Decoding Divides: The Role of Socioeconomic Status and Personality Traits in AI Divides and Educational Inequality

Synthesis: Wang and colleagues investigate whether integrating AI into primary education deepens the digital divide and educational inequality, and whether socioeconomic status (SES) or personality traits matter more for that risk. Using survey and national registry data from 4,497 Grade 6 students in the Netherlands, the study models the mediating roles of AI usage and digital literacy in the links between student background, personality, and academic performance.

Key Findings

• Digital literacy — not AI usage intensity — is the key mediator. The link between personality traits and academic performance operates through students' digital literacy rather than how intensively they use AI.
• A personality-driven digital skills divide emerges. Differences in digital literacy are driven more by personality traits than by socioeconomic status, complicating the classic SES-focused framing of the digital divide.
• SES advantages operate independently of AI engagement. Socioeconomic advantages on performance persist even apart from how students engage with AI, suggesting AI use is neither the main driver of inequality nor an automatic equalizer.
• The findings point to a more nuanced digital divide in AI-era primary education, where skills and dispositions — not just access or SES — shape who benefits.

What this means for practice

• For primary educators: Digital literacy and its development matter more than raw AI access or usage; teaching AI-era digital skills is an equity-relevant intervention.
• For policymakers: Equity policy should target digital-skills formation and personality-supportive learning environments, not merely device or AI-tool access.
• For researchers: The study models AI-usage and digital-literacy as distinct pathways, urging future work to separate "access/usage" from "skills" when measuring AI-related inequality.

Limitations

• The data are cross-sectional, drawn from the 2022–2023 wave of a regional monitor: the mediation from personality traits through digital literacy to achievement describes associations at a single point in time, so the authors cannot establish temporal precedence or rule out bidirectional relationships and common antecedents.
• The sample covers one Dutch province — 4,497 Grade 6 students in Limburg, from a monitor reaching about 90% of regional schools — so the near-flat link between family socioeconomic status (SES) and digital literacy may reflect the Netherlands' comparatively equitable digital infrastructure and school-choice system; systems with wider access disparities may show stronger SES-driven divides.
• The measures predate the spread of generative AI: the survey was collected in 2022–2023, before tools like ChatGPT entered classrooms, so the AI usage it captures is of an earlier generation of educational tools.
• AI usage was teacher-reported and digital literacy came from the monitor's items rather than a performance-based assessment; teacher reports capture institutionally sanctioned school tools but miss autonomous, out-of-school use of general-purpose AI, and the authors call for performance-based digital literacy measures and independent teacher ratings in future work.

Citation

Wang, Z., van Wetten, S., Segers, E., & Haelermans, C. (2026). Decoding divides: The role of socioeconomic status and personality traits in AI divides and educational inequality. Computers and Education: Artificial Intelligence, 10, 100566.