*연구과제명: Examining the Nuances of Consumer Racial Bias: An Analysis of Positive Consumer Response to Racial Representation on Instagram
-연구 기관: 우버 프레이트 우버 테크놀로지스, 네바다 대학교 마케팅학부, 고려대학교 심리학과 (연구팀)
-저자: Joon H. Ro, Jae‑Eun Namkoong, James M. Leonhardt, Eunsoo Choi
Abstract
While racial bias in the media has faced intense scrutiny, less attention has been paid to consumers’ personal biases when interacting with media content. Building on the literature on aversive racism, we investigate how consumers’ racial biases manifest in their responses to social media content based on race and other facial characteristics of individuals featured in brand posts. We employed a face-detection algorithm to process over 70,000 image-based posts made by ten major media brands on Instagram, extracting information from nearly 50,000 faces to examine their infuence on social media engagement. On the surface, racial bias appeared to favor Blacks—featuring more Blacks in posts was associated with greater post-liking. However, a more refned analysis revealed nuanced patterns of racial bias. Low-threat facial features (higher values of femi- ninity, age, and smiling) were key moderators in predicting positive response to posts featuring more Blacks (vs. other races), indicating the presence of racial bias associating Blacks with aggression. Additionally, aligning with mainstream beauty standards was increasingly important for predicting positive consumer response when posts included more Blacks. Finally, featuring more racial minorities, especially Blacks, in the center of images reduced positive response to a greater extent.
Keywords Racial bias · Stereotypes · Facial recognition · Social media · Consumer engagement
*연구과제명: Examining the Nuances of Consumer Racial Bias: An Analysis of Positive Consumer Response to Racial Representation on Instagram
-연구 기관: 우버 프레이트 우버 테크놀로지스, 네바다 대학교 마케팅학부, 고려대학교 심리학과 (연구팀)
-저자: Joon H. Ro, Jae‑Eun Namkoong, James M. Leonhardt, Eunsoo Choi
Abstract
While racial bias in the media has faced intense scrutiny, less attention has been paid to consumers’ personal biases when interacting with media content. Building on the literature on aversive racism, we investigate how consumers’ racial biases manifest in their responses to social media content based on race and other facial characteristics of individuals featured in brand posts. We employed a face-detection algorithm to process over 70,000 image-based posts made by ten major media brands on Instagram, extracting information from nearly 50,000 faces to examine their infuence on social media engagement. On the surface, racial bias appeared to favor Blacks—featuring more Blacks in posts was associated with greater post-liking. However, a more refned analysis revealed nuanced patterns of racial bias. Low-threat facial features (higher values of femi- ninity, age, and smiling) were key moderators in predicting positive response to posts featuring more Blacks (vs. other races), indicating the presence of racial bias associating Blacks with aggression. Additionally, aligning with mainstream beauty standards was increasingly important for predicting positive consumer response when posts included more Blacks. Finally, featuring more racial minorities, especially Blacks, in the center of images reduced positive response to a greater extent.
Keywords Racial bias · Stereotypes · Facial recognition · Social media · Consumer engagement