Relative Advantage and Compatibility as Drivers of ChatGPT Adoption in Latin American Higher Education: A PLS SEM Study Towards Sustainable Digital Education

Abstract

As Latin American universities pursue digitally and environmentally sustainable teaching models, understanding why students adopt generative AI is essential. We analyzed data from undergraduate students (n = 792) across five Latin American countries (Peru, Chile, Bolivia, Argentina, and Colombia). Grounded in the diffusion of innovations theory, the study evaluated the effects of relative advantage, compatibility, complexity, trialability, and observability on attitudes towards ChatGPT and examined the effect of attitude on intention to use among higher education students in the region. The reliability and validity of the measurement scale were confirmed, and structural relationships were tested using partial least squares structural equation modeling (PLS-SEM). The model explained 58.1% of the variance in attitude: relative advantage (β = 0.247) and compatibility (β = 0.246) exerted the largest effects, followed by trialability (β = 0.223) and observability (β = 0.167); complexity showed a weaker yet significant effect (β = 0.118). Attitude strongly predicted the intention to use ChatGPT (β = 0.777), accounting for 60.4% of its variance. All paths were significant (p < 0.001), and psychometric indicators exceeded recommended thresholds. These findings indicate that student adoption is driven more by perceived academic benefits and alignment with existing learning routines than by technical ease. Highlighting concrete, ethically delineated use cases and providing guided institutional spaces for experimentation may accelerate the responsible, long-term adoption of generative AI in quality higher education.

Ähnliche Arbeiten