What AI Qualitative Analysis Reveals About STEM Design Thinking

According to the PLOS One study, "Pre-service mathematics teachers’ integration of design thinking in STEM lesson plans", pre-service mathematics teachers commonly included the Define, Ideate, and Prototype phases of design thinking, but they used the Empathy and Test phases much less consistently.

The study suggests that AI qualitative analysis can help STEM researchers compare lesson-plan artifacts with participant reflections, identify uneven phase coverage, and connect numerical patterns with the reasons participants give for those patterns.

• According to the November 2, 2026, PLOS One study, 55 senior pre-service mathematics teachers participated in a qualitative case study at a state university in Türkiye.
• According to the study, the researchers analyzed 24 collaborative STEM lesson plans and questionnaire responses from all 55 participants after a 15-week course during the 2021 to 2022 academic year.
• According to the study, Define appeared in 95.8% of lesson plans, Prototype in 91.7%, and Ideate in 83.3%, while Empathy appeared in only 20.8% and Test in 54.2%.
• According to the study, 43 participants believed mathematics teachers should be able to develop design-thinking lesson plans, while 30 participants identified time consumption as a disadvantage.
• According to the study, independent coders reached 93.1% agreement when evaluating lesson-plan STEM features and 90% initial agreement when coding a 25% sample of questionnaire responses.

The November 2, 2026, PLOS One study found that pre-service mathematics teachers integrated design thinking into STEM lesson plans unevenly rather than applying all five phases as a sustained, iterative cycle.

The study examined five design-thinking phases: Empathy, Define, Ideate, Prototype, and Test. The authors treated design thinking as non-linear, meaning that testing can send learners back to problem definition or prototyping for further refinement.

The study found that lesson plans often emphasized product completion. Define, Ideate, and Prototype formed the dominant procedural core, while Empathy was frequently omitted and Test was often treated as a final check rather than as a source of evidence for revision.

The study’s qualitative analysis combined two data sources: 24 group-produced lesson plans and individually completed open-ended questionnaires. The researchers used an inductive four-stage thematic process consisting of familiarization, initial coding, category development, and theme construction.

Participants described both value and constraint. PMT 25 said, "The most challenging part was aligning the design with other disciplines and finding suitable learning outcomes." PMT 16 added, "I think design is necessary, but it should not take too much time because there is a program that needs to be covered."

The study reported that participants selected topics primarily around real-life problems and student learning. The most frequent topic-selection codes were problems from participants’ own experience, with 34 mentions, and student learning, with 19 mentions.

The November 2, 2026, PLOS One study indicates that STEM researchers should analyze both the presence and the quality of design-thinking phases.

A frequency count alone could show that a lesson plan contains a Test phase, but the study’s coding protocol required substantive implementation, such as measurement, quantitative analysis, or evidence-based prototype refinement. AI qualitative analysis should therefore distinguish nominal mentions from operationalized practices.

The study also shows why researchers should triangulate documents and participant accounts. Lesson plans revealed where phases appeared, while questionnaire responses helped explain barriers involving topic selection, discipline alignment, materials, timing, and assessment.

For teacher educators, the study suggests that scaffolding should focus especially on Empathy and Test. Participants understood the value of authentic, student-centered learning, but many plans did not translate that belief into user inquiry or iterative optimization.

• Use code definitions that separate a phase being named from a phase being enacted.
• Compare document frequencies with themes from interviews or questionnaires before drawing conclusions.
• Track negative cases, such as lesson plans that included mathematics but did not use mathematical reasoning to test or improve a design.
• Preserve quotations alongside codes so that claims about barriers remain connected to participant language.
• Treat AI-generated themes as an analytical aid that requires researcher review, not as a replacement for methodological judgment.

The November 2, 2026, PLOS One study shows that design-thinking research in mathematics education benefits from combining frequency analysis, thematic interpretation, document analysis, and participant quotations. The study’s most important analytical distinction is between naming a design-thinking phase and implementing that phase through meaningful mathematical reasoning, user inquiry, testing, and refinement. AI qualitative analysis can make these comparisons faster and more traceable when researchers keep the codebook, source excerpts, and methodological boundaries visible. Researchers analyzing STEM lesson plans, teacher reflections, or open-ended education surveys can try Evidano for free.