Chat-GPT in Higher Education, a Comparative Analysis within Business Students at Universities in two Peruvian Cities

Abstract

The article explores the use of ChatGPT in higher education through a comparative analysis among business students from universities in Lima and Arequipa. With a sample of 768 students, it seeks to achieve two main objectives: to compare students' perceptions of the usefulness and limitations of the educational use of ChatGPT and to validate the internal consistency and factorial structure of the questionnaire used to measure these perceptions. The results of the study indicate that ChatGPT is positively valued for its ability to improve productivity, save time and facilitate the understanding of theoretical concepts. However, students in Arequipa show a greater appreciation for the specialized applications and personalization of learning with ChatGPT, compared to students in Lima. Despite these positive perceptions, both groups agree in pointing out limitations, such as inaccuracy in the citation of scientific sources and the generation of inaccurate references. The psychometric analysis confirmed the excellent reliability of the questionnaire, with a McDonald’s Omega of 0.856, and validated three key factors: usefulness and efficiency, specialized applications and customization, and limitations and risks. In conclusion, although ChatGPT is perceived as a useful tool, its limitations need to be addressed to optimize its impact on learning. The study suggests future research in other disciplines and educational contexts, as well as a deeper analysis of the long-term risks of using artificial intelligence in academia.

Ähnliche Arbeiten