This data was generated via Remesh during a working session funded by the Peacetech and Polarization Lab at the University of Notre Dame on June 25, 2024 during the “Deliberative Technologies and Peacebuilding in Polarized Contexts Workshop.” During the working session, multiple collective dialogues were conducted with N300 representative samples of the US public recruited via Prolific. They each focused on a polarizing topic among the US public:
Raw and aggregate data is provided for each collective dialogue.
All participant-level demographic, response, and vote data:
- [*]_binary.csv is a table of all agreement votes. All votes are binary [agree or disagree]. The ID of the response voted on is Thought ID. The ID of the voter is Participant ID.
- [*]_preference.csv is a table of all pairwise preference votes. Most are simple pair votes -- choose [thought A or thought B] -- but it also includes votes for when a participant can’t decide: ie. [I’m not sure, I disagree with both, I agree with both].
- [*]_verbatim_map.csv is a table of all thoughts (verbatim particpant responses). It maps Thought IDs to the Thought Text it corresponds to, along with the ID of the participant who submitted the response.
- [*]_participants.csv is a table of all participants. It gives the demographic and poll data for each participant ID along with some other stuff.
Demographic segment-level aggregations of results:
- [*]_aggreagate.csv is a table of all responses and answer options to all prompts and questions, and what fraction of each participant segment selected or agreed* with them.
*note: Aggregate results for ‘opinion ask’ question types include inferences for missing agreement votes.
The data is licensed under CC By 4.0: https://creativecommons.org/licenses/by/4.0/