Replication Data for: Against Conglomeration

Metadata and code in Python for reproducing the results in "Against Conglomeration." Full texts are held by HathiTrust. Hathi IDs are provided with the metadata. Files include: the corpus metadata; a spreadsheet with a document-term matrix for the top 10k frequent words in the corpus; a Jupyter Notebook that ingests the metadata and DTM and replicates our classifier; a spreadsheet with our reported predictions for each novel; and a spreadsheet with our reported model (i.e. a list of tokens and regression coefficients)