Detecting Gender Bias to Enhance Inclusivity in Software Engineering Education

Abstract

Educational environments and course materials—such as textbooks, notes, slides, and examinations—are foundational elements that can either encourage or discourage students from pursuing studies and careers in STEM fields. Detecting gender bias in these materials is essential for fostering inclusivity and diversity in the field. Research shows that early exposure to inclusive and relatable course content significantly influences students’ interest and persistence in STEM fields, highlighting a direct connection between educational experiences and career choices. Building on this foundation, this study investigates whether course materials—with the focus on software engineering—exhibit a male, female, or neutral orientation through an automated approach that incorporates keyword extraction, word analysis, and classification. To ensure our findings accurately reflect the content’s gender orientation, we also consider the subject matter of the materials. This approach helps to distinguish between gendered terms tied to specific contexts and broader gender bias. By offering this analysis of gender bias, our approach supports efforts to create more inclusive and equitable learning environments.