AI Qualitative Analysis of Spain's Transformative Education

AI qualitative analysis helps researchers scale discourse analysis of complex policy histories, extract timelines, and surface contested signifiers such as "sustainability". This post refracts the PLOS study of Spanish education policy through the lens of AI-enabled qualitative research and shows practical steps for researchers and policy teams to reproduce, extend, and operationalize those findings with AI tools. The primary audience is qualitative researchers, policy analysts, and education researchers who need to analyze laws, interview transcripts, and advocacy texts at scale while preserving interpretive nuance. The guidance below pairs concrete statistics and quotations from the PLOS article with actionable methods and a short explanation of how Evidano's platform supports transcription, thematic coding, cross-segment analysis, and visualizations for studies like this one.

Key Takeaways

According to PLOS Sustainability and Transformation, published August 18, 2026, Spain's education policy discourse shows three historical periods and a partial opening to transformative sustainability education (TSE).

Ó Riada et al. identify three discursive periods in Spain: 1975–1990 (democratic transition), 1990–2018 (neoliberal OECD-shaped consolidation), and 2019–present (discursive openings toward TSE), as reported in the PLOS article published on August 18, 2026.
The authors triangulated a document analysis of national education laws from 1975 to the present with seven in-depth interviews conducted between February and September 2025, according to the Methods section of the PLOS article.
The PLOS analysis links the 2020 LOMLOE law (drafted 2020, rolled out 2021–2024) to increased curricular references to sustainability, while warning that neoliberal governance risks co-optation, as argued in Ó Riada et al., 2026.
Two representative quotes from the source: the LOMLOE preamble states that “schools must become a place of stewardship and care for our environment” (LOMLOE, cited in Ó Riada et al., 2026), and an interviewee said, “the whole system has been squeezed by competitiveness, by market orientation” (interview 4, cited in Ó Riada et al., 2026).

What happened and how the PLOS study measured it

Answer: The PLOS study applied discourse analysis to Spain's education laws and seven interviews to trace how human-capital narratives displaced then partly yielded to transformative sustainability education (TSE), according to Ó Riada et al., 2026.

The PLOS authors analyzed the preambles and relevant articles of national education laws from 1975 onward and triangulated those documents with seven purposively sampled interviews carried out between February and September 2025, as described in the Methods section of the article.

The PLOS analysis frames the OECD and PISA as major external drivers of a modernist, human-capital paradigm from 1990 to 2018, and identifies the 2019 global climate mobilizations and coalition politics around 2020 (LOMLOE) as dislocatory events opening discursive space for TSE, per Ó Riada et al., 2026.

Findings Snapshot

Date	Metric	Value	Implication
August 18, 2026	Article publication	Ó Riada et al., PLOS Sustainability and Transformation	Peer-reviewed synthesis of 50 years of policy discourse
1975–1990	Discursive period	Democratic transition and pedagogical renewal	Relational, emancipatory education appears in laws (LOGSE, LODE)
1990–2018	Discursive period	Neoliberal consolidation influenced by OECD/PISA	Quantification and accountability become dominant signifiers
2019–present	Discursive period	Partial openings for TSE (LOMLOE draft 2020; rollout 2021–2024)	Sustainability gains curricular legitimacy but faces co-optation risk
Feb–Sep 2025	Interviews	7 in-depth interviews	Triangulation of document analysis with key actor perspectives

Implications for qualitative researchers analyzing education policy

Answer: Researchers should combine document-based discourse analysis with targeted interviews and reproducible coding to show both historical shifts and present contestations, as the PLOS study demonstrates.

The PLOS team used an iterative NVivo coding approach on legal preambles and parliamentary minutes and triangulated results with seven interviews between February and September 2025; qualitative researchers should mirror that triangulation to strengthen claims (Ó Riada et al., 2026).

Practically, researchers aiming to trace signifiers like “sustainability” should (a) extract and timestamp law preambles and curriculum texts, (b) code for hegemonic vs. counter-hegemonic signifiers, and (c) map co-occurrence of actors and signifiers to show articulation and chains of equivalence, following the discourse methods described by Ó Riada et al., 2026.

How Evidano Helps: from transcripts and laws to thematic insights

Evidano: platform definition

Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents.

Evidano supports transcription with custom dictionaries and PII redaction, document ingestion for laws and curricula, iterative coding, cross-segment analysis, and visualizations such as co-occurrence networks and hierarchical codes-to-subcodes that match the PLOS study needs.

Problem: Large document sets and dispersed quotes

Answer: A common problem is locating and linking the same signifier across dozens of legal texts and interviews, which the PLOS authors had to do across laws from 1975 onward (Ó Riada et al., 2026).

Evidano ingests PDFs, transcripts, and parliamentary minutes, performs full-text search and context extraction, and surfaces every occurrence of a term like “sustainability” with source, date, and surrounding clauses so researchers can trace its articulation over time.

Problem: iterative, reproducible coding at scale

Answer: The PLOS team used iterative NVivo coding; reproducibility and multi-pass coding are essential to validate discourse claims (Ó Riada et al., 2026).

Evidano automates baseline thematic suggestions, supports human-in-the-loop code refinement, logs coding decisions for auditability, and exports reproducible codebooks and frequency tables to reproduce the exact analytical steps.

Problem: cross-segment comparisons (laws vs interviews)

Answer: The PLOS study triangulated documents with seven interviews to understand implementation gaps, a task that benefits from cross-segment analysis (Ó Riada et al., 2026).

Evidano performs cross-segment analyses to compare how signifiers appear in law preambles versus interview transcripts, and generates visualizations showing where discursive openings (e.g., LOMLOE mentions of stewardship) do not align with teacher accounts of capacity and resources.

Relevant Evidano features and where to start

Answer: For projects like the PLOS study, start with Evidano's transcription and document ingestion, then apply thematic coding and co-occurrence mapping to trace signifiers over time.

See the platform capabilities in Evidano Features and consider transcribing interview audio with Evidano's speech-to-text pipeline to speed coding and ensure consistent timestamps.

FAQ: AI qualitative analysis

How can AI qualitative analysis reproduce the PLOS discourse study?

Answer: AI qualitative analysis can reproduce the PLOS study by ingesting the same legal texts and interview transcripts, applying the same coding schema, and exporting comparable frequency and co-occurrence outputs.

Ó Riada et al., 2026 describe an NVivo-based iterative coding workflow combined with seven interviews from February–September 2025; researchers should match those inputs and document coding decisions to ensure reproducibility.

Can AI preserve interpretive nuance when coding contested signifiers like "sustainability"?

Answer: Yes, when AI is used with human-in-the-loop validation to flag ambiguous contexts rather than auto-assign codes blindly.

The PLOS study shows “sustainability” functions as a floating signifier whose meaning depends on articulation in context (Ó Riada et al., 2026); AI models that surface contexts, offer suggestions, and let coders accept or correct them preserve interpretive nuance.

What dataset size and interview count are minimally sufficient for a discourse mapping like the PLOS study?

Answer: The PLOS study analyzed all national education laws since 1975 and triangulated that corpus with seven purposive interviews, showing that a focused document corpus plus targeted interviews can suffice for a robust discourse mapping (Ó Riada et al., 2026).

Researchers should prioritize document completeness for the institutional record and use purposive interviews until thematic saturation is reached, as Ó Riada et al. did with seven interviews conducted February–September 2025.

Are AI transcripts reliable for legal and parliamentary texts?

Answer: AI transcripts are reliable when combined with a domain-specific dictionary and human review to capture legal terms and named entities.

The PLOS methods used careful transcription and translation steps for interview excerpts; Evidano supports custom dictionaries and human review workflows to ensure high-fidelity transcripts for legal and parliamentary language.

Conclusion & Next Steps

Answer: AI-enabled qualitative analysis makes it feasible to replicate and extend the PLOS discourse study by scaling document ingestion, standardizing iterative coding, and measuring co-occurrence patterns across time, as demonstrated in Ó Riada et al., 2026.

The PLOS study (published August 18, 2026) shows both openings for transformative sustainability education in the 2020 LOMLOE and persistent neoliberal continuities that risk co-optation; AI workflows help make those continuities visible and auditable.

To reproduce the analysis, compile the legal corpus (1975–present), transcribe interviews (Feb–Sep 2025 style sampling), implement an iterative codebook, and run cross-segment visualizations to map contestations and chains of equivalence.

If you want to accelerate this workflow, explore how Evidano handles transcription, coding, and visualizations on the Evidano Features page and Try Evidano for free to onboard a pilot project.

Topics

AI qualitative analysis
AI-enabled qualitative research
qualitative analysis education policy
LOMLOE discourse analysis

Keep reading

Browse all articles