Analizzare dataset ampi e non strutturati nella ricerca educativa con il supporto dell’Intelligenza Artificiale. Sfide e problematiche

In the educational field, there is a need to make sustainable the analysis of large datasets, which often contain unstructured texts and materials. In this research, we investigate the designs of 560 students in the third year of SdFP. The study aims to capture how the unexpected impacts action and validate the “design for the unexpected” proposal. For the analysis, we implemented three approaches. The first one involves two researchers reading and using the phenomenological method, the second implies an automatic term search method, and the third the use of ChatGPT. The paper will analyze the pros and cons of each mode and grasp how the different device impacts sustainability, research purposes, and the evidence gathered. Some issues that emerged from the work and some guidelines will be shared