The Role of Social Identity in Shaping Biases Against Minorities in Software Organizations

TL;DR

This study investigates how social identity influences biases against minorities in software organizations, revealing prevalent biases and demographic disparities through a vignette-based survey.

Contribution

It applies Social Identity Theory to quantify and analyze specific biases in software workplaces, highlighting demographic factors affecting bias prevalence.

Findings

Career development and task selection biases are most common.

Women face over three times higher bias risk than men.

Marginalized ethnic groups are more targeted by identity attacks.

Abstract

While systemic workplace bias is well-documented in non-computing fields, its specific impact on software engineers remains poorly understood. This study addresses that gap by applying Social Identity Theory (SIT) to investigate four distinct forms of bias: lack of career development, stereotyped task selection, unwelcoming environments, and identity attacks. Using a vignette-based survey, we quantified the prevalence of these biases, identified the demographics most affected, assessed their consequences, and explored the motivations behind biased actions. Our results show that career development and task selection biases are the most prevalent forms, with over two-thirds of victims experiencing them multiple times. Women were more than three times as likely as men to face career development bias, task selection bias, and an unwelcoming environment. In parallel, individuals from marginalized ethnic backgrounds were disproportionately targeted by identity attacks. Our analysis also confirms that, beyond gender and race, factors such as age, years of experience, organization size, and geographic location are significant predictors of bias victimization.