Key Findings: Qualitative Analysis of Design Thinking in STEM

According to the PLoS One study published on November 2, 2026, design thinking was integrated unevenly into mathematics-focused STEM lesson plans created by 55 senior pre-service mathematics teachers. The qualitative analysis found that Define, Ideate, and Prototype were common, while Empathy and Test were substantially less developed.

• The November 2, 2026 PLoS One study analyzed 24 collaboratively created lesson plans and 55 individual open-ended questionnaires.
• The study found that Define appeared in 95.8% of lesson plans, Prototype in 91.7%, and Ideate in 83.3%.
• The Empathy phase appeared in only 20.8% of lesson plans, while Test appeared in 54.2%.
• The study reported that 18 of 24 lesson plans included mathematics as a central STEM feature, while 18 included creative application.
• The study found that 43 participants believed mathematics teachers should be able to create design thinking lesson plans, but 30 participants identified time consumption as a major disadvantage.

The PLoS One study used a qualitative case study design to examine one cohort of senior pre-service mathematics teachers enrolled in a 15-week course during the 2021–2022 academic year. The research combined group lesson-plan artifacts with individual reflections so the analysis could compare planned teaching practices with participants’ reported experiences.

According to the study’s methods section, the researchers analyzed 24 lesson plans with a STEM integration rubric and coded the presence and quality of five design thinking phases: Empathy, Define, Ideate, Prototype, and Test. Two raters independently assessed the lesson plans, reaching 93.1% initial agreement before resolving discrepancies through discussion.

The researchers analyzed questionnaire responses through four inductive stages: familiarization, initial coding, category development, and theme construction. A second researcher independently coded 25% of the questionnaire data, representing 14 of the 55 participants across eight prompts, and the initial agreement was 90%.

The study distinguished between a phase being mentioned and a phase being meaningfully implemented. A robust phase required evidence that mathematical reasoning, quantitative analysis, or evidence-based justification shaped the design activity, rather than appearing only as a label.

The PLoS One study indicates that STEM teacher education should assess not only whether design thinking phases are named, but also whether each phase produces substantive mathematical reasoning. A lesson plan can include a Test heading without requiring students to collect data, measure performance, or revise a prototype.

The study’s results suggest that researchers should analyze design thinking as an iterative process rather than a checklist. The predominance of Define, Ideate, and Prototype, combined with the lower occurrence of Empathy and Test, was interpreted by the study as evidence of a tendency toward product completion rather than sustained user-centered refinement.

The participant responses also show why mixed qualitative datasets matter. The PLoS One study found that participants selected real-life problems and valued student engagement, yet they reported difficulty aligning disciplines, choosing materials, designing assessments, and fitting design work into curriculum time.

For qualitative researchers, the study offers a practical analysis pattern: code artifacts for observable practices, code participant responses for perceived barriers and benefits, then triangulate the two datasets. AI can accelerate the repetitive stages of this workflow, but researchers still need to define codes, inspect context, review counterexamples, and document analytic decisions.

The November 2, 2026 PLoS One study shows that design thinking can support mathematics-centered STEM education, but the quality of integration depends on whether teachers move beyond naming phases to implementing empathy, testing, measurement, and iteration.

The study’s 55-participant qualitative case and 24 lesson plans also demonstrate the value of combining document analysis with individual reflections. AI-enabled qualitative research can make that triangulation more manageable by bringing thematic, frequency, and cross-segment analysis into one workflow.

Researchers who want to analyze education documents and open-ended responses more efficiently can try Evidano for free.