The effect of artificial intelligence supported coding training on computational thinking skills and attitude towards coding in gifted students

This study aims to examine the impact of artificial intelligence-supported applications developed with MIT App Inventor on gifted students' computational thinking skills and attitudes towards coding. The research employs a single-group pre-test/post-test experimental design, a quantitative research method. The study group comprises students from the Individual Talent Recognition (BYF) and Special Talent Development (ÖYGP) programs at Tokat Turhal Science and Art Center during the 2023-2024 academic year. The research was conducted over an 8-week period, with 2-hour sessions per week, totaling 16 hours of instruction. Data collection instruments include the "Computational Thinking Scale (for Middle School Level)" adapted by Korkmaz et al. (2015) and the "Attitude Towards Coding Scale" developed by Akkuş et al. (2019). The study group consists of 28 gifted students (13 females, 15 males). Normality tests were conducted for data analysis. According to Shapiro-Wilk values, both pre-test and post-test scores for Computational Thinking Skills demonstrated normal distribution, while for Attitude Towards Coding, the pre-test showed normal distribution, but the post-test did not. Based on these results, Kruskal Wallis H, t-test, and ANOVA tests were applied to analyze the responses to the Attitude Towards Coding and Computational Thinking Skills scales. Findings reveal significant improvements in students' attitudes towards coding (p = .001) and computational thinking skills (p = .000) between pre-test and post-test scores. Regarding attitudes towards coding, 6th and 7th-grade students showed higher mean ranks in terms of grade level. However, variables such as grade level, computer usage duration (in years), and daily computer usage time did not yield statistically significant differences. In terms of computational thinking skills, 7th-grade students scored higher in several sub-dimensions. Significant differences among grade levels were particularly observed in algorithmic thinking, collaboration, and critical thinking sub-dimensions. These results indicate that computational thinking skills vary by grade level, with a notable increase in the 7th grade. However, variables such as grade level, computer usage duration (in years), and daily computer usage time did not produce statistically significant differences.