Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Decoding Political Polarization in Social Media Interactions

Giulio Pecile, Niccolò Di Marco, Matteo Cinelli, Walter Quattrociocchi

TL;DR

The paper addresses how political polarization manifests in social media by quantifying selective exposure in large-scale Facebook interactions with news pages across five bias categories. It uses Shannon entropy to measure concentration of engagement and two null models—strong and weak randomizations—to test whether political bias drives page-level selectivity beyond simple popularity. Key findings show pervasive leaning-driven selectivity, particularly among center-left users, with pages and biases interacting to shape engagement patterns; center-left and centrist likes are especially informative of political affiliation. The work provides a robust framework for analyzing ideological fragmentation online and offers methods for distinguishing bias-driven behavior from random or popularity-based effects, with implications for understanding echo chambers and designing interventions.

Abstract

Social media platforms significantly influence ideological divisions by enabling users to select information that aligns with their beliefs and avoid opposing viewpoints. Analyzing approximately 47 million Facebook posts, this study investigates the interactions of around 170 million users with news pages, revealing distinct patterns based on political orientations. While users generally prefer content that reflects their political biases, the extent of engagement varies even among individuals with similar ideological leanings. Specifically, political biases heavily influence commenting behaviors, particularly among users leaning towards the center-left and the right. Conversely, the 'likes' from center-left and centrist users are more indicative of their political affiliations. This research illuminates the complex relationship between social media behavior and political polarization, offering new insights into the manifestation of ideological divisions online.

Decoding Political Polarization in Social Media Interactions

TL;DR

The paper addresses how political polarization manifests in social media by quantifying selective exposure in large-scale Facebook interactions with news pages across five bias categories. It uses Shannon entropy to measure concentration of engagement and two null models—strong and weak randomizations—to test whether political bias drives page-level selectivity beyond simple popularity. Key findings show pervasive leaning-driven selectivity, particularly among center-left users, with pages and biases interacting to shape engagement patterns; center-left and centrist likes are especially informative of political affiliation. The work provides a robust framework for analyzing ideological fragmentation online and offers methods for distinguishing bias-driven behavior from random or popularity-based effects, with implications for understanding echo chambers and designing interventions.

Abstract

Social media platforms significantly influence ideological divisions by enabling users to select information that aligns with their beliefs and avoid opposing viewpoints. Analyzing approximately 47 million Facebook posts, this study investigates the interactions of around 170 million users with news pages, revealing distinct patterns based on political orientations. While users generally prefer content that reflects their political biases, the extent of engagement varies even among individuals with similar ideological leanings. Specifically, political biases heavily influence commenting behaviors, particularly among users leaning towards the center-left and the right. Conversely, the 'likes' from center-left and centrist users are more indicative of their political affiliations. This research illuminates the complex relationship between social media behavior and political polarization, offering new insights into the manifestation of ideological divisions online.
Paper Structure (12 sections, 6 equations, 5 figures, 2 tables)

This paper contains 12 sections, 6 equations, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related works
  3. Polarization and Echo Chambers
  4. The interplay between social media and polarization
  5. Materials and Methods
  6. Entropy and selective exposure
  7. Strong and weak randomization
  8. Results
  9. Measuring Selective exposure
  10. Page and Bias Selectivity
  11. Bias and Page Selectivity
  12. Conclusion

Figures (5)

  • Figure 1: In the tripartite representation of the interactions, the strong randomization process affects the user-pages interactions, while the weak one affects the page-bias affiliations.
  • Figure 2: Number of Facebook pages and users grouped by political affiliation.
  • Figure 3: Number of pages among which users divide their attention.
  • Figure 4: Distributions of bias entropy of users compared with the strongly randomized scenario.
  • Figure 5: Distribution of users' bias entropy compared with the benchmark values obtained via weak randomization.