Pandemic Fatigue: An Analysis of Twitter Users’ Sentiments Against the COVID-19 in Indonesia

Abstract

Indonesian society is undergoing a shift of behavioral patterns towards the coronavirus (COVID) 19 indicating the symptoms of the pandemic fatigue phenomenon. Pandemic fatigue is defined as a gradual demotivation to adhere to recommended protective behaviors. Pandemic fatigue might reduce the effectiveness of health protocols and accelerate the spread of the virus. This study aims to examine the pandemic fatigue sentiment of Indonesian twitter-users chronologically and the factors causing the development of pandemic fatigue sentiment. The method of this study includes digital ethnographic theory using sentiment analysis based on the Valence Aware Dictionary for Sentiment Reasoning (VADER) algorithm and topic modelling using Latent Dirichlet Allocation (LDA) analysis. The results showed a pattern of sentiment degression towards health protocols indicating the pandemic fatigue phenomenon. The factors causing the sentiment degression were influenced by three themes: (1) public criticism of the government's efforts to handle the spread of COVID-19, (2) experience in implementing health protocols, and (3) statements that against the government’s efforts to handle the spread of COVID-19. Based on the results of sentiment analysis and topic modelling, this study presents a public policy design referencing the World Health Organization (WHO) framework for community reinvigoration in the midst of pandemic fatigue that could be used for the government to undertake broader reforms to public health and social care for Indonesian society.