Greater female first author citation advantages do not associate with reduced or reducing gender disparities in academia

Abstract

Ongoing problems attracting women into many Science, Technology, Engineering and Mathematics (STEM) subjects have many potential explanations. This article investigates whether possible under-citation of women associates with lower proportions of, or increases in, women in a subject. It uses six million articles published 1996-2012 across up to 331 fields in six mainly English-speaking countries: Australia, Canada, Ireland, New Zealand, UK and USA. The proportion of female first and last authored articles in each year was calculated and 4968 regressions were run to detect first author gender advantages in field normalised article citations. The proportion of female first authors in each field correlated highly between countries and the female first author citation advantages derived from the regressions correlated moderately to strongly between countries, so both are relatively field-specific. There was a weak tendency in the USA and New Zealand for female citation advantages to be stronger in fields with fewer women, after excluding small fields, but no other association evidence. There was no evidence of female citation advantages or disadvantages to be a cause or effect of changes in the proportions of women in a field for any country. Inappropriate uses of career-level citations are a likelier source of gender inequities.