Exploring Gender Bias in Requirements Engineering: Challenges and Mitigation From Systematic Literature Review

Aim/Purpose: This study investigates gender bias, an important social dimension in Requirements Engineering (RE), with a special emphasis on identifying challenges and mitigation strategies through a Systematic Literature Review (SLR). Background: Though an important activity, RE is heavily influenced by stakeholders’ social dynamics, particularly gender bias in software requirements. The less represented genders in RE may negatively influence the process, including a lack of inclusiveness in software products and limited diversity of perspectives. However, researchers and practitioners are exploring a comprehensive understanding of gender bias across various RE activities, and this research area is still emerging. Methodology: We conducted an SLR and explored the challenges and the mitigation strategies, techniques, and solutions reported in the literature to address gender bias in RE. This review investigates the ongoing challenges of gender bias in RE and the potential to mitigate them. Due to bias, the software design may lead to poorer outcomes, and the needs of diverse groups may be overlooked. The Kitchenham SLR guidelines are followed in this review to investigate relevant research studies, as they help identify the challenges, mitigation techniques, and the effectiveness of existing validation methods. During initial scanning, almost 3,889 studies on gender bias were found from computing sources. These studies are further screened for inclusion and exclusion criteria to select the most appropriate studies for formulating research questions. After applying the three major steps of Kitchenham SLR guidelines, gender bias challenges and mitigation techniques are identified. Contribution: The contribution of our work is multifaceted: (1) conducting a systematic review of existing literature, (2) identifying key challenges of gender bias in RE, and (3) highlighting the effectiveness of gender bias mitigation techniques to overcome the challenges. Findings: The findings indicate the persistent challenges posed by gender bias in RE and highlight existing mitigation strategies. Furthermore, gender bias in RE can result in poorly designed software and neglected diverse user needs. Recommendations for Practitioners: It is recommended that practitioners integrate inclusive practices across RE activities, adopt models and frameworks that cater to diverse users, and actively address gender bias. Recommendation for Researchers: Researchers should further investigate the validation of available mitigation techniques that address gender bias in RE. Also, new methods are needed to address gender bias in RE-related activities to ensure equity and fairness in software products. Impact on Society: The outcome of this study is helpful in addressing gender bias in RE, integrating inclusivity into software systems, and developing equitable systems well-aligned with the needs of diverse user groups. Future Research: In the future, more empirical studies will be conducted to validate gender-inclusive methods and tools across different RE scenarios. Promote continuous gender-sensitization and awareness programs for all stakeholders involved in RE to ensure long-term impact.