AI-enabled qualitative research: sand mining governance

Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents. According to the August 13, 2026 PLoS One paper, PLoS One, sand mining governance in Ghana shows widespread illegal activity and enforcement gaps revealed through 32 interviews, two FGDs, and documentary analysis. The primary keyword for this post is "AI-enabled qualitative research" and this article explains how AI workflows accelerate coding, extractable statistics, and stakeholder mapping for governance research, using the PLoS One study as a worked example.

Key Takeaways

According to the August 13, 2026 PLoS One paper, PLoS One, the legal framework for sand mining in Ga South and Gomoa East, Ghana, is poorly applied because of weak monitoring, overlapping mandates, and political interference.

The PLoS One study published on August 13, 2026, used 32 key informant interviews and two FGDs held between March 3 and May 20, 2021, with follow-up interviews from January 10 to January 21, 2022.
The PLoS One authors reported that over 80% of sand mining in Ghana is carried out illegally, and that the study areas produce an estimated 4.55 million m3 of sand annually (data cited in the paper).
The PLoS One interviews document practical enforcement gaps, including understaffed regulators, outdated sanctions (for example, a GHS 200 fine noted in 2021), and missing local bylaws.
AI-enabled qualitative workflows can make these findings reproducible: automated transcription, thematic coding, frequency counts, and cross-segment comparisons compress weeks of manual work into hours.

What happened: methods and core findings from the PLoS One study

The PLoS One study analyzed attitudes toward sand mining laws in two Ghanaian local government areas using qualitative interviews, FGDs, and document review.

According to the PLoS One paper, researchers interviewed 32 key informants and ran two FGDs of 7–12 participants, with primary fieldwork from March 3 to May 20, 2021 and additional follow-ups January 10–21, 2022.

The PLoS One authors reported specific governance failures: licensing and compensation rules were poorly enforced, complaint channels were ineffective, monitoring staff were insufficient, and coordination among the Minerals Commission, the EPA, and local authorities was weak.

The PLoS One paper quoted local actors directly to illustrate dynamics, for example a landowner observed, "most landowners believe that while the lands belong to them, regulatory agencies accrue gains from sand mining" (Landowner in Gomoa Buduatta, 2021), and an EPA official noted, "The sanction for sand mining offenses... is GHS 200 ($20)" (Key informant, 2021).

Findings snapshot

Date	Metric	Value / Citation	Implication
August 13, 2026	Publication	PLoS One	Peer-reviewed source for governance evidence
Mar 3–May 20, 2021; Jan 10–21, 2022	Fieldwork period	Primary interviews and FGDs conducted as reported in PLoS One	Temporal context for respondents' statements
2021 (reported in study)	Key informant interviews	32 interviews	Sample reached saturation for homogeneous actor groups
Cited in paper	Illegal share of sand mining	>80% illegal (PLoS One citation)	Legal framework not effectively enforced
Cited in paper	Annual sand volume from study areas	4.55 million m3	High extraction pressure near Accra

Implications for qualitative researchers studying governance

Qualitative researchers should adopt AI-enabled workflows to handle multi-source governance data like the PLoS One study.

According to the PLoS One paper, complex governance findings combine interviews, FGDs, and documentary review; AI tools let researchers unify these sources into a single coded corpus for thematic and frequency analysis.

AI-assisted coding accelerates reproducible theme extraction (for example, licensing gaps, compensation failures, and inter-agency fragmentation) and supports cross-segment queries such as comparing regulators versus landowners on the same code.

When a study reports hard numbers (32 interviews, >80% illegal, 4.55 million m3) AI-extracted tables and dashboards make those statistics computable and citable for policy briefs and stakeholder meetings.

How Evidano Helps

Problem: hours to transcribe and code interviews

Solution: Evidano provides automated transcription with custom dictionaries and PII redaction to convert raw audio into research-ready text.

Researchers working on cases like the PLoS One study can upload interview audio, use Evidano Speech-to-Text for fast, accurate transcripts, and preserve sensitive details via built-in redaction.

Problem: stitching interviews, FGDs, and documents into one analysis

Solution: Evidano ingests transcripts, documents, and spreadsheets and performs thematic, content, frequency, and cross-segment analyses so themes such as "licensing" or "compensation" are comparable across actor groups.

The PLoS One authors used NVivo for thematic analysis; researchers can run equivalent or deeper queries in Evidano and export reproducible codebooks and visualizations via Evidano Features.

Problem: verifying quotes, counts, and timelines for policy audiences

Solution: Evidano links coded excerpts to source metadata (speaker, date, location) so statements like the GHS 200 sanction quote and the "over 80% illegal" statistic remain traceable.

Traceability reduces editorial friction when preparing evidence briefs and helps meet E-E-A-T expectations for policy and academic audiences.

Problem: iterating with stakeholders and non-English sources

Solution: Evidano supports translation with custom dictionaries and AI chat over your documents so you can validate themes with local stakeholders and translate quotes precisely.

For multi-actor governance studies, these features speed stakeholder review cycles and help researchers implement the consultative governance recommendations that the PLoS One paper makes.

FAQ: AI-enabled qualitative research

How can AI speed thematic analysis of interviews like in the PLoS One study?

Answer: AI speeds thematic analysis by automating transcription, preliminary coding, and frequency counts while leaving interpretive synthesis to humans.

According to the PLoS One paper, the study combined interviews, FGDs, and document review; AI can unify those sources, produce code frequency tables, and surface co-occurrence networks so researchers spend less time on manual tagging and more time on interpretation.

Can AI replace human qualitative coders in governance research?

Answer: No, AI cannot replace human coders for interpretive judgments, but AI can augment human work and increase reproducibility.

The PLoS One study relied on human-led thematic analysis; best practice is to use AI to draft codes and identify patterns, then have subject-matter experts validate, refine, and interpret themes for policy relevance.

What data and metadata does AI need to reproduce the PLoS One analysis?

Answer: AI needs verbatim transcripts, interview metadata (actor role, date, location), and the documentary texts cited in the study.

The PLoS One authors reported 32 interviews and two FGDs with dates and locations; providing those metadata fields to an AI system enables cross-segment queries such as regulator versus landowner perspectives.

Is using AI platforms ethical for sensitive interview data?

Answer: Yes, when platforms use PII redaction, controlled access, and do not share data with third-party models.

Evidano provides PII redaction and encrypts data and does not use customer data to train third-party models, which aligns with the confidentiality expectations reported by the PLoS One authors who restricted raw data for ethical reasons.

Conclusion & Next Steps

The PLoS One study published on August 13, 2026 demonstrates that qualitative governance research produces rich, actionable findings but also creates large, multi-source datasets that are time-consuming to synthesize.

AI-enabled qualitative research workflows compress transcription and coding time, maintain traceability for quotes and counts (for example, the PLoS One study's 32 interviews and >80% illegal estimate), and support the stakeholder-mapping and policy-facing outputs that governance teams need.

If you want to pilot an AI-assisted analysis on interviews, FGDs, and policy documents like those in the PLoS One study, explore Evidano Features and secure transcription options via Evidano Speech-to-Text.

Ready to try it? Try Evidano for free.

Topics

AI-enabled qualitative research
qualitative analysis Ghana sand mining
AI thematic analysis
transcript analysis AI

Keep reading

Browse all articles