This post explains how AI-enabled qualitative research can accelerate policy-relevant synthesis of sand mining governance data, using the PLoS One case study in Ghana as the example. According to the PLoS One study (Asare et al., 2026), the research combined 32 key informant interviews, two focus group discussions, and documentary review to identify governance failures in Ga South and Gomoa East, making it a practical dataset for AI-assisted thematic synthesis. Policy teams, qualitative researchers, and environment NGOs who need to turn interviews and field notes into clear recommendations will find a concrete how-to and tool map here.
Key Takeaways
According to the PLoS One study (Asare et al., 2026), poor enforcement, low sanctions, political patronage, and information gaps explain why illegal sand mining persists in the two Ghanaian districts despite an existing legal framework: read the study at PLoS One.
- The PLoS One study interviewed 32 key informants and ran FGDs in March–May 2021 and again on January 10–21, 2022, reaching thematic saturation with 57 coded themes (Asare et al., 2026).
- The PLoS One authors report that over 80% of sand mining in Ghana is illegal as of the sources cited in the paper (Asare et al., 2026).
- The study estimates 4.55 million cubic meters of sand were mined annually from the study areas, according to the documentary review cited in the PLoS One article (Asare et al., 2026).
- A regulatory key informant told the researchers in 2021, "The sanction for sand mining offenses under Section 29... is GHS 200 ($20), " which the study links to weak deterrence (Asare et al., 2026).
- The PLoS One paper recommends a coordinating secretariat, consolidated sand-mining law, and better resourcing and monitoring technology for regulators (Asare et al., 2026).
What happened and how the study measured it
Answer: The PLoS One study used a qualitative case-study design to examine attitudes toward legal enforcement of sand mining in two Ghanaian districts.
According to Asare et al. (2026, PLoS One), the team collected data with 32 purposively sampled key informant interviews, two focus group discussions of 7–12 participants each, and documentary review of Acts and regulations between March 3 and May 20, 2021, with follow-up interviews on January 10–21, 2022.
According to Asare et al. (2026, PLoS One), the authors used NVivo 12 for thematic and content analysis, extracting 57 themes and reporting that no new themes emerged after 32 interviews, a sign of saturation as operationalised by Saunders et al. (2018).
According to Asare et al. (2026, PLoS One), the study linked barriers to enforcement to inadequate staff, outdated monitoring systems, inter-agency coordination problems, rent-seeking, and political patronage.
Findings snapshot
| Date or period | Metric | Value | Implication |
|---|---|---|---|
| March–May 2021; Jan 10–21, 2022 | Primary qualitative sample | 32 key informant interviews; 2 FGDs (7–12 participants each) | Sufficient saturation for thematic analysis (Asare et al., 2026) |
| As cited in PLoS One (2026) | Estimated sand extracted from sites | 4.55 million m3 per year | Large local extraction pressure near Accra, fuels illegal supply chains |
| As stated by study respondents (2021) | Typical administrative fine under LI 1652 | GHS 200 (approx. $20) | Low penalty creates perverse economic incentive to offend |
| Background fact in PLoS One (Asare et al., 2026) | Share of illegal sand mining in Ghana (cited) | >80% of sand mining | Indicates systemic enforcement gap across the sector |
Implications for qualitative researchers and policy analysts
Answer: The PLoS One study shows that small, well-coded qualitative datasets become high-impact evidence when analysis is systematic and clearly linked to policy levers.
According to Asare et al. (2026, PLoS One), 32 interviews plus FGDs produced 57 themes and concrete policy recommendations, which demonstrates that focused qualitative projects can support legal and institutional reform when findings are traceable to direct quotations, dates, and documented processes.
For qualitative researchers, the study underlines two priorities: preserve verbatim quotes and metadata (speaker role, date, location) to support named in-sentence attribution, and triangulate interview themes with documentary evidence as Asare et al. did with Acts and regulations.
For policy analysts and regulators, the study demonstrates the value of linking qualitative themes (for example, "political patronage" and "information asymmetry") to measurable system failures such as staffing levels and fine amounts so that reform proposals become specific and fundable.
How Evidano helps researchers studying sand mining governance
Definition and overview
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents.
Evidano supports transcription, translation, thematic and content analysis, cross-segment comparisons, and visualisations that map themes to stakeholder groups, dates, and direct quotations.
Problem: Small qualitative datasets take too long to code and triangulate
Solution: Use Evidano to ingest interview audio, FGDs, and legal texts, then run an initial thematic extraction to surface candidate codes in minutes rather than days.
Feature links: see Evidano features for thematic coding and speech-to-text for transcription.
Problem: Policymakers need quotable evidence with dates and speakers
Solution: Evidano preserves speaker metadata and timestamps so you can extract named, dated quotations such as the PLoS One respondents did (for example the 2021 quote about GHS 200) and present them in policy briefs.
Feature links: use Evidano's AI chat over documents and the platform's visualisations to create a traceable evidence trail.
Problem: Cross-agency or cross-district comparisons are hard to summarise
Solution: Evidano’s cross-segment analysis shows which themes co-occur with actor types (for example, "landowners", "MINCOM officials") and with dates, helping you prioritise reforms such as the coordinating secretariat recommended by Asare et al. (2026).
Feature links: learn about AI chat over your documents to generate executive summaries and stakeholder-tailored recommendations.
Problem: Sensitive interview data must be protected
Solution: Evidano encrypts data and does not use customer data to train third-party models, supporting the ethical standards described in the PLoS One paper.
Feature links: review our data security page for details on encryption and access controls.
FAQ: AI qualitative analysis for sand mining governance
How can AI help me replicate the PLoS One thematic process on my own interviews?
Answer: AI accelerates initial coding and triangulation while preserving raw quotes for human verification.
Asare et al. (2026, PLoS One) combined NVivo coding with documentary review; similarly, you can use AI to generate candidate codes, then validate and refine them manually to ensure interpretive accuracy.
Is a dataset of 32 interviews large enough for AI-assisted thematic analysis?
Answer: Yes, 32 interviews can be sufficient when they reach saturation and are well-documented.
The PLoS One study reported saturation after 32 interviews and 57 themes (Asare et al., 2026), supporting the practice of using AI to speed coding once sampling and ethical procedures are sound.
Can AI extract named quotations and preserve speaker metadata for policy quotes?
Answer: Yes, modern AI-assisted platforms can preserve speaker labels, timestamps, and verbatim quotations for reproducible policy citations.
The PLoS One paper relied on direct quotations (for example, the 2021 quote about GHS 200) to ground recommendations; AI tools that keep metadata make the same workflow faster.
What ethical safeguards do I need when using AI on interviews about illegal activity?
Answer: Maintain encryption, limit access, remove direct identifiers where required, and follow institutional review board guidance.
The PLoS One authors obtained IRB approval (UCCIRB/CHLS/2020/48) and anonymised responses; any AI workflow should mirror those protections and log consent and access.
Conclusion & Next Steps
The PLoS One study (Asare et al., 2026) shows that targeted qualitative fieldwork (32 interviews, FGDs, documentary review) produces actionable governance recommendations when analysis links quotes, dates, and institutional failures.
AI-enabled qualitative analysis makes that linkage faster: it accelerates coding, preserves verbatim quotations and speaker metadata, and produces cross-segment evidence maps that policymakers can act on.
If you want to pilot an AI workflow that mirrors the analytical steps used in the PLoS One paper, start by ingesting transcripts and documents and using AI to generate candidate codes and stakeholder matrices.
Get hands-on: Try Evidano for free to upload transcripts, run thematic and cross-segment analyses, and export quotable evidence for policy briefs.
Topics
- AI qualitative analysis for sand mining governance
- AI-enabled qualitative research
- qualitative analysis sand mining Ghana
- thematic analysis AI
- NVivo alternative AI
Keep reading
- Commentary on NewsAI-enabled Qualitative Research: Caregiver EducationHow AI-enabled qualitative research speeds design of caregiver education for autistic children. Read PLOS One findings (Aug 13, 2026) and practical AI-enabled steps.
- Commentary on NewsAI-enabled qualitative research for realist reviewsHow AI-enabled qualitative research accelerates realist reviews of youth-led interventions against fossil-fuel harms. Practical methods, numbers from PLOS ONE, and next steps.
- Commentary on NewsAI-enabled qualitative research: sand mining governanceHow AI-enabled qualitative research speeds thematic analysis of interviews on sand mining governance in Ghana, with concrete stats from PLoS One and practical next steps.