Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Examining (Political) Content Consumption on Facebook Through Data Donation

Joao Couto, Kiran Garimella

TL;DR

The paper addresses how political information is consumed on Facebook under restricted data access. It introduces a data donation approach paired with CrowdTangle enrichment to capture public interactions, yielding a dataset from 1,261 US users and over 32 million posts. It combines a fine-tuned POLUSA-based political-content classifier, BERTopic topic modeling, and propensity-weighted reweighting to produce population-level estimates of political and non-political content consumption. The findings reveal that political content comprises about 17 percent of consumption and varies by age, ethnicity, and gender, with younger individuals and Hispanics showing lower engagement and older men higher engagement. The work advances a privacy-preserving, platform-agnostic framework for studying social-media usage in the post-API era and provides data resources to researchers and policymakers.

Abstract

This paper investigates the usage patterns of Facebook among different demographics in the United States, focusing on the consumption of political information and its variability across age, gender, and ethnicity. Employing a novel data donation model, we developed a tool that allows users to voluntarily share their interactions with public Facebook groups and pages, which we subsequently enrich using CrowdTangle. This approach enabled the collection and analysis of a dataset comprising over 1,200 American users. Our findings indicate that political content consumption on Facebook is relatively low, averaging around 17%, and exhibits significant demographic variations. Additionally, we provide insights into the temporal trends of these interactions. The main contributions of this research include a methodological framework for studying social media usage in a privacy-preserving manner, a comprehensive dataset reflective of current engagement patterns, and descriptive insights that highlight demographic disparities and trends over time. This study enhances our understanding of social media's role in information dissemination and its implications for political engagement, offering a valuable resource for researchers and policymakers in a landscape where direct data access is diminishing.

Examining (Political) Content Consumption on Facebook Through Data Donation

TL;DR

The paper addresses how political information is consumed on Facebook under restricted data access. It introduces a data donation approach paired with CrowdTangle enrichment to capture public interactions, yielding a dataset from 1,261 US users and over 32 million posts. It combines a fine-tuned POLUSA-based political-content classifier, BERTopic topic modeling, and propensity-weighted reweighting to produce population-level estimates of political and non-political content consumption. The findings reveal that political content comprises about 17 percent of consumption and varies by age, ethnicity, and gender, with younger individuals and Hispanics showing lower engagement and older men higher engagement. The work advances a privacy-preserving, platform-agnostic framework for studying social-media usage in the post-API era and provides data resources to researchers and policymakers.

Abstract

This paper investigates the usage patterns of Facebook among different demographics in the United States, focusing on the consumption of political information and its variability across age, gender, and ethnicity. Employing a novel data donation model, we developed a tool that allows users to voluntarily share their interactions with public Facebook groups and pages, which we subsequently enrich using CrowdTangle. This approach enabled the collection and analysis of a dataset comprising over 1,200 American users. Our findings indicate that political content consumption on Facebook is relatively low, averaging around 17%, and exhibits significant demographic variations. Additionally, we provide insights into the temporal trends of these interactions. The main contributions of this research include a methodological framework for studying social media usage in a privacy-preserving manner, a comprehensive dataset reflective of current engagement patterns, and descriptive insights that highlight demographic disparities and trends over time. This study enhances our understanding of social media's role in information dissemination and its implications for political engagement, offering a valuable resource for researchers and policymakers in a landscape where direct data access is diminishing.
Paper Structure (16 sections, 22 figures, 1 table)

This paper contains 16 sections, 22 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Data Donation
  4. Data Processing
  5. Labeling of Political and Non-Political Content
  6. Detection of Political Content
  7. Topics
  8. Re-weighting to obtain population level estimates
  9. Political content
  10. Political content prevalence
  11. Political content consumption over time
  12. Political leanings of various demographics
  13. Overall content consumption
  14. Topics disproportionately consumed by specific demographics
  15. Heterogeneity in content consumption across groups
  16. ...and 1 more sections

Figures (22)

  • Figure 1: Facebook data donation flowchart. Panel 1 -- Welcome page where the user sees a 'Login with Facebook' button. Panels 2--4 allow the users to select what public Pages and Groups they wish to donate. Identifying details have been anonymized.
  • Figure 2: Fraction of users by ethnicity and gender.
  • Figure 3: Fraction of users by ethnicity and age.
  • Figure 4: Political content consumption by: (a) ethnicity, (b) age group, (c) gender, (d) gender and ethnicity.
  • Figure 5: Time series. The data spans from October 2022 to October 2023.
  • ...and 17 more figures