This book relies on resources that a lot of people have made available, usually for free.
The raw data comes from JSTOR’s Data for Research program. It wouldn’t really be possible without the work they did to make that set available.
I initially transformed the JSTOR Ddta into something that could be read in R via some scripts from John Bernau. His paper Text Analysis with JSTOR Archives describes some techniques for modeling trends in sociology journals using the JSTOR archives.
Once I had the data in R, I analyed it using the topicmodels package by Bettina Grün and Kurt Hornik.
The idea for analyzing this data using topicmodels to analyze the data in this way comes from What Is This Thing Called Philosophy of Science? A Computational Topic-Modeling Perspective, 1934–2015, by Christophe Malaterre, Jean-François Chartier, and Davide Pulizzotto (2019).
As well as those sources, I learned a lot about how to use the topicmodels package from Text Mining with R: A Tidy Approach by Julia Silge and David Robinson, and from some articles in Towards Data Science, including those by Shashank Kapadia (2019) and by Farren tang. (2019).
The citation data I use here are from Google Scholar, and I accessed them via “Publish or Perish” (Harzing 2007). Happily, a Mac version of Publish or Perish came out recently; in the past I had set up PC emulators just so I could run it.
The whole book was put together using the bookdown package, primarily built by Yihui Xie (2020). This in turn is built on the Pandoc language, which was originally built by John MacFarlane. And the code uses tools from the tidyverse package throughout, which was originally built by Hadley Wickham.
Michigan Publishing was incredibly helpful both with copyediting, which fixed both many mistakes and some dubious stylistic choices that I had originally made, and with the very difficult task of getting a book this unusual into something like print.
And my daughter Nyaya has also helped catch a lot of errors, though given the way I write, this is an endless task.