Researchers and practitioners want scalable methods to identify who leads knowledge-making in self-tracking studies. The primary keyword for this post is personal science qualitative analysis. The 18 August 2026 Frontiersin.org review by Sara Riggare offers a concrete, coded framework and numeric benchmarks that make that assessment possible, and this post shows how AI-enabled qualitative research tools can operationalize that framework for teams analyzing interview transcripts, n-of-1 logs, and device exports.
Key Takeaways
According to the Frontiersin.org review by Sara Riggare (published 18 August 2026), an agency-based framework separates self-tracking that is merely measurement from personal science that is self-directed, and the framework classified 721 search hits into 241 full papers and 46 self-tracking studies for detailed analysis (Frontiersin.org).
- 721 records were retrieved in April–May 2025 and screened, yielding 241 full papers for analysis, as reported in the Frontiersin.org review (published 18 August 2026).
- Among the 46 articles that used self-tracking as the data-collection method, the Frontiersin.org review (18 August 2026) classified 17 as low-agency, 10 as mixed-agency, and 19 as high-agency.
- The Frontiersin.org review found a publication peak of 38 articles in 2018 and noted an uptick to 13 articles in 2024, suggesting shifting terminology and renewed interest (Frontiersin.org, 18 August 2026).
- The Frontiersin.org authors conclude that the difference between measurement and knowledge production often turns on “who owns the question, ” and they argue that “personal science can be seen as a high-agency dimension of citizen science for health” (Sara Riggare, Frontiersin.org, 18 August 2026).
What Happened and how the review measured agency
Answer: The Frontiersin.org review applied a four-phase framework to assess epistemic agency in self-tracking: Questioning, Observing, Reasoning, and Discovering, then coded papers into three agency levels (low, mixed, high).
According to the Frontiersin.org paper (published 18 August 2026), the authors searched Web of Science and PubMed in April–May 2025 for the term “quantified self, ” retrieved 721 unique records, screened abstracts, and retained 241 full papers for analysis.
According to the Frontiersin.org article (18 August 2026), only 46 of those 241 full papers used self-tracking as the data-collection method and were therefore eligible for agency coding using the four-phase framework.
According to the Frontiersin.org review (18 August 2026), the authors assigned agency conservatively: when phases were ambiguous they assigned the lower level unless clear textual evidence showed the individual performed that phase.
The Frontiersin.org review also reported reviewer disagreement rates and resolution methods: during abstract screening 13% of cases (_n_ = 80) required discussion, and during the final categorization 17% of disputed cases (_n_ = 8) were resolved by group consultation (Frontiersin.org, 18 August 2026).
Findings snapshot
| Date / Period | Metric | Value | Implication |
|---|---|---|---|
| April–May 2025 | Records retrieved for “quantified self” | 721 | Large, heterogeneous literature that required filtering to identify empirical self-tracking studies (Frontiersin.org, 18 Aug 2026). |
| April–May 2025 | Full papers retained after screening | 241 | Corpus for conceptual analysis and agency classification (Frontiersin.org, 18 Aug 2026). |
| After screening (as reported 18 Aug 2026) | Papers using self-tracking as data-collection method | 46 | Subset eligible for the four-phase epistemic agency coding (Frontiersin.org, 18 Aug 2026). |
| Among the 46 self-tracking papers (reported 18 Aug 2026) | Agency breakdown (low / mixed / high) | 17 / 10 / 19 | Shows roughly equal representation of low- and high-agency practices, indicating diversity in who controls inquiry (Frontiersin.org, 18 Aug 2026). |
| Publication years (Fig. 3 in Frontiersin.org) | Peak year and recent counts | 38 articles in 2018; 14 in 2021; 13 in 2024 | Trend suggests an early peak with a 2024 uptick, possibly due to terminology shifts (Frontiersin.org, 18 Aug 2026). |
Implications for qualitative researchers and UX teams: personal science qualitative analysis
Answer: Researchers and design teams should code epistemic agency explicitly when analyzing self-tracking data because ownership of question-setting changes interpretation and inference.
According to the Frontiersin.org review (18 August 2026), low-agency studies treat participants as “sensor platforms, ” while high-agency studies show individuals conducting Questioning, Observing, Reasoning, and Discovering themselves.
According to the Frontiersin.org authors (18 August 2026), when self-trackers co-author papers it is a strong indicator that the Discovering phase was participant-led and that analysis represented personal science practices.
Practical step: in qualitative coding plans add a binary or ordinal “agency” code (e.g., low, mixed, high) linked to evidence tokens for Questioning, Reasoning, and Discovering; this is the core of personal science qualitative analysis because the unit of analysis expands from text to authorship and method-sharing behavior.
Ethics note: If your work touches on health, state that findings are research-focused and non-diagnostic and follow institutional guidance; the Frontiersin.org review (18 August 2026) highlights that personal science often operates outside formal ethical oversight.
How Evidano helps: problem to AI-enabled solution
Problem: locating agency across mixed document types
Answer: Mixed corpora (interviews, device logs, forum posts, publications) make it time-consuming to surface who authored questions or methods.
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents.
Evidano feature: multi-document ingestion and cross-document entity linking to tag authorship signals and method-sharing statements across transcripts and papers, enabling quick identification of Questioning, Reasoning, and Discovering phases.
Problem: synthesis of n-of-1 logs and qualitative reflections
Answer: Combining time-series device exports with narrative notes requires alignment and mixed-method codebooks.
Evidano feature: thematic, content, frequency, and cross-segment analyses that merge transcripts and spreadsheet-style logs so you can run N-of-1 pattern discovery and extract statements that show ownership of the research question.
See the product capabilities at the Evidano features page: Evidano features.
Problem: reproducible, auditable agency coding
Answer: Manual coding of agency is slow and non-reproducible without structured workflows.
Evidano feature: hierarchical codes and AI-assisted code suggestions that speed conservative assignment rules like those used in the Frontiersin.org review (assign lower level when ambiguous), and exportable audit trails for publication or team review.
Problem: transcription, PII, and multi-language corpora
Answer: Self-tracking projects produce audio logs and multilingual notes that require accurate transcription and privacy controls.
Evidano feature: built-in transcription with custom dictionaries and PII redaction plus translation support to standardize corpora before agency coding, reducing manual preprocessing time.
FAQ: personal science qualitative analysis
What is personal science qualitative analysis and how does it differ from standard qualitative coding?
Answer: Personal science qualitative analysis centers the person-as-researcher, coding not only for themes but for who authored the question, method, and interpretation.
Standard qualitative coding focuses on participants as respondents; personal science qualitative analysis treats the self-tracker as both subject and investigator and therefore adds agency codes for Questioning, Reasoning, and Discovering.
How can teams operationalize the Frontiersin.org four-phase framework with AI?
Answer: Operationalize the framework by creating code labels for Questioning, Observing, Reasoning, and Discovering and training or prompting AI to surface textual evidence for each label.
The Frontiersin.org review (18 August 2026) provides decision rules (assign lower level when ambiguous) that you can encode into AI-assisted workflows so model suggestions follow conservative human coding practices.
Does the Frontiersin.org review provide numeric benchmarks I can use to compare my corpus?
Answer: Yes, the Frontiersin.org review reports concrete counts useful as benchmarks: 721 retrieved records, 241 full papers, and 46 self-tracking studies with a 17/10/19 agency split (Frontiersin.org, 18 Aug 2026).
You can use these counts to calibrate expectations when your corpus is similar in scope or when you focus on extracting high-agency personal science cases.
Can AI help identify when a self-tracker is a co-author, and why does that matter?
Answer: Yes, AI can detect authorship statements and metadata that indicate co-authorship and flag those cases as strong evidence of participant-led Discovering.
The Frontiersin.org review (18 August 2026) notes that co-authorship often signals that the self-tracker performed the Discovering phase and that this is a reliable marker of high agency.
Conclusion & Next Steps
Answer: The Frontiersin.org review (Sara Riggare, 18 August 2026) shows that personal science and epistemic agency are measurable concepts and that roughly half of coded self-tracking studies exhibit high participant agency (19 of 46).
AI-enabled qualitative research can operationalize the four-phase framework (Questioning, Observing, Reasoning, Discovering) at scale by linking transcripts, device logs, and publications to surface who owns the question and who performed the analysis.
If you analyze self-tracking or n-of-1 data, use structured agency codes, conservative assignment rules from the Frontiersin.org review, and AI-assisted synthesis to speed discovery and improve reproducibility.
To try this approach on your corpus, see how Evidano ingests mixed documents and runs thematic plus cross-segment analyses at the Evidano features page, or start a free trial today: Try Evidano for free.
Topics
- personal science qualitative analysis
- epistemic agency self-tracking
- AI qualitative research
- personal science analytics
Keep reading
- Commentary on NewsImprove Developer Data Compliance: AI-enabled Qualitative ResearchTranslate PLoS One findings on developer data compliance into actionable research with AI-enabled qualitative analysis. Learn methods, stats, and next steps.
- Commentary on NewsAI for Qualitative Analysis of Hospital Pharmacy CrisisHow AI speeds qualitative analysis of hospital pharmacy crisis management studies, using LSE 2025 data and practical methods to scale synthesis for researchers.
- Commentary on NewsInterview vs Survey: Qualitative Analysis of HyperphagiaHow a 2026 mixed-methods study found underreported severe hyperphagia in Bardet-Biedl Syndrome and how AI qualitative analysis improves interviews, coding, and synthesis.