AI-enabled Qualitative Research: Participatory Evaluation

Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents. According to the PLOS ONE case study (published August 20, 2026), strategy mapping software guided a four-workshop participatory evaluation with 15 healthcare professionals and produced ten participant-developed recommendations. The primary lesson for AI-enabled qualitative research is that structured, visual mapping plus lightweight AI support can increase stakeholder ownership and produce actionable outputs if recommendation clarity and facilitator involvement are managed. This post explains what the PLOS ONE study did, gives the concrete numbers and quotes you can cite, and maps those findings to practical AI-enabled qualitative research workflows for evaluation teams.

Key Takeaways

According to PLOS ONE (published August 20, 2026), a multi-stage case study used strategy mapping software across four workshops with 15 participants between April and June 2023 and generated ten recommendations that participants largely considered useful.

15 participants: the study involved all 15 employees of a communicable disease unit and ran four workshops on April 3, April 6, April 24, and June 12, 2023, according to PLOS ONE (Aug 20, 2026).
10 recommendations: the workshops produced ten participant-developed recommendations, and the study conducted ten follow-up interviews in spring 2024 to assess perceived use, according to PLOS ONE (Aug 20, 2026).
AI-assisted translation: the authors used a large language model (M365 Copilot, GPT-5) to translate and recreate strategy maps from Swedish to English, with author verification, as reported in PLOS ONE (Aug 20, 2026).
Trade-offs noted: interviewees praised the method’s structure but flagged recommendations as sometimes vague and four workshops as potentially too long, according to PLOS ONE (Aug 20, 2026).

What Happened / How the PLOS ONE study worked

The PLOS ONE study used the strategy mapping methodology inside browser-based Strategyfinder software to run a participatory evaluation of pandemic crisis management, and the study reports concrete dates, counts, and methods.

According to PLOS ONE (published August 20, 2026), the evaluation targeted one Swedish regional communicable disease and infection control unit and involved all 15 staff members across four workshops between April 3 and June 12, 2023.

According to PLOS ONE (Aug 20, 2026), the workshop process was: blind individual idea-gathering on April 3, 2023, collaborative causal linking and subsystem creation on April 6 and April 24, 2023, and an in-person validation session on June 12, 2023, producing ten final recommendations.

According to PLOS ONE (Aug 20, 2026), the authors conducted semi-structured follow-up interviews with ten of the original 15 participants in spring 2024, transcribed with speech-to-text and then manually corrected, and analyzed via directed content analysis.

The PLOS ONE authors explicitly reported AI use: "a large language model (M365 Copilot, based on GPT-5 architecture) was used to translate statements from Swedish to English and to recreate the visual structure of the original strategy maps, " with outputs verified by the authors.

Findings Snapshot

Date (absolute)	Metric	Value	Implication
April 3, 2023	First workshop (blind gather + causal mapping)	15 participants, initial risk system created	Individual-to-collective elicitation generated a visible causal map for later analysis
April 6, 2023	Second workshop (validation)	Five subsystems created from centrality analysis	Validation tightened clusters but authors later questioned if facilitator-led validation could shorten process
April 24, 2023	Third workshop (recommendation development)	Participants generated recommendations and rated effectiveness/feasibility	Interactive rating produced a shortlist of candidate recommendations for validation
June 12, 2023	Fourth workshop (final validation)	Ten final recommendations presented in the evaluation report	Final validation improved participant buy-in but length raised time-cost concerns
Spring 2024	Follow-up interviews	10 interviews conducted, analyzed	Majority of interviewees reported that several recommendations had been started or used, but views on clarity and novelty varied

Implications for evaluation researchers and UX/qual teams

Participatory mapping software can increase ownership and produce actionable recommendations but teams must manage clarity, facilitation, and time investment.

Design decision: According to PLOS ONE (Aug 20, 2026), four workshops produced richer causal maps and learning but some participants found the cadence too slow; evaluation teams should balance depth with calendar constraints.
Facilitator role: According to PLOS ONE (Aug 20, 2026), interviewees recommended more facilitator involvement in polishing recommendations for clarity and actionability, suggesting a hybrid approach where participants draft and evaluators refine.
Scalability: According to PLOS ONE (Aug 20, 2026), participant ratings (importance, feasibility, effectiveness) were helpful and can be adopted even without full strategy mapping software to prioritize follow-up work.

How Evidano helps in AI-enabled qualitative research

Problem: Slow transcription and error-prone translation → Solution: fast, verifiable transcripts

Evidano integrates speech-to-text and translation tools that accelerate multi-session workshop transcription and bilingual map verification, reducing manual correction time.

For teams replicating the PLOS ONE workflow, Evidano’s transcription features can replace multi-step speech-to-text editing; see Evidano speech-to-text for details.

Problem: Manual coding and synthesis across sessions → Solution: thematic + cross-segment analysis

Evidano automates thematic coding, frequency counts, and cross-segment comparisons so facilitators can quickly see which risks or recommendations are most central across participants.

Evidano’s document ingestion means facilitators can load transcripts, meeting exports, and the strategy map text and get an analyst-ready codebook in hours rather than days; see Evidano features.

Problem: Unclear recommendations → Solution: AI-assisted rewrites with human verification

Evidano offers AI-assisted summarization and recommendation polishing that preserves participant voice while increasing clarity and actionability, enabling the facilitator to refine participant drafts and then re-validate them with stakeholders.

Evidano’s AI chat over your documents helps facilitators generate candidate wording, then collect quick participant approvals, which addresses the PLOS ONE recommendation to increase facilitator involvement without undermining ownership.

FAQ: AI-enabled qualitative research

What is AI-enabled qualitative research and why use it for participatory evaluation?

AI-enabled qualitative research uses AI tools to speed transcription, translation, coding, and synthesis while keeping human-in-the-loop verification.

According to PLOS ONE (Aug 20, 2026), the case study used an LLM for translation and author verification, illustrating a useful pattern: AI for mechanical tasks, humans for judgment and actionability.

Can AI replace the facilitator in participatory mapping?

No, AI cannot replace the facilitator for stakeholder ownership and nuanced recommendation crafting.

According to PLOS ONE (Aug 20, 2026), authors recommend increased facilitator involvement to improve clarity, and the study used AI only for translation and figure recreation, with authors verifying outputs.

How many workshops are needed to run a participatory strategy mapping evaluation?

Two to four workshops are common, but the exact number depends on depth and time constraints.

According to PLOS ONE (Aug 20, 2026), the four-workshop sequence produced deep reflection and ten recommendations, but several interviewees suggested a shorter two-workshop variant could preserve benefits while saving time.

How should teams measure whether recommendations were used?

Define 'use' up front and collect both process-use and instrumental-use indicators.

According to PLOS ONE (Aug 20, 2026), the authors found variance in interviewee responses because the study did not pre-define what counted as use; they recommend explicit operational definitions and structured follow-ups.

Conclusion & Next Steps

The PLOS ONE case study (published August 20, 2026) shows that strategy mapping software plus participatory workshops can produce ten stakeholder-owned recommendations and measurable learning, but clarity and facilitator polishing matter.

If you plan a participatory evaluation, define use metrics up front, decide facilitator polishing points, and consider AI for transcription and translation with human verification as the PLOS ONE team did.

To prototype the workflow described here (fast transcription, thematic coding, recommendation polishing, and stakeholder re-validation) try a hands-on trial with automated tools.

For a fast start, explore how Evidano handles multi-session transcripts and coding; when you are ready, Try Evidano for free.

Topics

AI-enabled qualitative research
participatory evaluation
strategy mapping software
qualitative analysis for crisis management

Keep reading

Browse all articles