Evaluating the Feasibility and Effectiveness of an Instagram Intervention to Combat Hate: A Randomized Controlled Trial

• In Event: Using Experimental Methods to Explore Criminological Topics

Abstract

Anti-Asian hate and hate crimes have surged since the onset of the COVID-19 pandemic. Exposure to hateful content has been associated with various negative social and behavioral outcomes, including heightened anxiety, fear, depression, defensive and hyper-vigilant attitudes, and, in some cases, engagement in aggressive or violent behavior. This study leverages large language models and prompt engineering techniques to develop an Instagram-based intervention aimed at encouraging reporting and counteractions against hate speech and behavior while mitigating the adverse effects of experiencing or witnessing such incidents. A randomized controlled trial conducted among a sample of Asian American and Pacific Islander college students at a large Southeastern university (N = 187) evaluates the effectiveness of this digital intervention. Preliminary findings suggest that the intervention effectively engages participants and shows promise in achieving its intended goals. Insights from this study can inform the development of similar digital interventions targeting other harmful behaviors, such as cyberbullying and online phishing.

Authors

• Beidi Dong, George Mason University

• Talley Bettens, George Mason University

• Xiaoquan Zhao, George Mason University

• Cher Chen, George Mason University

• Yoolee Kim, George Mason University