JMIR Publications

ABSTRACT

Background:

Social media platforms, such as Facebook, allow individuals of various identities to disseminate race- and ethnicity-related content through image-based posts. Social media and image-based posts allow racial minorities to engage in content creation to counter racial silencing and discrimination. Nevertheless, racism and stereotyping are present in social media posts. Racism on social media can have negative outcomes on both the physical and mental health of marginalized groups on social media due to psychosocial stressors caused by online messages, as well as their manifestations in the physical world. Difficulty identifying and blocking such content arises from techniques such as humor masking of stereotypical posts. Therefore, it becomes essential to understand how social media users post about racial and ethnic topics, as this can reveal common public attitudes regarding race and ethnicity.

Objective:

This study aims to describe qualitative content themes in Facebook posts with images and text regarding race and ethnicity.

Methods:

We used CrowdTangle to randomly sample 100 facebook posts each year between 2019 and 2023 (n=500). Each post included an image and text and was required to contain one or more race-related key terms. The images and corresponding text were uploaded to GitHub, and each post’s URL was uploaded to Label Studio for coding. A codebook was developed to code race and ethnicity, continuum of race-related discussions and content topics. All posts were double-coded until interrater agreement reached 80%. The remaining 20% of posts were settled by the coders meeting and reaching a complete consensus.

Results:

We found that the count data on the frequencies of posts mentioning various racial and ethnic groups from 2019 to 2023 did not display a particular pattern. Three categories constituted greater than 10%, Hispanic, multiple races and ethnicities, as well as no specific race mentioned. We found several common content themes unique to each racial and ethnic group in the posts, with the most highly represented themes throughout the sample and across several races and ethnicities being cultural pride, racism, antiracism, and U.S. politics.

Conclusions:

It can be complicated to interpret image based posts because of the subtle ways in which an image may reference race and ethnicity but does not explicitly mention it, or when there is a contradiction in the ideas portrayed in the image versus the text. Decoding this process on Facebook can help researchers boost the positive impacts and reduce the harmful effects of racism on social media.