Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
The Relationship between ChatGPT reliance, critical use, and critical reading skills among Indonesian undergraduate students
0
Zitationen
4
Autoren
2026
Jahr
Abstract
The increasingly frequent use of ChatGPT in higher education raises questions about its impact on how students critically read a range of academic texts. However, empirical evidence examining the relationship between the dependence and critical use of ChatGPT and college students' critical reading ability is limited. Three aspects were investigated in this study: students' dependence on ChatGPT, their critical use of ChatGPT, and how these relate to students' critical reading skills. This study aimed to examine university students, specifically 79 students at a public university in Indonesia. Students were given a critical reading test, which was in the form of an essay, and there was also a questionnaire to measure students' dependency and critical use of ChatGPT. The results were descriptive, showing that the students' level of ChatGPT dependency was moderate (mean = 3.10) and their level of critical use was quite good (mean = 3.87). For critical reading scores, results ranged from 31.25 to 92.19, with an average in the moderate category. The results of multiple linear regression show that neither critical use nor dependence on ChatGPT has a statistically significant relationship with students’ critical reading skills (β = 0.164, p = 0.151; β = −0.071, p = 0.530). Overall, these results indicate that students' engagement with ChatGPT has not had a clear or significant effect on their critical reading performance. Further research is needed to confirm these findings. These findings indicate that reliance on AI and critical use of ChatGPT alone may not be sufficient to influence students' critical reading abilities, suggesting that other factors may also play a role. Therefore, future research should involve a more diverse population and develop more specific, contextually relevant instruments, particularly those linking ChatGPT use to critical reading skills, while also considering other mediating variables.
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.316 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.177 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 7.575 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.776 Zit.
Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
2018 · 5.468 Zit.