Skip to main content

News & Media

News & Media Front Page

Study Finds Children Carry Implicit Bias Towards Peers Who Are Overweight

Participants more likely to think positively of children who are healthy weight

Study participants first viewed an image of a child engaging in an activity, followed by a neutral, abstract image. They then rated the abstract image “good” or “bad.”

Contact

Duke Health News Duke Health News
919-660-1306

DURHAM, N.C. – Even children as young as 9 years old can carry a prejudice against their peers who are overweight, according to a new study led by Duke Health researchers. They might not even realize they feel this way.

The study, published online June 23 in the journal Pediatrics, sheds important insight into implicit weight bias in children and could serve as a starting point for further studies on the subject.

“When children are stigmatized for being overweight, it can cause further weight gain and other health consequences,” said Asheley Skinner, Ph.D., associate professor of medicine at Duke University School of Medicine, member in the Duke Clinical Research Institute and the study’s lead author. “Given that, we felt that it was important to determine if we could identify unconscious attitudes towards weight in this 9-to-11 age group.” 

The study included 114 children. The authors used a research method that primes subjects by using quick flashes of a series of carefully selected images that depict the study topic, juxtaposed with neutral images. Skinner said the study is the first to use this method, known as the Affect Misattribution Procedure, to consider attitudes that children have about weight.

The image used to measure bias showed a child engaging in an activity that was either related or unrelated to weight, like running or participating in class. Photographs alternately showed children involved in the activities who were either healthy weight or overweight, but were otherwise similar. 

Study participants first viewed an image of a child engaging in an activity, followed by a neutral, abstract image. They then rated the abstract image “good” or “bad.”

After study participants saw these photographs, they were shown a neutral, abstract image, and asked to rate it as “good” or “bad.” On average, participants rated 64 percent of the abstract images preceded by images of children who were healthy weight as “good,” but did so for only 59 percent of second images preceded by children who were overweight, a difference that was statistically significant among the participants.

In the absence of bias, the study’s authors suggest that there would have been equal percentages of "good" ratings, since the preceding images of children performing the same activity were similar except for weight. The gap between "good" ratings therefore represented an overall implicit bias rate of 5 percent against the children who were overweight, according to the authors.

Previous studies by Keith Payne, Ph.D., of the University of North Carolina at Chapel Hill (UNC) and a co-author on the current study, have measured similar rates of implicit racial bias among adults. In this study, participants who were healthy weight had higher rates of implicit bias than those who were underweight or overweight.

“The main takeaway is that weight bias and a preference for thin people appears to start at a fairly young age,” Skinner said. “Knowing that this kind of implicit bias exists among children this age allows us to potentially be more aware of the unintended ways that children who are overweight might be stigmatized.”

“Implicit bias is important because it may underlie decisions among children about friendship, participation on sports teams and even bullying,” added senior author Eliana Perrin, M.D., professor of pediatrics at Duke and the study's co-principal investigator. “It’s essential to raise awareness about this kind of bias because it can have real consequences for children.”

Perrin said the work benefited from an interdisciplinary approach, with a research team that included physicians and social scientists, including co-principal investigator Andrew Perrin, Ph.D., a sociologist at UNC.

In addition to Skinner, E. Perrin, Payne and A. Perrin, co-authors include Janna Howard, Abigail Panter, Anna Bardone-Cone, Cynthia M. Bulik, and Michael J. Steiner.

The study was supported by the National Institutes of Health/National Cancer Institute (R24CA186212). The authors report no financial conflicts.

###

News & Media Front Page