Evidano is an AI-powered qualitative data analysis platform that ingests transcripts, runs thematic and cross-segment analyses, and generates visual reports. This PLOS One study, published 8 June 2026, compares specialist plastic waste enterprises with general/mixed waste collectors in Greater Accra (Ghana) and Kisumu (Kenya) using 11 focus group discussions (n=87). The qualitative component identifies price volatility, theft of separated plastics, seasonal sachet-water flows, gendered labour patterns, and rising faecal contamination risks. Read the original paper at PLOS One. For platform support, see Evidano.
Findings snapshot
| Item | Value | Source / Note |
|---|---|---|
| Publication date | 8 June 2026 | PLOS One |
| Cities studied | Greater Accra (Ghana), Kisumu (Kenya) | Fieldwork: Accra 26 Sep–5 Oct 2022; Kisumu 2–3 Oct 2023 |
| Qualitative sample | 11 FGDs, n=87 participants | Plastic main/sub-collectors, general collectors, intermediaries, apex traders |
| Key risks reported | Price volatility; theft; harassment; occupational injuries; faecal contamination | Seasonal sachet water demand and imported plastics increase volatility |
| Demographic pattern | Accra: older women among plastic sub-collectors; Kisumu: younger, more educated workers | Impacts safety, formalisation prospects |
What happened: methods & main themes
This section describes the study methods and the main thematic findings from the focus group discussions. The study used focus group discussions to compare specialist plastic waste enterprises with general or mixed waste collectors across two cities, with Accra hosting six FGDs and Kisumu hosting four FGDs plus a small apex-trader meeting; audio was transcribed, translated where needed, and thematically coded using NVivo.
- 11 FGDs, n=87 participants; Accra fieldwork: 26 Sep–5 Oct 2022; Kisumu: 2–3 Oct 2023.
- Shared challenges reported included societal stigma, harassment and extortion, hazardous items and rising diaper/faecal contamination, lack of equipment, and transport or fuel costs.
- Plastic-specific challenges in Accra included price volatility driven by seasonal sachet-water use and bulk imports, and theft of stored separated plastics; many sub-collectors were elderly women facing musculo-skeletal risks.
- Mixed-waste-specific challenges included illegal undercutting by unregistered collectors and higher disposal fees, and slower separation where households do not separate waste.
- Participant requests included PPE, secure storage, transport support, representative associations, price-regulation or local recycling agents, and occupational health programmes.
So what for researchers, UX teams and policy analysts
Researchers
This subsection explains what researchers should change in study design and analysis. Researchers should plan for longitudinal or repeated cross-sections to capture seasonal and import-driven volatility, and disaggregate by role (sub-collector vs main collector) and by city to capture demographic and institutional differences.
UX / Product teams (waste tech or marketplaces)
This subsection explains product design implications derived from the findings. UX and product teams should design user journeys mindful of older female users in Accra and younger, tech-savvy intermediaries in Kisumu, and include payment guarantees and proof-of-delivery to reduce payment delays and defaulting noted in Kisumu.
Policy & program teams
This subsection explains policy and program priorities informed by the study. Policy and program teams should combine fair-pricing mechanisms, secure storage, and occupational health supports, and support association-building (for example, PWCA and KIWAN) and local recycling agents to reduce harassment and stabilise prices.
Do more, faster with Evidano
This section explains how Evidano addresses common problems in qualitative analysis
Evidano ingests transcripts, audio, and survey sheets to accelerate thematic and cross-segment analyses and reporting.
Problem: scattered FGDs & seasonal effects → Solution
Import FGD transcripts, audio, and survey sheets into Evidano to run thematic, frequency, and cross-segment analyses (Accra vs Kisumu; sub-collector vs main collector).
Problem: inconsistent coding across teams → Solution
Use Evidano's codebook import and AI-assisted coding to apply a reproducible thematic framework and generate hierarchical themes and subthemes (for example, 'price volatility' → 'seasonal sachet demand', 'imported bulk shipments').
Problem: multilingual audio & noisy recordings → Solution
Apply Evidano transcription and translation with custom dictionaries for local terms (for example, 'sachet rubbers') and PII redaction before coding.
Problem: convincing stakeholders with evidence → Solution
Create co-occurrence networks, word-clouds, and segment comparisons to show which themes co-occur with outcomes (for example, 'theft' and 'insecure storage') and export stakeholder-ready reports.
Security & compliance
Evidano encrypts your data end-to-end and does not use customer data to train third-party models, suitable for sensitive field transcripts and donor-funded projects.
Run-book: reproduce this analysis in two weeks
This run-book gives seven practical steps to turn transcripts into policy-ready insight using the same inputs and themes as the PLOS One study.
- 1) Gather inputs: audio files, transcripts, registration lists, and the PLOS paper for reference: PLOS One.
- 2) Auto-transcribe and translate in Evidano; add custom dictionary entries for local terms such as 'aboboya' and 'sachet rubbers'.
- 3) Import or create a codebook mirroring the PLOS study themes; run AI-assisted coding across transcripts.
- 4) Run cross-segment queries (city, role, gender, season) to quantify theme frequency and extract representative quotes per segment.
- 5) Visualize findings: co-occurrence network (for example, 'price volatility' ↔ 'seasonal demand'), hierarchical themes, and word clouds.
- 6) Validate results with member-check extracts and export interactive reports for stakeholders.
- 7) Schedule recurring imports to monitor seasonal volatility and evaluate interventions longitudinally.
FAQ: qualitative analysis of plastic waste enterprises
How do I compare Accra vs Kisumu reliably?
Use cross-segment frequency and co-occurrence analysis to compare Accra and Kisumu reliably: Evidano shows normalized theme rates and sample quotes per segment to control for sample size and code prevalence.
How do I handle local language terms?
Add custom dictionary entries during transcription and translation so the AI preserves meaning for local terms, for example mapping 'rubbers' to sachet packaging.
Is this approach ethical for sensitive topics (for example, contamination or child pickers)?
Use anonymisation and PII redaction to protect participants, and treat research outputs as program design evidence; if clinical or child protection issues arise, route to appropriate services because this research is not a diagnostic tool.
Wrapping up: what to do next
This section summarises recommended next steps to turn FGDs like the PLOS One study into policy-ready insights. If teams need to extract policy-ready insights from FGDs (11 FGDs, n=87), run reproducible thematic and cross-segment analysis, visualise co-occurrence, and set up longitudinal monitoring for price and flow volatility.
- Start a pilot: import two to three transcripts, auto-transcribe, auto-code, and generate a cross-segment summary to test assumptions.
- Secure the data and speed decision cycles, and Try Evidano for free to link your next qualitative project to evidence that policymakers and funders can act on.