Artificial intelligence-enhanced phenomenon-based learning: An embedded mixed methods design example

The present study is aim to examine the effects of the artificial intelligence (AI)-enhanced phenomenon-based learning (PhenoBL) approach on students' critical thinking and problem-solving skills, as well as to identify the strengths and weaknesses of this approach. In line with this aim, the study was conducted using an embedded mixed-methods design. The quantitative dimension of the embedded design followed a single-group pre-test–post-test model, while the qualitative dimension employed a case study approach. The research was carried out with 24 second-grade primary school students. In the quantitative dimension, the data collection tools included the "Critical Thinking Disposition Scale for Primary School Students," the "Problem Solving Inventory for Children," and a "Personal Information Form." For the qualitative dimension, a "Participant Observation Form" and a "Self-Assessment Form" were used. Additionally, field notes were recorded by the researcher throughout the entire implementation process. Following the implementation, it was found that the AI-enhanced phenomenon-based learning approach significantly improved students' overall critical thinking disposition scores. Observations made during and after the intervention also revealed notable improvements in students' critical thinking skills. Specifically, in the subdimension of skepticism, the difference between pre- and post-test scores was statistically significant. Weekly observation data showed that skepticism-related codes increased both numerically and in depth over time. In the open-mindedness subdimension, a statistically significant increase was observed between the pre- and post-tests. Qualitative data also indicated that open-mindedness codes not only increased in frequency but evolved from simple acceptance to more sophisticated and inclusive perspectives. In the curiosity subdimension, post-test scores were significantly higher than pre-test scores. Qualitative analysis revealed that while curiosity in the first and second weeks was limited to simple information seeking and tool exploration, by the sixth and eighth weeks, both the number and depth of questions had increased. Students began to explore mechanisms, historical origins, interdisciplinary contexts, and design alternatives. Quantitative analysis also showed a statistically significant improvement in the objectivity subdimension. Qualitative findings supported this increase, indicating a steady rise in the number of students generating objective arguments, as well as an expansion in the scope and depth of evidence used. In terms of total problem-solving scores, the post-test improvement was not statistically significant. However, qualitative findings revealed the opposite: more students participated in the problem-solving process, and the quality of the proposed solutions improved. In the problem-solving confidence subdimension, quantitative data showed a statistically significant increase, supported by qualitative findings. For the self-regulation subdimension, quantitative and qualitative results diverged. While students rated themselves lower in self-regulation in the post-test, qualitative data showed that more students demonstrated democratic role-sharing, created communication rules, maintained long-term focus, and exhibited task-centered behaviors as the weeks progressed. There was no statistically significant change in avoidance scores according to the quantitative data. However, qualitative observations indicated a behavioral decrease in avoidance tendencies. The strengths of the AI-enhanced phenomenon-based learning approach were clustered under: a climate of critical inquiry, creative and collaborative problem-solving, interdisciplinary transfer, increased motivation, and improved self-assessment. The weaknesses were categorized under: technical and infrastructural limitations, student resistance, and pedagogical design challenges. Artificial Intelligence, Phenomenon Based Learning, Criticial Thinking, Problem Solving