Online course instructors are 94 percent more likely to respond to discussion forum comments made by students with names connoting that they are white and male compared to other race-gender groups, according to research published in March.
The experimental study, “Bias in Online Classes: Evidence from a Field Experiment,” was conducted by researchers from Stanford University, the University of California, Irvine, and Vanderbilt University who posted comments in 124 online course discussion forums in 2014.
Researchers placed eight randomly assigned comments in each forum. Each comment was made by fictitious students whose names were indicative of a specific race (white, black, Indian or Chinese) and gender (male or female).
Authors found that instructors' response rate was almost double for white males than for any other race-gender combination.
“That’s a really dramatic finding because we think the discussion forums are important settings where student engagement and learning motivation gets supported and catalyzed,” Thomas Dee, co-author of the report and director of the Center for Education Policy Analysis at Stanford University, said. “If there's this type of bias in who's getting the engagement with the instructor, that, in a way, just reinforces patterns of inequity that many of us care about.”
The overall instructor response rate was 7 percent, but comments made by researchers' white male students received a 12 percent response rate. Indian males were the least likely to receive a response from instructors.
“These learning environments are going to damage particular groups of students over others,” Dee said. “And many of us have this mindset that education is supposed to be this equalizer that provides opportunity to all. So my concern is that that is going to go unrealized.”
Online classes have become increasingly common in education, a trend that served as one reason that motivated researchers to conduct the study, Dee said.
From 2012 to 2016, the number of graduate and undergraduate students taking at least one distance education course, which includes online classes, increased by 17.2 percent, according to a 2018 report from the Babson Survey Research Group. In the fall of 2016, 31.6 percent of all enrolled students in post-secondary institutions were taking at least one distance education course, while 14.9 percent were taking exclusively distance courses, the report states.
Dee said that most design thinking for online classes has been focused on learners as individuals but hasn’t acknowledged that learning environments are also social settings. As a result, some of the dynamics that exist in conventional K-12 classrooms might be relevant in online classrooms, he added.
“And in particular, the design of most online classes is asynchronous. The students aren’t there at the same time; they watch recorded lectures at different times," Dee said. "In that kind of environment, the discussion forum plays a really critical role."
Despite the inequity, Dee noted that that it's easier to redesign online learning environments to address issues compared to brick-and-mortar classrooms because there is greater control over the way students engage with each other and the instructor.
Researchers recommended that the concerns raised in the study be included in the design process for online learning environments. Without making any specific recommendations, some examples of possible solutions researchers offered include anonymizing student identities to mitigate bias and providing appropriate professional training for online instructors.
Dee said he hopes this research will catalyze some interest among those who design online learning spaces to consider issues of equity and fairness, and that it will broaden the conversation around the designing of effective online classrooms.