Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
AI-CHATGPT, STUDENTS’ ENGAGEMENT AND PROBLEM-SOLVING ABILITY IN UNDERGRADUATE PHYSICS: AN EXPERIMENTAL STUDY
0
Zitationen
2
Autoren
2026
Jahr
Abstract
The integration of Artificial Intelligence (AI) into science education can transform instructional methods by providing supporting, interactive and personalized learning experiences. This study examines the effects of AI-ChatGPT on students’ engagement and problem-solving ability in physics at the undergraduate level. A quasi-experimental design was adopted, with a study sample of 115 physics students divided into an experimental group (ChatGPT-Integrated Instruction, CGPT-II) and a control group (Traditional Lecture Method, TLM). Data were collected using Pre-Engagement Scale, Post-Engagement Scale, Pre-Problem Solving Test, and Post-Problem-Solving Test, with reliability coefficient values, α = 0.76, 0.78, 0.79, and 0.81, respectively. Collected data were analyzed using descriptive, t-test and Analysis of Covariance (ANCOVA) statistics at a 0.05 significance level. Results showed that students exposed to CGPT-II demonstrated significantly higher engagement scores (MS = 3.98, SD = 0.62) compared to the control group (MS = 2.77, SD = 1.06) with a t-test value of t(113) = 7.49, p &lt; 0.05. Similarly, the experimental group outperformed the control group in problem-solving ability (MS = 71.61%, SD = 6.80) compared to the control group (MS = 59.40%, SD = 8.52), yielding t(113) = 8.52, p &lt; 0.05. ANCOVA analysis indicated a significant main effect of the teaching method on post-test engagement {F(1, 113) = 27.67, p &lt; 0.001, partial η² = 0.20} and problem-solving ability {F(1,112) = 69.48, p&lt; 0.001, partial η² = 0.382} scores between the experimental and control groups. These findings indicate that ChatGPT significantly enhanced both engagement (behavioral, emotional, and cognitive) and problem-solving ability of students in physics. It is therefore recommended that school authorities should organize training workshops and seminars for science lecturers on the use of AI-ChatGPT, and support them with the required facilities.<p> </p><p><strong> Article visualizations:</strong></p><p><img src="/-counters-/soc/0807/a.php" alt="Hit counter" /></p>
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.422 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.300 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 7.734 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.781 Zit.
Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
2018 · 5.519 Zit.