This (22 page PDF) is a "mixed methods analysis of 112 student reflections from a 12-week course." As the abstract states,"we examined experiences with GenAI tools... Sentiment analysis showed 86% positive responses, though ethical concerns generated significant negative sentiment (62%). Findings demonstrate that GenAI, when pedagogically scaffolded, augments rather than replaces human judgment. Students evolved from passive users to critical evaluators, developing strategies for bias detection and source validation." This kind of analysis is a super-common type of publication, is obviously a narrow snapshot, and when reading this type of article I evaluate it as an opinion piece (where the opinions are those of the students, which are being collected and summarized by the authors). Because the sample is so small, the percentages mean nothing, and what I do is sort the responses into buckets of 'possible responses to AI'. The literature review is also a useful if opinionated summary of previous work pointing to mechanisms for identifying and categorizing the discussion; I liked the mapping of generative AI integration into design thinking stages (illustrated). This is not nothing, and there's value in collecting a wide range of opinions.
Today: Total: [Direct link] [Share]
