## What are these things? ## This is the data! Many things jumbled together here: 1. The actual raw texts used. They're all zipped together in a file called `Corpora.zip`. It's pretty big. There's a lot of files in there. 2. The dictionaries used to score each corpus. `moral_foundations_dictionary.dic` scores each for the various foundations in aggregate, `authority_ingroup.dic` identifies frequencies for each Authority and Ingroup words, and `MFD2.0.dic` is the revised dictionary used in a footnote to the paper. If you want to see the words, you can open each file in any text-editor. If you want to rescore things yourself, each can be loaded directly into LIWC. Maybe other programs work as well? Let me know if you find another one? 3. The output of the dictionary scores for each corpus. Those CSVs that end in `-MF.csv` report the moral foundations scores for each, those CSVs that end in `-.AI.csv` report the frequencies for each Authority and Ingroup word, and the one that ends in `-MF2.0.csv` reports the moral foundation 2.0 scores for the State of the Union. 4. The sheet that has the population stats for the US (and, with way more missing data, Canada and New Zealand, if you can come up with a better imputation approach than we could), called `RuralPct.csv`. Within those files, - Canada = Canadian Speeches from the Throne - Inaugural = US Inaugural Addresses - NZ = New Zealand Speeches from the Throne - PP = US Party Platforms - SC = Supreme Court Opinions - Senate = Senate Speeches - Songs = Lyrics of Top-5 songs - SotU = US States of the Union - Stanford = Stanford Commencement Addresses - UVa = University of Virginia Commencement Addresses