Percepciones sobre el uso académico de ChatGPT y su impacto en el rendimiento escolar: un análisis logístico en Nayarit, México

Abstract

The integration of generative artificial intelligence tools, such as ChatGPT, in educational environments raises questions about their effects on learning. This study analyzes the relationship between the academic use of ChatGPT and the perceived improvement in school performance among high school and university students in Nayarit, Mexico. A quantitative, experimental, cross-sectional, and correlational approach was used with a sample of 224 students who reported using ChatGPT for school purposes. The instrument, validated by experts, was administered online and included items on attitudes, uses, and satisfaction with the tool. An exploratory factor analysis identified three main components: perceptions and attitudes toward artificial intelligence, academic use and functionality, and overall satisfaction. A binary logistic regression model was then applied, showing that these factors significantly influence the likelihood of perceiving an improvement in academic performance. The findings suggest that the conscious and formative use of ChatGPT may support learning and study organization processes, highlighting its educational potential when critically and pedagogically integrated.