Political Emotional Message Sharing

We replicate a successful arm of the intervention by Athey et al. (2023) and apply it to political content from TikTok. In a field experiment by Athey et al. (2023) using text message courses in Kenya, emotion-based interventions reduced impulsive sharing of misinformation. The emotional course also outperformed reasoning-based approaches. While these findings suggest the effectivness of emotion-based interventions, there is a gap in applying such strategies to political content, especially on platforms like TikTok, where emotionally charged videos are widespread. Political videos often exploit emotions like fear, anger, or superiority, leading to impulsive sharing and rapid misinformation spread. TikTok’s short-form, engaging format amplifies this issue, as users may share content without fully considering its accuracy. Given these insights, we propose testing an emotion-based intervention tailored to political video sharing on TikTok. This will educate users about emotional manipulation in political content and provide tools to recognize and resist these impulses. We note an important difference. Athey et al. (2023) have control over the content of the online posts they provide to respondents, allowing them to explicitly target misinformation and emotional content by design. In contrast, we use actual online content from political parties. As a result, we do not specifically target misinformation but instead use a random sample of videos and code them based on the emotional language used in the videos.