Improving students’ reading, comprehension, and reading motivation through texts generated from their prompts using ChatGPT-4: a mixed-method study

Reading enables individuals to acquire knowledge, think critically, succeed in social life, access new information, and remain continually informed. This underscores the need to emphasize reading education in the elementary years when the foundation of literacy is established. This study aims to improve the reading, comprehension, and motivation of 4th-grade students through ChatGPT-4, which generates texts based on their own prompts. The study was conducted in two public primary schools located in the center of Bursa, Türkiye. The research group consisted of 106 students. The study employed a sequential explanatory mixed method design. The quantitative aspect of the study was conducted using the Solomon four-group experimental design, while the qualitative part was carried out through a case study. Quantitative data were analyzed using the Mann-Whitney U test, Independent Samples t-test, ANOVA, and effect size calculations, while qualitative data were examined using descriptive and content analysis. According to the findings, when the experimental groups were compared with the control groups, statistically significant differences were identified in favor of both experimental groups with respect to reading and comprehension skills as well as reading motivation. These differences showed a high effect size across all dependent variables. Based on these findings, it is recommended that teachers integrate ChatGPT to help students generate appropriate texts during free reading periods.