms them. What this does tell me is that the antidote [to the bias] is not as simple as I was hoping. If only parents and teachers knew and understood each other better, maybe that would help mitigate this. I don’t think that is the case. I think they need to know and understand each other, and if they are of different races, they might need a little help understanding each other.

Q: Zooming out: Is there a policy solution to this? People have found implicit bias in other professions as well. Do teachers need more pre-service training to end the overuse of suspension and expulsion?

I would be all for more pre-service training if we had any evidence that pre-service training had any relationship to it. I don’t know if it does.

When we were looking at just expulsion in general, we found at the very beginning of our work on this 10-15 years ago, that there is no relationship whatsoever between the degrees that a teacher has or the credentials that a teacher has and the likelihood of that teacher to expel a child or suspend a child. No relationship whatsoever. We looked at that variable in every possible way you could imagine, but we saw nothing.

Then, I started interviewing teachers and what we heard was that if a teacher has nothing more than a high school diploma, which some of our preschool teachers do, the chances are that they have absolutely no training in how to handle challenging behaviors in the classroom. And if a teacher has a master’s degree in early care and education, chances are they have no training in how to handle challenging behaviors in the classroom. If the content of that training doesn’t actually address challenging behaviors, what difference does it make?

I think that it is the same thing with bias. I just don’t know if making sure that all of our teachers have a master’s degree is going to get rid of the problem. Nor am I convinced that having them take a class on cultural sensitivity is going to get rid of it because I don’t have any evidence to suggest that those classes work.

We need to know for a fact what it is that we can do that really reduces the likelihood of teachers having these kinds of biases. The fact that we haven’t really adequately studied this, I find appalling.

Q: There is still pushback in the field around the banning of suspension and expulsion in the early years and grades. Why do you think that is? Is there just a need for greater education around this issue?

I think that it’s a need for greater education. This is an important issue when you take into the account the potential role of implicit bias. Why is it that we think expelling and suspending children helps, when all the evidence suggest that it doesn’t? The only thing that I am left with is that either: 1) we haven’t adequately articulated the whole nature of the problem and the fact that there is no data to suggest that it actually helps or 2) that maybe what it is, is that it is the symptom of a workforce that’s afraid that what’s going to happen is that we are going to pass a rule that says you can’t expel or suspend and give them nothing in return that actually could help.

Expulsion and suspension doesn’t help kids. It just relieves the current teachers from having to deal with the problem any longer. And, I don’t say that critically because I can appreciate the fact that it’s really difficult to have other children in your classroom that you have to take care of and there is a child in your classroom that you feel requires some additional attention. And, there is no additional support.

We have been suggesting that we do need policies that make it much more difficult to expel and suspend children because it is way too easy in preschool. There are no rules on how to do this at all. At the same time, we need to be able to also provide teachers with some solutions, some services, or some support. So that when they aren’t able to expel and suspend, they have an alternative.

One of those solutions we have been pointing to lately is early childhood mental health consultation. It takes advantage of mental health providers that are already in the community, and just makes them available to preschool teachers and child care providers in a fairly low cost, efficient way. And it works, it’s an effective intervention. But to have the ban without the intervention is probably not going to help. In fact, it may even make things worse.

There is also a role for the parent in all of this. Any intervention, whether it’s early childhood mental health consultation or some other type of intervention, has to have a component where we make better connections between the parents and the teachers.

One of the things that I’ve never seen is a child expelled from a preschool program where the teacher and parent knew and liked one another. Building stronger relationships between the home and the provider may not solve all the problems that all of our children face, but will probably build a stronger bridge between the home and the preschool in order to give some time for those problems to be solved. It’s just so easy to expel a child, when you don’t really like the parent.

But, if you like that parent and you understand where the challenges are coming from or you’re empathic to the fact that this parent needs this child care in order to be able to go to work, it makes it so much harder to want to expel a child, which then makes it so much more likely that we’ll look for another solution.