Better Learning Insights: AI Qualitative Analysis

According to the PLOS One study on physical exhibits and medical student learning, combining quantitative scores with qualitative student accounts showed that a hands-on physiology project was associated with higher post-test performance and perceived gains in teamwork, application, and conceptual understanding.

AI qualitative analysis can help medical education researchers code open-ended feedback, compare themes across student groups, and connect participant explanations with quantitative outcomes, but the PLOS One study itself used researcher-led deductive and inductive thematic analysis rather than AI.

• In a study conducted between February 17, 2023, and October 31, 2023, 50 first-year MBBS students in Rajkot, Gujarat, created four physical physiology exhibits in four teams.
• According to the PLOS One authors, median post-test scores increased from 12, with an interquartile range of 10 to 14, to 16, with an interquartile range of 15 to 20, and the improvement was statistically significant at p < 0.01.
• The qualitative component included open-ended feedback from all 50 students and a 96-minute focus group discussion with 10 purposively selected students.
• The researchers used a predefined framework of enabling factors, hindering factors, and learning gains for the open-ended responses, then used inductive thematic analysis for the focus group transcript.
• The source reported that collaboration, faculty support, application of theory to practice, planning difficulties, and material procurement were central parts of the student experience.

The PLOS One study found that creating physical physiology exhibits was associated with improved higher-order cognitive test scores and positive student perceptions. The intervention took place in the Department of Physiology at a tertiary care hospital in Rajkot, Gujarat, India, between February 17, 2023, and October 31, 2023.

The researchers included all 50 participating first-year MBBS students through convenience sampling. The students were divided into four teams of 12 or 13 students, and the teams created exhibits about ascending sensory pathways, light and conjunctival reflexes, the micturition reflex, and the automatic bladder.

The researchers administered a validated 30-item multiple-choice test before and after the three-week exhibit-building intervention. The PLOS One article reported median scores of 12 before the intervention and 16 afterward, with p < 0.01 using a Wilcoxon signed-rank test.

The study used an explanatory sequential mixed-methods design. The quantitative results were collected and analyzed first, and the qualitative focus group was then used to explain students' experiences in greater depth.

The authors summarized the learning value through a student quotation: "Learning through such tasks is way better and easier than anything else because this is an application-based learning method." The quotation connects the numerical improvement with the students' interpretation of how learning occurred.

The study does not establish that physical exhibits alone caused long-term improvements in clinical reasoning. The authors identified limitations including the absence of a control group, the lack of long-term retention measures, and the absence of individual-level outcome assessment within the team-based intervention.

The PLOS One study shows how medical education research becomes more informative when test-score changes are interpreted alongside student accounts of collaboration, application, planning, and learning. AI qualitative analysis can reduce the manual burden of organizing open-ended feedback and focus-group transcripts while preserving a transparent link between themes and source evidence. Researchers should use AI as an auditable assistant, not as an independent decision-maker, and should protect participant privacy throughout the workflow. Explore the evidence in your own interviews, surveys, and documents with Try Evidano for free.