Evidano is an AI-powered qualitative data analysis platform that can ingest transcripts, auto-transcribe and auto-translate, and deliver thematic, frequency, and cross-segment analyses to accelerate Phase 2–3 interpretation. Problem: refugee homestay hosting is expanding in Canada but evidence on why citizens host, what supports they need, and how to sustain hosting is thin. Payoff: this post breaks down the July 13, 2026 PLOS One protocol "Why do Canadians host refugees? " and shows a reproducible AI-enabled qualitative analysis workflow you can run on interview transcripts, survey open-ends, and co-design notes. The 36-month study (published 13 July 2026) combines a national survey (target n≥200), 20–25 interviews, and group concept mapping co-design. Read the protocol on PLOS One. If you already collect transcripts, use Evidano to ingest them and accelerate analysis.
Key Takeaways
Evidano is an AI-powered qualitative data analysis platform that speeds transcription, coding, synthesis, and visualization for mixed-methods refugee-hosting research. The PLOS One protocol (published 13 July 2026) outlines a 36-month sequential explanatory mixed-methods study combining a national survey (target n≥200), 20–25 interviews, and group concept mapping to co-design host supports.
- PLOS protocol design: 36-month sequential explanatory mixed-methods study, with Phase 1 survey, Phase 2 interviews, and Phase 3 group concept mapping and co-design.
- Survey and timeline: target survey n≥200 to support SEM, recruitment runs 1 Sept 2026 to 1 Sept 2028, survey results expected March 2027.
- Practical acceleration: Evidano ingests LimeSurvey exports and interview audio to produce auto-themes, codebooks, cross-segment comparisons, and exportable visuals for rapid synthesis.
Fast take: what the PLOS protocol sets out to do
The PLOS protocol sets out to run a 36-month sequential explanatory mixed-methods study to map relationships and co-design supports for refugee hosts. The study protocol (Al-Hamad et al., published 13 July 2026) proposes Phase 1 = national survey of current/former hosts, Phase 2 = 20–25 semi-structured interviews, and Phase 3 = group concept mapping and co-design with 20–25 participants, with key aims to map how hospitableness, empathy, attitudes, readiness, satisfaction, and advocacy relate and to co-design an actionable host support toolkit and policy recommendations.
- Study start/end details: recruitment runs 1 Sept 2026 → 1 Sept 2028; survey results expected March 2027.
- Sample & power: target survey n≥200 to support structural equation modeling.
- Ethics & funding: REB# 2026−152; SSHRC Insight Grant #435-2026-1595.
Study snapshot (quick reference)
| Date / Metric | Value | Source / Note | Implication for qualitative researchers |
|---|---|---|---|
| Published | 13 July 2026 | PLOS One protocol | Use protocol details to align timelines and ethics submissions |
| Design | 36-month sequential mixed-methods | Phase 1→2→3 | Sequential design lets survey results shape interview guides |
| Phase 1 target | n ≥ 200 | Power for SEM; pilot n≈15 | Plan quantitative-driven coding priorities for open-ends |
| Phase 2 sample | 20–25 interviews | Purposive; hosts + service providers | Expect thematic saturation; prepare rich coding schema |
| Phase 3 | 20–25 co-design participants | Group concept mapping + 4 sessions | Produce prioritized, actionable clusters for toolkits |
What the protocol does and why it matters
The protocol pairs a national LimeSurvey questionnaire with interviews and group concept mapping so quantitative associations can guide qualitative probing and feed participatory co-design. The team will run a national LimeSurvey questionnaire (online/phone/paper), analyze quantitative measures (SPSS, SmartPLS) and open-ends (NVivo), then follow with interviews transcribed verbatim and group concept mapping (GroupWisdom).
Measures of interest: hospitableness, attitudes toward refugees, empathy, readiness to assist, satisfaction, and advocacy.
Analysis plan: SEM for mediation testing (α=0.05), thematic analysis (NVivo) with inductive and deductive coding, and multidimensional scaling plus hierarchical clustering for concept mapping.
Practical outputs promised: a host support toolkit, service guidelines, and policy recommendations.
So what for qualitative researchers and program teams
Researchers (methods & reproducibility)
Researchers should use the survey to surface high-frequency issues and adapt interview guides for explanatory follow-up. The sequential explanatory design is ideal when you need measurement plus contextual explanation, so use the survey to surface high-frequency issues (for example, boundary-setting or trauma supports) and adapt interview guides to probe causal stories.
Prepare a codebook that maps to the survey constructs (hospitableness → codes for welcome practices; empathy → reflexive narratives) to enable mixed-methods integration.
Service providers & program designers
Service providers should replicate the protocol's co-design model to produce prioritized, feasible interventions rather than unranked wish-lists. The protocol emphasizes co-design with hosts and providers, a model you can replicate to produce prioritized, feasible interventions.
Expect questions on trauma- and gender-responsiveness; integrate safeguards and consenting processes into your facilitation plans early.
Policy analysts & funders
Policy analysts and funders can use links between satisfaction and re-hosting or advocacy as evidence to invest in standardized host supports. If the study finds links between satisfaction and re-hosting or advocacy, that creates evidence to invest in standardized host supports.
Plan for rapid knowledge mobilization (executive summaries, short films) as the authors intend.
Do more, faster with Evidano (map to the protocol)
Phase 1: Survey open-ends → Auto-thematic analysis
Evidano ingests LimeSurvey exports and extracts themes, frequencies, and co-occurrence networks to speed open-end coding. The problem is that open-ended responses are high-volume and slow to code.
How Evidano helps: ingest LimeSurvey exports, auto-clean text, run thematic extraction and frequency counts, and produce co-occurrence networks to see which concerns cluster with 'satisfaction' or 'boundaries'.
Outputs: exportable codebook, most-cited quotes, and segment comparisons (for example, provinces, prior hosting vs current).
Phase 2: Interview transcripts → Fast coding & validation
Evidano auto-transcribes audio, redacts PII, and applies imported codebooks to produce thematic hierarchies and memos to accelerate interview analysis. The problem is that 20–25 interviews need consistent coding and member-checks.
How Evidano helps: auto-transcribe audio (custom dictionary for names/terms), redact PII, apply imported codebook, and generate thematic hierarchies with subcodes. Use AI chat over transcripts to produce summary memos and pull verbatim quotes for the toolkit.
Security note: Evidano encrypts data and does not use your data to train third-party models.
Phase 3: Co-design outputs → Prioritization & visualization
Evidano converts sorted and rated statements into cluster maps and ranked action lists to speed co-design synthesis. The problem is translating brainstorms into prioritized actions is labor-intensive.
How Evidano helps: ingest sorted/rated statements from GroupWisdom or spreadsheets, produce cluster maps, and generate a ranked action list by importance and feasibility. Export visuals (cluster maps, word clouds) for stakeholder sessions.
Extra: Multilingual & distributed teams
Evidano supports translation with a custom dictionary so interviews in other languages can be translated and analyzed consistently. This aligns with the protocol’s equity emphasis and helps include non-English hosts in follow-ups.
Checklist: five steps to reproduce this workflow in two weeks (pilot)
This checklist gives five steps to pilot the protocol's qualitative parts in two weeks using Evidano.
Day 1: Collect 10 survey open-ends and 3 recorded 45–60 min interviews; gather consent forms.
Day 2: Upload recordings and survey CSV to Evidano; enable PII redaction and custom dictionary.
Day 3–4: Auto-transcribe and auto-translate where needed; review transcripts and correct OCR errors.
Day 5–7: Auto-generate preliminary themes, frequency tables, and quote bundles; import a draft codebook to refine AI-assisted coding.
Day 8–10: Run cross-segment comparisons (for example, province, prior host vs current) and export visualizations (word cloud, co-occurrence network).
Day 11–14: Prepare a 2-page stakeholder brief and a short slide deck with visuals; schedule a 60-minute member-check.
FAQ: qualitative analysis of refugee hosting
How many interviews for saturation?
The protocol targets 20–25 interviews given heterogeneous participants (hosts and service providers). Monitor information sufficiency and recruit further if new themes keep emerging.
Can AI bias thematic coding?
AI can surface patterns but should be paired with human validation. Pair AI-assisted coding with human-in-the-loop checks, import a draft codebook, run AI-assisted coding, and refine codes collaboratively (Evidano supports collaborative code editing).
How to integrate survey SEM results with themes?
Use the survey to identify strong associations, then map qualitative themes that explain mechanisms behind those associations. Use the survey to identify strong associations (for example, hospitableness → advocacy), then map qualitative themes that explain mechanisms behind those associations, and use cross-segment reports to make the mapping explicit.
Wrapping up: next steps and CTA
The PLOS protocol provides a reproducible blueprint for pairing quantitative measurement with deep qualitative and participatory co-design. For teams running the interview and co-design phases, AI-enabled workflows cut hours from transcription, coding, synthesis, and visualization while preserving human validation.
Ready to pilot? Upload transcripts, survey open-ends, or concept-mapping CSVs to Evidano and get a draft thematic report within 48 hours.
Try Evidano for free or book a walkthrough at Evidano; Evidano encrypts your data and does not use your content to train third-party models.