Revealing Unfairness in social media contributors’ attention to vulnerable urban areas during disasters

Abstract

Emerging influential contributors (EICs) on social media platforms—meteorologists, news reporters, journalists, and employees of public sector agencies—play a pivotal role in disaster public information and warning by consistently disseminating disaster situational information public. However, the fairness in Emerging Influential Contributors' (EICs') attention to different urban areas when posting disaster situational information has not been examined. It is unknown whether vulnerable communities where low-income, minority, and older individuals live are omitted from the content of posts made by EICs. The study used tweets from Houston, Texas, related to Hurricane Harvey to test the proposed approaches and examine the existence of bias in the content of social media posts made by EICs. This study first formulated the problem by identifying its correspondence with the well-explored areas of fairness assessment and then proposed two fairness assessment methods for evaluation of fair information dissemination on social media. This study's findings show that biases exist in the amount of attention given to areas in which vulnerable populations in the content of posts made by EICs on Twitter to inform about disaster situations. Such biases negatively affect the representativeness of social media content disseminated by EICs to inform emergency response and relief efforts. This study's outcomes address important knowledge gaps related to fairness in disaster informatics and information dissemination biases in online social networks during crises.