Implicit Gender Biases in Professional Software Development: An Empirical Study

Abstract

It has been well-known that the software development profession lacks gender diversity, particularly in the technical leadership positions. Researchers and practitioners have spent tremendous efforts on identifying the problems and finding solutions. However, most of the existing software engineering literature focuses on the explicit gender biases but ignores implicit gender biases. To fill this gap, the study sought to empirically investigate whether professional software engineers hold implicit gender biases related to women in the software development profession, and examine whether these implicit biases predict discriminatory decision-making. Using data from 142 professional software engineers in seven organizations, our study yields a rich set of concerning findings. First, we find that implicit biases were pervasive-both male and female software engineers implicitly associated software development professions, particular technical leadership roles, with men, not women, and also associated women with the home and family. Besides, people often cannot resist their implicit gender biases and make decisions in gender-neutral ways while they do well in resisting their explicit gender biases.