From Workshops to Insight: AI Qualitative Data Analysis

According to PLOS One, a 2026 co-development study used qualitative workshop data to build an Implementation Research Logic Model for whole-school physical activity in special education settings.

The study shows that AI qualitative data analysis can support, but should not replace, theory-informed human interpretation when researchers must connect stakeholder experiences to implementation decisions.

• The November 13, 2026 PLOS One study involved 16 participants from six special schools and related education and community organizations in Northern Ireland.
• The researchers identified 31 implementation determinants, including 18 facilitators, 12 barriers, and one factor that functioned as both a barrier and a facilitator.
• The researchers converted the qualitative findings into eight pragmatic implementation strategies and eight causal mechanisms.
• The study identified 11 implementation outcomes, seven school-level outcomes, and six pupil-level outcomes.
• The researchers used deductive content analysis guided by CFIR 2.0, ERIC, the Implementation Outcomes Framework, COM-B, and Normalization Process Theory.

The 2026 PLOS One study used two four-hour, in-person co-development workshops to examine how a whole-school approach to physical activity could work in special education settings.

The study included parents, teachers, classroom assistants, school leaders, Education Authority Northern Ireland representatives, and community organization professionals. The researchers recorded and transcribed workshop discussions, combined transcripts with flipchart mind maps and fieldnotes, and applied deductive content analysis.

The research team coded the material against predefined theoretical domains, grouped similar responses, and resolved coding disagreements through discussion. The researchers then mapped determinants to CFIR 2.0, matched determinants with ERIC strategies, and classified outcomes using the Implementation Outcomes Framework.

The authors described the resulting model as a way to connect determinants, strategies, mechanisms, and outcomes. The study states that "The model illustrates the interactions and causal mechanisms between implementation determinants and implementation strategies to achieve the intended outcomes."

The researchers also cautioned that the model should be treated as "a flexible framework rather than a prescriptive model, " because special education systems and stakeholder priorities differ across settings.

The 2026 PLOS One study suggests that education researchers should analyze stakeholder data across multiple levels instead of treating implementation as an individual behavior problem.

The study identified staff training, collaborative networks, family engagement, adaptive resources, and physical activity champions as important parts of implementation planning. Researchers can therefore use qualitative analysis to ask not only what stakeholders think, but also which conditions may enable a program to become part of routine practice.

The study also demonstrates why framework-aware coding matters. CFIR 2.0 helped organize determinants, ERIC helped identify implementation strategies, COM-B helped explain individual behavior mechanisms, and Normalization Process Theory helped explain the social work required to embed new practices.

AI-enabled qualitative research can make this workflow more manageable by grouping similar statements, comparing stakeholder segments, surfacing recurring barriers, and tracing relationships between codes and subcodes. Human researchers still need to validate interpretations, examine context, and protect the meaning of participant accounts.

The study's Northern Irish setting limits direct generalization. The authors noted that the findings should be adapted with local stakeholders before use in other education systems, and the source did not include pupils directly in the co-development workshops.

The November 13, 2026 PLOS One study shows how structured qualitative analysis can turn stakeholder workshops into an implementation logic model with 31 determinants, eight strategies, eight mechanisms, and 24 outcomes.

AI qualitative data analysis can help researchers manage similar evidence bases, compare stakeholder perspectives, and connect themes to implementation frameworks without removing human oversight.

Researchers can use Evidano to analyze transcripts, documents, and survey responses in one qualitative workflow, then validate findings against source passages and stakeholder context. Start your next project with Try Evidano for free.