Evidano is an AI-powered qualitative data analysis platform that ingests freelist text, survey CSVs, and transcripts to speed cleaning, salience and sentiment analysis, and exportable visuals. Low enrollment among Black, Hispanic/Latinx, women, and rural populations undermines trial generalizability. This post shows a repeatable workflow for qualitative analysis of research participation using the July 1, 2026 PLOS study (n=101, data collected May–Sep 2023) as an example. You will learn which freelisting signals matter (for example, ubiquitous salience of “scary”), how to prioritize outreach to people never asked to join research, and exactly how AI tools speed coding, cross-segment comparison, and visualization. Practical payoff: reduce synthesis time and produce stakeholder-ready outputs you can action in weeks, not months. See how to import freelist text, run thematic and cross-segment analyses, and export quotes and visuals using Evidano (Evidano). Note: this is research-focused guidance, not clinical advice.
Key Takeaways
This post describes a two-week, repeatable workflow to turn freelist data (n=101, May–Sep 2023) into actionable recruitment tactics, prioritizing people never previously asked and neutralizing fear and burden signals.
Use freelisting to detect culturally salient terms, then follow up with interviews, salience and sentiment analyses, and AI-enabled cleaning and visualization to iterate messaging quickly.
- PLOS freelisting (n=101) found mixed sentiments about research and universally salient fear-related terms including “scary” (published July 1, 2026).
- Prior research experience (32% previously participated) was associated with more positive sentiment; 58% of respondents reported never being asked to participate.
- Operational levers: ask more people (especially those never approached) and address fear/burden with plain-language onboarding and low-effort incentives.
- Evidano ingests text and spreadsheets, auto-groups synonyms, builds editable parent-term dictionaries, runs salience-style and cross-segment analyses, and exports stakeholder-ready deliverables.
Fast take: what the PLOS study found
The PLOS freelisting study found mixed sentiments about clinical research with salient fear-related terms across demographic groups.
In brief: the PLOS study used web-based freelisting (May–Sep 2023) to probe perceptions of clinical research among underrepresented groups; 101 participants completed the freelists (56% Black, 23% Hispanic/Latinx, 80% women, 46% rural). Key, shared salient terms included “study, ” “knowledge, ” “search, ” and worryingly, “scary.” Full paper: PLOS.
- Publication: July 1, 2026; data collection: May–Sep 2023; n=101
- Core finding: mixed sentiments overall; more negative sentiment among people never previously asked to participate
- Unique signal: “research misconduct” salient among Black respondents, a trust cue that recruitment must address
Findings snapshot
| Metric | Value | Source / Note |
|---|---|---|
| Sample size | 101 | PLOS study (May–Sep 2023) |
| Demographics | 56% Black; 23% Hispanic/Latinx; 80% women; 46% rural | Participants could belong to multiple groups |
| Prior research experience | 32% previously participated; 58% never asked | Prior experience → more positive sentiment |
| Shared salient terms | study, knowledge, search, scary | Across demographic groups |
| Publication | July 1, 2026 | PLOS |
What happened: methods & key results
The study ran a web-based freelisting survey at five health systems and a community organization across Philadelphia, Atlanta, and Washington, DC between May–Sep 2023.
The team ran a web-based freelisting survey at five health systems and a community organization across Philadelphia, Atlanta, and Washington, DC (May–Sep 2023). Participants listed terms in response to three prompts: “research, ” “being asked to participate, ” and “being a participant.” Responses were cleaned into parent terms and analysed with Smith’s salience index to identify the most culturally salient items.
- Salience combines frequency and order in lists; investigators used scree plots to select salient terms.
- Sentiment coding (positive/neutral/negative) showed overall neutral views of the word “research, ” positive sentiment when asked to participate, and mixed or negative sentiments about being a participant, especially for those never asked before.
- Contextual gap: freelisting identifies priorities but not deep reasons, authors recommend follow-up interviews or concept mapping.
Implications for researchers, UX teams, and trial ops
For recruitment teams
The study implies recruitment teams should prioritize outreach to people never previously asked because first invitations shift sentiment toward positive.
Prioritize outreach to people who’ve never been asked, the study shows that first invitations shift sentiment toward positive. Treat initial contact as a trust-building intervention with clear purpose, low burden, and tangible incentives.
Address universal fear signals (for example, ‘scary’ or ‘lab rats’) with plain-language FAQs, short explainer videos, and demonstration of participant protections.
For UX / qualitative researchers
The study implies UX and qualitative researchers should use freelisting as a rapid pulse and follow up with interviews for context.
Use freelisting as a rapid pulse to detect which concepts are top-of-mind across segments, then follow up with interviews for context.
Compare freelist salience across segments (for example, prior experience versus never asked) to prioritize messaging experiments.
For policy & community teams
The study implies policy and community teams must address structural trust deficits where signals like “research misconduct” appear.
‘Research misconduct’ salient among Black respondents signals structural trust deficits. Invest in long-term community partnerships and transparent data-sharing commitments.
Track whether system changes (transport stipends, flexible hours) measurably reduce salience of ‘effort’ and ‘burden’ in repeat freelists.
How Evidano helps: map from problem to AI-enabled solution
Problem: scattered text inputs (freelists, transcripts, surveys)
Evidano ingests varied text formats and normalizes terms with an editable dictionary to reproduce the PLOS cleaning step at scale.
Solution: ingest all formats (text, spreadsheets, transcripts) and normalize terms with a shared dictionary to reproduce the PLOS cleaning step at scale.
Problem: time-consuming cleaning & codebook mapping
Evidano auto-groups synonyms and builds an editable parent-term dictionary to reduce manual coding time.
Solution: Evidano auto-groups synonyms, builds a parent-term dictionary you can edit, and exports a validated codebook, reducing manual coding hours.
Problem: comparing segments (e.g., prior participant vs never asked)
Evidano runs frequency, salience-style, and cross-segment analyses in one workflow and visualizes differences.
Solution: run frequency, salience-style, and cross-segment analyses in one workflow; visualize differences with co-occurrence networks and hierarchical code → subcode trees.
Problem: stakeholder-ready deliverables
Evidano exports clickable quotes, shareable visuals, and an executive brief, and supports AI chat to draft outreach language.
Solution: export clickable quotes, shareable visuals, and an executive brief. Use AI chat over your corpus to draft recruitment scripts, consent language, and community-facing FAQs.
Security & governance
Evidano uses encrypted storage and proprietary LLMs tuned for qualitative research, and customer data is not used to train third-party models.
Evidano uses encrypted storage and proprietary LLMs tuned for qualitative research; customer data is not used to train third-party models (see Evidano for details).
Two-week runbook: from freelists to action
This two-week, 10-step runbook reproduces the study’s insights and converts them into recruitment tactics.
Follow this 10-step, two-week plan to reproduce the study’s insights and convert them into recruitment tactics.
- Day 1: Import freelists and survey CSVs into Evidano; attach metadata (age, race, prior participation).
- Day 2–3: Auto-clean and review parent-term dictionary; confirm merges for local vocabulary.
- Day 4: Run frequency and salience-style ordering; generate scree plots and pick salient elbow terms.
- Day 5–6: Tag sentiment and validate on a 10% sample; correct mismatches.
- Day 7: Compare segments (for example, never asked vs. previously participated) and export top divergent terms.
- Day 8–10: Build visuals (co-occurrence network, hierarchical codes) and extract representative quotes.
- Day 11: Draft targeted outreach scripts and FAQ language with Evidano’s AI chat.
- Day 12–14: Pilot two messaging variants with community partners; collect quick feedback and rerun freelisting to measure signal change.
Quick QA / ethics note
The PLOS study was IRB-approved, and teams must secure consent and local approvals before recruitment or data collection.
Ethics: the PLOS study was IRB-approved; recommendations here are research-focused and non-diagnostic. Always secure consent and local approvals before recruitment or data collection.
- Freelisting is fast but shallow, use interviews for context before policy changes.
- Be mindful of priming, the authors noted their example prompt may have amplified fear-related terms.
FAQ: Qualitative analysis of research participation
What did the PLOS freelisting study find about perceptions of clinical research?
The PLOS freelisting study found mixed sentiments overall with salient terms such as “study, ” “knowledge, ” “search, ” and “scary.”
The study reported that these terms were shared across demographic groups and that 'research misconduct' emerged as especially salient among Black respondents, indicating trust concerns. The study sample was n=101, data collected May–Sep 2023, and the paper was published July 1, 2026 (PLOS).
How was the freelisting study conducted?
The freelisting study was conducted as a web-based survey across five health systems and a community organization in three cities between May and September 2023.
Participants listed terms in response to three prompts about research and participation, responses were cleaned into parent terms, and Smith’s salience index with scree plots identified culturally salient items.
Who were the participants and what were the key demographics?
The study included 101 participants with 56% identifying as Black, 23% Hispanic/Latinx, 80% women, and 46% rural representation.
The sample allowed multiple group memberships; 32% previously participated in research and 58% reported never being asked to participate, which correlated with more negative sentiment among those never asked.
How can teams reproduce these analyses quickly?
Teams can reproduce the analysis in about two weeks by importing freelist text, auto-cleaning, running salience and sentiment analyses, and piloting messaging variants.
Follow the runbook steps: import data, confirm parent-term merges, run salience and scree plots, validate sentiment on a sample, compare segments, build visuals, extract quotes, draft outreach language, and pilot with partners.
What ethical cautions should teams heed when using freelisting for recruitment insights?
Teams must obtain IRB approval and informed consent and remember that freelisting is fast but does not replace interviews for deep context.
The PLOS authors cautioned about priming effects in prompts and recommended follow-up interviews or concept mapping before making policy changes.
Conclusion: turn freelists into equitable recruitment
Turn freelists into equitable recruitment by asking more people and neutralizing fear and burden signals with clear, low-effort onboarding and trust-building steps.
The PLOS freelisting study (published July 1, 2026) highlights two operational levers: ask more people (especially those never approached) and neutralize fear and burden signals with clear, low-effort onboarding. AI-enabled qualitative workflows compress cleaning, cross-segment comparison, and visualization into days, letting teams test messaging and access changes iteratively.
- Next step: replicate the study pattern on your cohort, import freelists or exit-survey text, run salience and sentiment, then pilot targeted outreach.
- Ready to try it? Start a pilot and accelerate synthesis with Evidano: Try Evidano for free.