Challenging Implicit Bias in Classroom Observations

Recently, all of the employees at Teachstone completed a course on identifying implicit biases in our individual thinking and our collective work. After the course, we engaged in small group discussions on how our biases impact us and the teachers and children that we serve. As our core mission is understanding and improving interactions, our discussion focused a lot on what and how we observe in a classroom. Simply put, if you have an implicit bias against a particular culture, you may misinterpret a specific behavior which can then affect how you score a classroom. 

The Role of Double Coding

In thinking about solutions, we identified double coding as one way to actively address implicit biases in the way that we observe classroom interactions. Double coding is when two observers code the same classroom at the same time (or video of the same classroom) and then compare their scores and the evidence they noted to support those scores. This process helps to ensure reliability in observations and strengthens observers’ understanding and use of the CLASS measure. Although double coding has often been used as a way to support new coders or assess the quality of data collected, it can also be a powerful tool for challenging implicit bias and promoting observer growth. 

A great example of this comes from a recent study of American Indian / Alaska Native (AI/AN) Head Start programs. To ensure that observers would be able to accurately capture cultural variations across tribal communities, the researchers provided training on ways that interactions vary culturally in AI/AN classrooms. Then observers reviewed videos that had been double coded with AI/AN program staff to provide the perspective of AIAN children and families. The double coding process consisted of 2-4 people watching the videos, comparing observation notes, and coming to a consensus on both scoring and cultural nuances in the video. After completing this group review, another group provided a cross-review. Both review teams provided information that was used as talking points for the training of observers. Following the cultural training and video discussion, the observers coded a live classroom together and discussed their evidence, scoring, and the ways they had grown in their understanding of culture and CLASS observation. 

Some of the key differences noted through this process center on the understanding that AI/AN and other Indigenous children are more likely to learn through cycles of quiet observation, reflection, and joining activities with high levels of collaboration often marked by heavy reliance on non-verbal communication. For someone not familiar with AI/AN cultures, a classroom with children learning through this cycle might be interpreted as disengaged. However, careful reflection through double coding can provide a very different interpretation: that the children are highly engaged in the learning process through this cycle. Note that centering the observation on the children’s experiences, not the observer’s reaction to the experience, often helps counteract implicit bias in the coding process, and double coding is a powerful mechanism for moving away from the observer’s reaction toward the children’s experiences. Because of this process, observers were better able to understand and capture teacher-child interactions in tribal settings. 

Using Double Coding to Address Implicit Bias

• Use this process only for observer growth. The point is not to shame or use as a measure of observer performance, but to support observers’ abilities to capture classroom interactions appropriately. 
• Encourage all of your observers to participate. Everyone has implicit biases, whether they recognize them or not, and everyone needs support in addressing them.
  • Check out the Implicit Association Tests created by researchers from Harvard if you need a way to help people begin identifying implicit biases.   
• Create an atmosphere of safety and openness so that each participant is able to share their perspective and hear from other perspectives. 
• Ensure each pair or group of people has representatives from the culture you are going to observe. Allowing people to experience coding with people from a variety of perspectives can help to address different forms of bias (e.g., race, culture, ability, etc.) if all parties are truly open and listening to one another. 
• Before coding, provide guidelines for the process. Ensure that everyone understands the purpose of the exercise, the definitions and examples of implicit bias, and the boundaries to keep all discussions professional and respectful. It may be helpful to create group norms or ways of interacting that everyone agrees to follow. 
• Provide background information about the culture being observed. This should include some general information as well as cultural ways of interacting in the classroom and valued learning styles. Such information is important for participants both from that culture and from outside of the culture, as implicit bias can be internalized from the dominant culture. 
• Provide guiding questions for observers to use in their discussion. Examples include: 
  • How was that interaction different from what you have experienced in the past?
  • What was your own reaction to specific interactions? Was it positive or negative? Why? 
  • How does your own cultural background influence the way you interpret the interactions you just saw?
  • Was that interaction effective for all of the children in the group? Why or why not? 
  • How did the children experience that interaction? What evidence do you have to support your understanding of the children’s experiences in that interaction?

Know that this list is only the beginning. It is also important for observers to engage in both professional development on implicit bias and training that promotes the understanding of how the cultural context of the classroom shapes the learning experience for the children. Challenging and changing an implicit bias is hard. It can often take years of intentional work. However, if we are truly going to make an impact for each and every child, we must take this step. 

Guest Blogger: The post was co-authored by Jessica Barnes-Najor, Ph.D. Jessica is a developmental psychologist and director for Community Partnerships in University Outreach and Engagement at Michigan State University. She is also a co-PI for the Tribal Early Childhood Research Center (TRC), partnering with tribal Head Start, Home Visitation, and Child Care grantees to promote community-engaged research and enhance early childhood program evaluation and research-to-practice activities across the nation.