Measuring Political Activity How Online and Offline Political Activities Connect

Kivonat, rövid leírás

In recent decades, social media has emerged as a pivotal platform for political involvement, influencing how both individuals and political entities engage in political activities. Its widespread use encompasses diverse aspects of political engagement, from organizing protests to election campaigning, making it an indispensable tool in the realm of politics. This dissertation delves into two key topics regarding online political engagement to offer a comprehensive understanding of social media's role in politics. First, the thesis assesses the effect of online political activities on offline political outcomes. The study employs the Bayesian Update method to establish a connection between these two domains based on prior research; further bolstered by an analysis of a national Hungarian panel survey to account for the unique context of the research. The findings affirm the link between online and offline political engagement, particularly within the Hungarian context. After grounding the theory of this connection, the focus of the analyses is on the information dissemination on social media regarding political topics. Prior research suggest that sharing political information can serve as a link between the online and offline political activities, as it can be a form of voter mobilization, and it can be linked to offline popularity. Thus, the second part of the thesis investigates two factors that can have an impact on the act of sharing on social media. Amongst the factors that can affect sharing behavior on social media, this thesis focuses on the role of emotions and network structures. Emotional valence and diversity were linked to different sharing behavior on social media, but prior research conclude different results. The role of emotions on sharing is analyzed in this thesis using posts from Hungarian political actors on Facebook. The emotional content of a post affects how users share them. The research uses Facebook Reactions as indicators of a post's emotional content, and results show that negative emotions increase the number of shares on a post. The analysis uses a time-dependent method, and proves that over time, the effect of reactions can vary. Specifically, after a post reached half its total shares, the additional number of new shares overall decrease, but the presence of Angry reactions increases the additional number of shares in the following timesteps even in the second half of a post’s lifespan. This can be the consequence of the fact that on social media, the content of a post is not the only factor that determines the reactions given to a post; other factors, such as social influence and the algorithm can affect it too. The influence of reactions to each other thus is also analyzed, and the results show that over time, more unanimous reactions are coming to a post, than in earlier stages of its lifespan. Social media platforms, like Facebook, are essentially built upon interpersonal networks. However, the specific network structures of these platforms are often undisclosed. To explore how these attributes influence the dissemination of information, this thesis employs an agent-based model to simulate the dynamics of the social media environment. The simulation highlights how homophily affects information spreading differently in the different types of social networks, while supporting the importance of small world networks in information spreading. This research presents novel contributions to the field by employing diverse analytical methods, thereby contributing new statistical evidence to the existing knowledge base. The examination of the time-dependent nature of reactions on Facebook, as well as the impact of negative emotions on information propagation substantiated using actual Facebook data, and network simulations aims to give a better understanding of online political activities on social media, which might contribute to the application of their analyses in the future.