Instructors in online courses demonstrate gender and racial bias in their interactions with students, a new study finds.
White men are more likely than all other racial and gender groups to receive teacher responses to questions and comments they posted on course message boards, according to a working paper from the Stanford Center for Education Policy Analysis.
The study only included college-level courses, but the results have broader relevance, the study’s authors write—online courses are gaining in popularity at all levels of education, including in K-12. It isn’t clear exactly how many K-12 students are taking supplemental online courses currently, but best estimates suggest that the number is around 2.7 million.
Researchers conducted the study in 124 college-level massive open online courses, or MOOCs—virtual classes accessible to anyone on the internet, regardless of university affiliation. Of the courses included in the study, 56.5 percent were STEM-focused, and 58 percent had a white male instructor.
They then posted comments in the discussion sections of these courses, registered under student profiles with names commonly associated with women and men of different racial and ethnic identities: white, black, Indian, and Chinese.
Overall, instructors responded to seven percent of all comments posted as part of the study. But they responded to more than 12 percent of all messages posted by white men—almost double the share of messages they responded to for other racial and gender groups.
Other students in the MOOCs are also able to respond to student-generated posts on discussion boards—researchers found that white women are significantly more likely to respond to posts by other white women.
Taking a closer look at teacher bias, the instructor’s identity was especially significant: White men are even more likely to get a response when the teacher is also a white man. White men also see a greater advantage when posting questions and comments with a social focus—for example, a post asking about where people in the class were from. Questions with a purely academic focus—about due dates, or how to complete an assignment—aren’t addressed by professors at statistically different rates across race and gender.
This finding suggests that implicit bias may be motivating instructors’ behavior, the researchers write. Professors may be more likely to see addressing an academic question as a contractually obligated duty, while engaging in social conversation as a choice. They may be subconsciously more comfortable and willing to forge that connection with a person from their “in-group.”
These results show that some of the patterns of bias teachers demonstrate in physical classrooms can be reproduced in virtual learning spaces, the authors say.
A long line of research demonstrates that teachers show racial and gender bias in interactions with students in brick-and-mortar schools. Some examples the study cites: White teachers are more likely to rate black students’ behavior as negative than similar behavior from white students, and teachers give more attention to, and interact more with, boys than girls in elementary school classrooms.
An online classroom is still a social space, the researchers write, and students’ identities still matter. Teacher expectations about student character and academic potential can have serious consequences, affecting students’ future achievement, as Education Week has covered.
So how can schools and educators address bias in online courses?
Instead of erasing students’ identities—for instance, making all students anonymous—the researchers recommend that architects of these virtual learning environments can use technology to intentionally create a more equitable learning space. Developers can include design features that make instructors more aware of who they favor in course interactions, and guide them to devote equitable attention to all of their students.
A version of this news article first appeared in the Digital Education blog.