Analysis of tweets regarding psychological disorders before and during the COVID-19 pandemic: The case of Turkey

Abstract This study aimed to examine the effects of the COVID-19 pandemic on Turkish society in relation to obsessive-compulsive disorder, anxiety disorder, and depression via content mining of tweets. Tweets were obtained by searching selected keywords via Twitter application programming interface in Python. The tweets were then filtered for psychopathology-related keywords. The sample consisted of 65,031 publicly available tweets that cover the period between 2 December 2019 and 31 May 2021. Latent Dirichlet allocation, was performed to uncover the latent semantic structures in the tweets. Data transformation and analysis were performed by using open-source R (version 4.0.2). As a result of the analysis, there were statistically significant differences in the total number of tweets, mean number of comments, likes, and retweets per tweet between the pre-pandemic and pandemic periods. From the topic modeling, it was also found that semantic strings of the tweets differed in the pandemic period compared to the pre-pandemic period. Topic analysis of social media shares can provide information on the mental health conditions of individuals and the use of tweet content can contribute to the research of psychopathologies, especially during the pandemic.