Utilizing Time Series Analysis to Understand the Effects of Social Media Activities on Public Opinion over Time

This study explores the dynamic relationship between social media activities and public opinion through time series analysis, utilizing extensive datasets from 2018 to 2024 that encompass over 1.5 million social media posts and opinion poll data. Using Vector Auto regression (VAR) models and structural time series analysis, the research reveals significant patterns indicating that specific types of social media engagement can precede shifts in public sentiment by 2 to 8 days, particularly in areas that provoke political polarization. The results highly recommend focusing not on the frequency of posts but on engagement metrics as a basis for the proper assessment of public opinion changes. This shows the crucial role of social media in developing public opinion that immediately affects political campaigning, advertising, and social science, among other fields. The methodology of the research served as a basis for creating a reliable tracking and, possibly, predicting mechanism that can trace the changes in social consensus due to the popularity of some topics in digital space.