Geometry of Culture

Github site with code and data associated with the ASR paper on the Geometry of Culture. The full paper can be found here: https://journals.sagepub.com/doi/full/10.1177/0003122419877135

Data

Word Embedding Models

• We provide 2 pre-trained word embedding models that are used in our analyses.
  
  • Google News embedding: https://www.dropbox.com/s/5m9s5326off2lcg/google_news_embedding.zip?dl=0
    
  • Google Ngrams US, 2000-12:

Google Ngrams Raw Text

• For our historical analyses and contemporary validation, we train embedding models on the full Google Ngrams US corpus for particular time periods. The Google Ngrams US corpus is publicly available for download and is hosted here:

Survey of Cultural Associations

• We also provide results from the Mechanical Turk survey of cultural associations. Data files include mean associations on race, class, and gender dimensions for 59 terms. We provide files with and without poststratification weights. These files are hosted here on github in the "Survey of Cultural Associations" directory. Details of the survey can be found in Appendix A of the article.