Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Leadership and Engagement Dynamics in Legislative Twitter Networks: Statistical Analysis and Modeling

Carolina Luque, Juan Sosa

Abstract

In this manuscript, we analyze the interaction network on Twitter among members of the 117th U.S. Congress to assess the visibility of political leaders and explore how systemic properties and node attributes influence the formation of legislative connections. We employ descriptive social network statistical methods, the exponential random graph model (ERGM), and the stochastic block model (SBM) to evaluate the relative impact of network systemic properties, as well as institutional and personal traits, on the generation of online relationships among legislators. Our findings reveal that legislative networks on social media platforms like Twitter tend to reinforce the leadership of dominant political actors rather than diminishing their influence. However, we identify that these leadership roles can manifest in various forms. Additionally, we highlight that online connections within legislative networks are influenced by both the systemic properties of the network and institutional characteristics.

Leadership and Engagement Dynamics in Legislative Twitter Networks: Statistical Analysis and Modeling

Abstract

In this manuscript, we analyze the interaction network on Twitter among members of the 117th U.S. Congress to assess the visibility of political leaders and explore how systemic properties and node attributes influence the formation of legislative connections. We employ descriptive social network statistical methods, the exponential random graph model (ERGM), and the stochastic block model (SBM) to evaluate the relative impact of network systemic properties, as well as institutional and personal traits, on the generation of online relationships among legislators. Our findings reveal that legislative networks on social media platforms like Twitter tend to reinforce the leadership of dominant political actors rather than diminishing their influence. However, we identify that these leadership roles can manifest in various forms. Additionally, we highlight that online connections within legislative networks are influenced by both the systemic properties of the network and institutional characteristics.
Paper Structure (14 sections, 7 figures, 9 tables)

This paper contains 14 sections, 7 figures, 9 tables.

Table of Contents

  1. Introduction
  2. 117th U.S. Congress
  3. Methodology
  4. Data collection and network structure
  5. Methods
  6. Network topology
  7. Exponential Random Graph Model
  8. Stochastic Block Model
  9. Results
  10. Characterization of Network Actors
  11. Connectivity and Information Flow
  12. Exponential Random Graph Model
  13. Stochastic Block Model
  14. Discussion

Figures (7)

  • Figure 1: 117th Congress Twitter network of the U. S. In each graph, the actors with the highest and lowest scores in-degree and -strength metrics are highlighted. The size of the prominent nodes is proportional to the value of the metric, while the other nodes have a fixed size. Nodes representing members of the Republican Party are colored red, Democrat Party members are colored blue, and independents are colored yellow.
  • Figure 2: 117th Congress Twitter network of the U. S. In each graph, the actors with the highest scores in centrality measures are highlighted. The size of these nodes is proportional to their metric values, while the other nodes remain fixed in size. In graphs (c)–(e), only some actors are visualized due to label overlap. The red color corresponds to members of the Republican Party, blue to the Democrat Party, and yellow to independents.
  • Figure 3: Maximal cliques in the network of the 117th Congress of the U. S. All the six subgroups are composed of members of the Republican Party.
  • Figure 4: Markov chain and log-likelihood distribution for the number of edges in Model 6. The convergence diagnostics for the other predictors in the model exhibit similar behavior.
  • Figure 5: Integrated conditional likelihood. The red line indicates the optimal number of communities.
  • ...and 2 more figures