Project Scope

The interface is intended for public exploration and research review. It is not a live social-media monitoring tool, and it should not be treated as a complete record of every communication by every state legislator. It is a structured view into a restored analytical corpus.

The site emphasizes reproducible filtering and comparison. Users can move from broad summaries to individual legislators, states, topics, and moments, while preserving the active filters in the URL when possible.

Methods

CivicWatch is based on a restored analytical corpus, not a live feed. The corpus covers January 1, 2020 through December 31, 2024 and contains approximately 22.2 million posts, 5,927 legislators, and 22 topic categories.

Aggregations are served from PostgreSQL base tables, restored materialized views, and local exploration views prepared after database restore. These prepared views support faster summaries for the public interface while retaining links back to the underlying entities where available.

Topic 999 is labeled Uncategorized and remains visible by default. Missing legislator metadata is shown as missing rather than silently excluded, so users can see where coverage or public-record fields are incomplete.

Data Notes

Legislator names, party labels, chambers, state codes, topic classifications, engagement counts, and post text may be incomplete or uneven across the corpus. Where fields are absent, the interface displays a no-data marker or an explicit unknown category.

Topic, state, and legislator summaries are analytical conveniences. They should be interpreted as summaries of the dataset as restored, not as official legislative records or claims about legislative behavior outside the observed posts.

Contact

For questions about CivicWatch, the dataset, or the research context, contact Quinn K Wolter.

QuinnKWolter@pitt.edu

Acknowledgements

  • University of Pittsburgh Computational Social Dynamics (PICSO) Lab Research lab
  • Quinn K Wolter Lead Researcher & Developer
  • Professor Yu-Ru Lin Primary Advisor
  • Radhika Purohit Developer
  • Chase Lahner Developer
  • Andrew Aquilinas Developer
  • Yongsu Ahn Research Advisor
  • Ahana Biswas Research Advisor
  • Yuehong Cassandra Tai Original Dataset Author
  • Nitheesha Nakka Original Dataset Author
  • Khushi Navin Patni Original Dataset Author
  • Sarah Rajtmajer Original Dataset Author
  • Kevin Munger Original Dataset Author
  • Bruce A. Desmarais Original Dataset Author

Funding

National Science Foundation logo

We are grateful for the support of the National Science Foundation (NSF #2318461) and the Bright Initiative. Any opinions, findings, and conclusions or recommendations expressed in this material do not necessarily reflect the views of the funding sources.