Education and Employment Through ChatGPT: Student Perceptions in Latin America and Asia

This article examines higher education students’ perceptions of ChatGPT in relation to skills development, anticipated labor-market transformation, and job readiness across ten countries in Latin America and Asia. The analytical sample included 3085 students for the skills-development dimension (PQ29) and 3083 for the labor-market transformation (PQ30) and skills-mismatch reduction (PQ31) dimensions. Analyses included descriptive statistics by country and discipline, adjusted general linear models, correlations, network analyses, and regional comparisons using Welch’s t tests, Cohen’s d, and covariate-adjusted models. Country-level differences remained statistically detectable after adjustment for discipline, gender, and level of study, whereas adjusted discipline-level mean differences were not statistically significant. Latin America–Asia differences were negligible for PQ29 and PQ30 and small for PQ31; after adjustment, the PQ30 regional difference was no longer statistically significant. The findings therefore describe students’ perceptions and associations, not demonstrated gains in skills, employability, or employment outcomes.