JSTOR Data for Research allows researchers to text mine content in JSTOR’s extensive archive of scholarly literature;  journal articles, primary sources, and books; textual content (metadata or OCR-ed full text) can be assembled into a downloadable dataset for analysis with one’s own tools. Arthur Netto, a Ph.D. student at the University of São Paulo and fellow of the Center for the History of Political Economy at Duke University, started using R bibliometric tools to make NGRAM plots of JSTOR data, but found he needed – and then created – more functions for analyzing the content. He wrapped everything into a single R package – tidyJSTOR – and uploaded the package to Github in order to disseminate it and receive comments on code and functions.

At this Digital Brown Bag talk, Arthur will talk about the research questions motivating his use of JSTOR’s Data for Research, the potential and limitations of existing R bibliometric tools, and the additional features he’s built into tidyJSTOR. As a self-taught R user and self-taught user of bibliometrics, he’s also eager to hear others’ approaches and suggestions. If you’d like to check out tidyJSTOR before the talk, please do! You can access the tool and the tutorial here: https://github.com/arthurbnetto/tidyJSTOR.