The Effect of ChatGPT Integration on Students' Scientific Argumentation Skills in Genetics Learning

Abstract

Scientific argumentation is an essential competency in Biology education, particularly in genetics, which requires students to analyze evidence, construct logical reasoning, and draw conclusions based on scientific concepts. However, students' scientific argumentation skills remain relatively low, highlighting the need for innovative instructional approaches that effectively facilitate scientific reasoning. This study aimed to analyze the effect of ChatGPT integration on students' scientific argumentation skills in genetics learning. The study employed a quantitative approach using a quasi-experimental design with a Nonequivalent Control Group Design. The research sample consisted of two twelfth-grade science classes at MAS Firdaus Tompobalang, designated as the experimental and control groups. Data were collected through a Toulmin-based scientific argumentation test, classroom observations, questionnaires, and documentation. The data were analyzed using descriptive and inferential statistics, including the Independent Samples t-test, Paired Samples t-test, normalized gain (N-Gain), and effect size analysis. The results revealed that the experimental group achieved a mean N-Gain score of 0.65, which was higher than that of the control group (0.35). Hypothesis testing indicated a statistically significant difference (p = 0.001 < 0.05) with a Cohen's d value of 0.96, indicating a large effect size. Improvements were observed across all components of scientific argumentation, including claim, evidence, reasoning, backing, and rebuttal, although the rebuttal component still requires further enhancement. Therefore, the integration of ChatGPT had a positive and significant effect on students' scientific argumentation skills and is recommended as an innovative approach to Biology instruction, while maintaining the teacher's central role in guiding the critical and responsible use of artificial intelligence.