Understanding Concerns, Sentiments and Disparities of Population Groups During the COVID-19 Pandemic: A Cross-Sectional Study Based on Large-Scale Twitter Mining

This article was originally published here

J Med Internet Res. 2021 Feb 18. doi: 10.2196/26482. Online ahead of print.


BACKGROUND: Since the outbreak of the COVID-19 pandemic in late 2019, its far-reaching impacts have been globally witnessed across all aspects of human life, such as health, economy, politics and education. Such widely penetrating impacts cast significant and profound burdens on all population groups, incurring varied concerns and sentiments among them.

OBJECTIVE: This study aims to discern the concerns, sentiments and disparities of various population groups during the COVID-19 pandemic through a cross-sectional study conducted via large-scale Twitter mining.

METHODS: The study conducted in this work consists of three steps: first, tweets posted during the pandemic are collected and preprocessed on a large scale; then the key population attributes, concerns, sentiments and emotions are extracted via a collection of natural language processing procedures; at last, multiple analyses are conducted to reveal concerns, sentiments and disparities of population groups during the pandemic. Overall, this study implements a quick, effective and economical approach for analyzing population-level disparities during public health events. The source code developed in this study is released for free public use at https://github.com/cyzhang87/EmulatedQuestionnaireOnTwitter.

RESULTS: 1,015,655 original English tweets posted between August 7 to 12, 2020, were acquired and analyzed to obtain the following results. Organizations are significantly more concerned about COVID-19 (OR=3.48 (95%CI: 3.39-3.58)) and have more ‘Fear’ and ‘Depression’ emotions than individuals. Females are less concerned about COVID-19 (OR=0.73 (95%CI: 0.71-0.75)) and have less ‘Fear’ and ‘Depression’ emotions than males. Among all age groups (below eighteen, nineteen to twenty-nine, thirty to thirty-nine, and above forty years old), the attention ORs of COVID-19, ‘Fear’ and ‘Depression’ increase significantly with age. It is worth noting that not all females pay less attention to COVID-19 than males. In the age group of above forty years old, females are more concerned than males, especially in the economic and education topics. Besides, males above forty and below eighteen years old are the least positive. Lastly, in all sentiment analyses, the sentiment polarities over political topics are always the lowest among the five concern topics across all population groups.

CONCLUSIONS: Through large-scale Twitter mining, this study reveals that meaningful differences regarding concerns and sentiments on COVID-19 related topics exist among population groups during the study period. Therefore, specialized and varied attention and supports are in need for different population groups. In addition, the efficient analysis method implemented by our publicly released code can be utilized to dynamically track the evolution of each population group during the pandemic or any other major events for better informed public health research and intervention.

PMID:33617460 | DOI:10.2196/26482