JMIR Publications

ABSTRACT

Background:

The COVID-19 pandemic caused a critical public health crisis worldwide, and policymakers are making lockdown policies to control the virus. However, there is a noticeable increase in aggressive social behaviors that threaten social stability. Lockdown measures might negatively affect mental health and lead to an increase in aggressive emotions. Discovering the relationship between lockdown and increased aggression is crucial to formulating appropriate policies that address these adverse societal effects. We employed Natural Language Processing (NLP) technology and Internet data to investigate the social and emotional impact of lockdowns.

Objective:

This research aims to understand the relationship between lockdown and increased aggression by using NLP technology to analyze three kinds of aggressive emotions: anger, offensive language, and hate speech, in spatiotemporal ranges of tweets in the USA.

Methods:

We conducted a longitudinal internet study of 11,455 Twitter users by analyzing the aggressive emotions in 1,281,362 tweets they posted from 2019 to 2020. We selected three common aggressive emotions on the internet as the subject of analysis: anger, offensive language, and hate speech. For emotion classification, we trained the BERT model to analyze the percentage of aggressive tweets in every state and every week. Then, we used the difference in differences estimation to measure the impact of lockdown status on increasing aggressive tweets. Since most other independent factors that might affect the result, such as seasonal and regional factors, have been ruled out by time and state fixed effects, a significant result in this difference in differences analysis not only can indicate a concrete positive correlation but also strongly suggest a causal relationship.

Results:

In the first 6 months of lockdown, aggression levels in all users increased compared to 2019. Notably, users under lockdown demonstrate greater levels of aggression than those not under lockdown. Our difference in differences estimation discovered a statistically significant positive correlation between lockdown and increased aggression (Anger P< 0.002, Offensive P< 0.000, Hate P < 0.005). It can be inferred from such results that there exist causal relations.

Conclusions:

Understanding the relationship between lockdown and aggression can help policymakers address the personal and societal impacts of lockdown. Applying NLP technology and utilizing big data on social media can provide crucial and timely information for this effort.