Social Media Reports of Anti-Asian Hate during the COVID-19 Pandemic: Incident Types, Victim Characteristics, and Emerging Themes

Background and Purpose: Online spaces have become sites of contention, reflecting both hostility and the lived experiences of those affected by violence. Social media coverage of violence against Asian Americans has attracted academic attention because it presents a broader range of perspectives, including those that diverge from dominant narratives in traditional media (Lee & Jang, 2021). This study examines anti-Asian hate discourse on X (formerly Twitter) by analyzing text data depicting anti-Asian hate incidents (AAHIs). This study addresses three key questions: (1) How are incident types characterized? (2) What are the common victim characteristics? (3) What themes emerge from the unstructured text data?

Methods: This research was conducted in partnership with The Asian American Foundation (TAAF). The dataset includes 9,844 posts collected between January 2020 and May 2022. Posts were selected if the perpetrator explicitly expressed anti-Asian animus during the act or if contextual evidence strongly indicated that the victim was targeted based on their Asian identity. A computer-mediated semantic network analysis examined the thematic structure of anti-Asian hate discourse (Nandwani & Verma, 2021; Smith & Humphreys, 2006). Additionally, a quantitative content analysis was conducted through systematic manual coding of text data using a structured framework. This approach captured contextual subtleties and underlying meanings, enriching the initial findings (Krippendorff, 2018).

Results: Regarding incident types, most incidents involved verbal harassment (76.5%), with discrimination (44.2%) and physical assault (14.6%) being less frequent, and that Asian communities, particularly Chinese individuals, were the primary targets of these attacks. In terms of victim characteristics, females were the most mentioned victims (29%), nearly twice as often as males (15.6%), though half of the posts did not specify gender. Most posts referenced victims as "Asian in general" (64.2%), with Chinese individuals most frequently referenced among specific ethnic groups. The most frequently mentioned terms, such as “Asian,” “racist,” and “Chinese,” highlighted the racial and ethnic focus of these discussions, while five distinct themes emerged: (1) everyday discrimination, (2) explicit racism, (3) racial identity, (4) pandemic-induced xenophobia, and (5) geopolitical influences.

Conclusions and Implications: The findings about victim characteristics suggests that anti-Asian sentiment during the pandemic was often overtly directed, but also disproportionately affected certain demographic groups (e.g., gender, ethnicity, etc), revealing patterns in victim reporting. Importantly, the study highlights how pervasive racialized discourse on social media both mirrors and amplifies societal prejudices, calling for proactive responses from policymakers and community leaders to address its systemic and psychological impacts. The results underscore that social workers can play a critical role in promoting bystander intervention strategies within online spaces and ensuring that victims of racialized hate have access to culturally responsive, trauma-informed mental health services. Additionally, leveraging social media analytics can support real-time tracking of racialized violence, enabling social workers to contribute to evidence-based policy development and community advocacy. By fostering interdisciplinary collaboration between technology platforms, law enforcement, and community organization, social workers are uniquely positioned to advocate for more effective interventions that address both the online and offline consequences of hate.