Facebook × Political communication

4 datasets that are both from Facebook and tagged Political communication.

4 datasets.

rtGender — Facebook Congress Analyzed

A Facebook Congress dataset of 548,341 posts, 13,866,507 responses, collected by Peixian and Zhang. Each record contains posts, responses, op_gender, responder_gender, sentiment, relevance. rtGender dataset for Facebook Congress — studies gendered responses in online discourse. Suitable for gender studies, sentiment analysis, response prediction.

14.4M records9.5M nodes10.3M edges2018

Facebook Pages — Government Category (Nov 2017) Analyzed

A Facebook dataset of 7,057 pages, 89,455 mutual likes, collected by Rozemberczki et al. via the Facebook Graph API (November 2017). Each record contains undirected edges between blue verified Facebook government pages that mutually liked each other, with node attributes including page name and raw Facebook ID. Graph of blue verified Facebook government pages, where edges represent mutual likes between pages, capturing how citizens engage with government content and official institutions. Suitable for community detection in social networks, graph embedding, government-citizen engagement studies, political communication analysis.

7.1K records7.1K nodes89.5K edges2019

Facebook Pages — Politician Category (Nov 2017) Analyzed

A Facebook dataset of 5,908 pages, 41,729 mutual likes, collected by Rozemberczki et al. via the Facebook Graph API (November 2017). Each record contains undirected edges between blue verified Facebook politician pages that mutually liked each other, with node attributes including page name and raw Facebook ID. Graph of blue verified Facebook politician pages, where edges represent mutual likes between pages, capturing how citizens engage with political figures and campaign content. Suitable for community detection in social networks, graph embedding, political communication analysis, politician influence studies, political polarization research.

5.9K records5.9K nodes41.7K edges2019

Facebook Social Connectedness Index Analyzed

Meta's Social Connectedness Index (SCI): the relative probability that two Facebook users in two places are friends, scaled so that the most connected pair is 1,000,000,000. This package is the release distributed on HDX, at several resolutions: country to country, GADM level 1 and 2 regions, geoBoundaries ADM1 and ADM2, EU NUTS 1 to 3 (2024), US counties and US ZIP code tabulation areas, with region-to-country lookups and the methodology paper. Suitable for research on social ties, migration, trade and economic spillovers between regions.

3.2K records3.2K nodes5.1M edges2026