Fundamental errors of data collection & validation undermine claims of ‘Ideological Intensification’

This is the submitted (author version) of an article accepted for publication in BioScience. The National Association of Scholars (i.e., NAS) recently published a report by Mason Goad and Bruce R. Chartwell which the authors claim is “the largest quantitative study of the growth of DEI-related language in the sciences” published to date. Having reviewed their code for harvesting online data and the resulting datasets, I conclude that they failed to conduct even the most rudimentary data validation procedures, and that as a result the “clean” data sets with which they conducted their analyses contain thousands of irrelevant records and duplications. Policy Institutes play a vital role in shaping higher education and research policy, but only when they put scholarship ahead of ideology. Goad and Chartwell’s conclusions regarding the prevalence of DEI in STEM were based entirely on visualizations of fundamentally flawed datasets. The NAS should retract their report immediately; failure to do undermines both the integrity and intellectual rigor that they and the report’s authors claim to espouse as fundamental principles.