Abstract:

Gender equality remains one of the most persistent challenges in STEM higher education, particularly in quantitative subjects where women’s participation, confidence, and academic self-efficacy continue to lag. At the same time, the rapid adoption of artificial intelligence (AI) tools in university teaching is transforming how students learn, interact with content, and engage with problem-solving tasks. Although international organizations emphasize the need to integrate gender perspectives into digital and AI-enhanced education, current research provides limited comparative evidence on how AI-mediated learning shapes gendered experiences in specific STEM subjects across different cultural and institutional contexts. In particular, there is a lack of empirical studies examining this intersection in undergraduate Applied Statistics courses in Latin America and Southern Europe.

This study addresses this gap through a comparative analysis of Spain, Mexico, and Chile—three countries that share Ibero-American academic traditions yet differ in institutional governance, digital maturity, and gender-equality policies. The research explores how female and male students in Applied Statistics courses perceive and use AI-based learning tools, and how these technologies relate to students’ self-confidence, engagement, and perceived fairness in assessment. It also examines whether institutional climate and pedagogical strategies promote or hinder gender-inclusive AI integration in STEM curricula.

In this research, a mixed-methods design was employed. Quantitative data were collected through structured surveys administered to undergraduate STEM students (n≈450) across the three countries, measuring attitudes toward AI-mediated learning, statistical self-efficacy, perceived institutional support, and concerns related to transparency, bias, and algorithmic fairness. Factor analysis and cross-country comparisons were conducted to identify gendered patterns. Qualitative data from semi-structured interviews with students and instructors (n≈24) provided deeper insights into classroom experiences, institutional expectations, and cultural meanings attributed to the use of AI in Statistics. Furthermore, triangulation ensured analytical robustness.

Preliminary findings suggest that while AI-based tools can support more adaptive and personalized learning, their benefits are not evenly distributed. Female students across all three contexts report higher concerns about fairness and bias, which correlate with lower usage and reduced confidence in advanced statistical tasks. Cross-national differences indicate that institutional digital strategies and gender-equality policies shape how AI is perceived and adopted. Spain shows closer alignment between AI integration and gender-inclusion initiatives, whereas Mexico and Chile present more heterogeneous patterns influenced by digital gaps, instructor training, and resource availability.

The study contributes to the international debate on equitable digital transformation in higher education by offering empirical evidence on how gender, AI, and STEM learning interact in culturally connected but structurally diverse systems. It provides actionable recommendations for designing gender-sensitive AI pedagogies, strengthening institutional support mechanisms, and aligning STEM curricula with emerging ethical and regulatory frameworks for responsible AI.

Keywords:

Gender education, equality education, STEM, artificial intelligence, higher education.