Women in the Workplace: Analyzing Gender Biases in Corporate Email Communications

Anonymous

Abstract: Gender disparities in the workplace hinder women in career advancement and equity. Communication within companies reflect gender norms and discrimination that affect organizational structures. Gender biases are exhibited in different forms, including unequal treatment, associations of gender with certain concepts, and stereotyping language. We approach different angles of considering linguistic gender biases to provide an extensive analysis on the role of gender in workplace emails (1) Determine how receivers’ genders affect language use in the emails through computational text analysis with LIWC and model explainability investigations. (2) Examine gender disparities through representation biases in word embeddings to find asymmetric associations of gender with profession words. (3) Identify biased emails in the workplace and create an NLP tool to identify and flag phrases in emails that express gender biases. We study corporate interactions through the Enron Corpus, a uniquely available database of 500K real workplace emails of Enron employees. Our results find significant presence of biases in all three paths, reveal gender inequalities through a case study of a corporation, and show the effectiveness of natural language processing methods to avoid such occurrences in further workplaces.

Paper Type: short

Research Area: Computational Social Science and Cultural Analytics