Atlas of Danish Facebook Culture

The Atlas of Danish Facebook Culture

The Atlas of Danish Facebook Culture is a research project undertaken by the Techno-Anthropology Lab (TANTLab) at the University of Aalborg in Copenhagen. The aim of the project is to understand how Danes have used Facebook as a public space in a six year period between 2012 and 2018. The project is led by Anders Kristian Munk ( and Asger Gehrt Olesen ( 


The purpose of the project

The purpose of the project is to map how public Facebook pages in Denmark post content and how users interact around that content. We want to understand how matters of public concern generate engagement, how different user groups organize around different issues, how sources are shared as part of the debate, and how these patterns break or create media bubbles. To do so we provide a combination of basemaps, that cover the entire public discourse on 72.000 Danish pages, and case studies that engage specifically with issues like vaccinations programs, food activism, waste disposal or the quality of urban spaces. We conduct these case studies in collaboration with a series of other projects. A subsidiary purpose of these collaborations is to find practical ways to make insights from big social data available to broader social research in ways that are ethically responsible and in compliance with GDPR. 

Below is some general stats for the dataset.


The dataset for the Atlas of Danish Facebook Culture was harvested in January 2018 through the open Facebook API for public pages. We do not collect content from personal profiles, closed groups, or other forums where users have a reasonable expectation of privacy. The dataset is thus strictly limited to self-published material (primarily comments or reactions to posts on public pages or events organised by these pages). 

For the purpose of treating the data we have scrambled the user ID’s which makes it impossible to directly associate these ID’s with actual profiles on the platform. However, given that we work with public material, and to the extent that users keep their comments visible on public pages, we also work under the assumption that a user ID can be de-anonymised wtih a simple Google serch, e.g. for the text of a comment. We therefore consider the the dataset as containing personally identifiable information and have accordingly registered the dataset with the Danish Data Protection Agency and taken appropriate steps to ensure that the storage and treatment of data is in compliance with GDPR guidelines. This includes, among other things, that:

  • we never share data containing anonymised user IDs outside the project
  • we aggregate data so that information contained in analysis shared with others can never be traced back to a specific user on the platform
  • we delete GDPR sensitive material once we have treated it for the purposes of the project
  • we allow users who wish to be removed from the dataset to do so (we keep a key that allows us to deanonymise specific user ID's for the purpose)

Below is a visual representation of the protocol used to construct the dataset.