Stereotypes are More Powerful When People Like to Agree With Each Other

Stereotypes Are More Powerful When People Like to Agree with Each Other

Michael Chwe
September 2016

This website provides supporting materials for my working paper "Stereotypes Are More Powerful When People Like to Agree with Each Other."

The data set of 119 academic fields used for the regressions in the paper is mainforanaylsis.csv (CSV format). The details of how it was derived are in main.xlsx (Excel spreadsheet). The data were derived from the following primary source documents.

Table 16 and Table 22 of Doctorate Recipients from U.S. Universities, 2011 by the National Science Foundation. Information about the survey of Science and Engineering Doctorates is available here and the complete collection of tables (downloadable zip file) is available here.
Table D1 of the working paper version of "The Skewness of Science in 219 Sub-fields and a Number of Aggregates,'' by Pedro Albarrán, Juan A. Crespo, Ignacio Ortuño, and Javier Ruiz-Castillo, Scientometrics (2011) 88: 385–397.
Table 4 (Extended), "General Test Percentage Distribution of Scores Within Intended Graduate Major Field Based on Seniors and Nonenrolled College Graduates," by the Educational Testing Service. The original location is here.

Other data used by the paper come from these sources.

Table S4 from the supplementary materials for "Expectations of brilliance underlie gender distributions across academic disciplines," by Sarah-Jane Leslie, Andrei Cimpian, Meredith Meyer, and Edward Freeland, Science (2015) 347: 262–265.
The HathiTrust Digital Library. Information on how to return search results using automated inquiries is in this July 27, 2016 email by Peter Organisciak.

The following widget allows you to perform the regressions in Table 4 yourself and try your own specifications. You have to download a (rather large) plug-in to make the widget work.