David Lazer, Oren Tsur, and I recently took a look at the use of the term “coal” in documents produced by members of the US Congress. Some numbers and visualizations based on this were published by Politico. Here, we post the data behind the network map of term co-occurrence in coal-related documents. The political statements these data were extracted from are helpfully provided by VoteSmart.
The files below describe the top coal-related terms for the 111th, 112th and 113th Congress.
The “NODE” files contain a list of the most frequent terms along with the percent of coal-related documents produced by Democrats and Republicans that contained those terms. The “RepDemDiff” column, computed as the difference in percent term use by Democrats vs Republicans, is used to score the right or left skew of each term.
The “EDGE” files contain information about the co-occurrence of the terms, specifically: the percent of Democratic and Republican documents (out of all coal-related Democratic/Republican documents) where the terms are used together.
- 111th Congress – NODES
- 111th Congress – EDGES
- 112th Congress – NODES
- 112th Congress – EDGES
- 113th Congress – NODES
- 113th Congress – EDGES