Fast take: Cantrell et al. (Published July 1, 2026) analyzed 154 substantive open-ended responses from 322 authors at four U.S. R1 institutions and identified an APC Trap composed of eight overlapping codes that describe cognitive dissonance about article processing charges (APCs). This post shows researchers, librarians, and policy teams how to reproduce that qualitative analysis at scale and turn it into stakeholder-ready visuals. Full study: PLOS ONE.
Key Takeaways
This guide shows how to reproduce Cantrell et al.'s qualitative analysis of APCs and convert open-ended responses into stakeholder-ready insights.
Cantrell et al. analyzed 154 substantive open-ended responses (from 322 surveys) collected Oct 31–Nov 21, 2023 and published Jul 1, 2026, identifying eight APC Trap codes.
- Cantrell et al. collected 322 survey responses, of which 154 were substantive open-ended responses (47.8%), during Oct 31–Nov 21, 2023 under IRB exempt Protocol 23–0216.
- The research team inductively created eight APC Trap codes and used independent triple/quad coding with consensus rules, applying a code when at least four coders agreed.
- Key quantitative patterns: 62% of responses had 1–2 codes, 28% had three or more codes, Financial Concern (34%) and Diversity/Equity (32%) were the top themes; automated sentiment tools were unreliable for this corpus.
- To reproduce the analysis, import Qualtrics exports, run thematic coding with human adjudication, and produce cross-segment frequency and co-occurrence reports and visuals for stakeholders.
Findings snapshot
| Item | Value | Why it matters | Source |
|---|---|---|---|
| Published | July 1, 2026 | Defines the reporting and peer-review window | PLOS ONE |
| Completed surveys | 322 total; 154 substantive open-ended responses (47.8%) | Qual corpus size for a targeted inductive codebook | Cantrell et al., 2026 |
| Core result | 8 APC Trap codes; common co-occurrences (e.g., Financial Concern + Restriction of Decisions) | Actionable themes for policy and library interventions | Study figures & tables (see article) |
| Data collection window | Oct 31–Nov 21, 2023 (Qualtrics) | Context for funding and agreements at time of responses | Methods, Cantrell et al. |
What happened (plain English)
Cantrell et al. used purposive sampling across four U.S. R1 public universities to collect APC-related responses.
Two open-ended prompts produced 154 substantive responses which were inductively coded into eight APC Trap codes: Financial Concern; Diversity/Equity; Threshold of Reasonableness; Restriction of Decisions; Grants Don't Cut It; Academic Obligation/Expectation; Academic Labor; Quality/Rigor.
- Survey period: Oct 31–Nov 21, 2023; multi-site IRB exempt (Protocol 23–0216).
- Coding: eight emergent codes, independent triple/quad coding with consensus rules (code applied when at least four coders agreed), iterative codebook refinement.
- Key quantitative patterns: 62% of responses had 1–2 codes; 28% had three or more codes; Financial Concern (34%) and Diversity/Equity (32%) were top themes.
- Automated sentiment tools (NVivo Auto Sentiment, SentiStrength) were tested but found unreliable for this nuanced corpus; rely on thematic coding with human adjudication.
So what for research teams: implications of a qualitative analysis of APCs
For librarians & OA program managers
The APC Trap codes indicate direct support opportunities for libraries and OA programs.
The eight APC Trap codes map to targeted interventions: create emergency APC funds to address Financial Concern, provide discipline-specific guidance for HSS where traps were stronger, and communicate waiver and transformative agreement policies clearly to reduce Restriction of Decisions.
For departmental PIs and grant managers
APCs can shift project budgets and require explicit accounting.
Quantify how APCs shift budgets because 'Grants Don't Cut It' and 'Threshold of Reasonableness' were common; use cross-segment frequency analysis to show how many projects face budgetary diversion and for how many years.
For research policy & equity teams
APC-related comments indicate structural equity concerns that policy teams must address.
Diversity/Equity codes point to structural exclusion; use co-occurrence networks to link demographic segments to APC codes and prioritize systemic interventions rather than ad-hoc waivers.
Do more, faster: mapped to this use case
Evidano definition
Evidano is an AI-powered qualitative data analysis platform that ingests Qualtrics exports, supports AI-assisted coding, and produces cross-segment analytics and visuals.
Ingest the corpus
Evidano ingests survey exports and preserves respondent metadata so teams can reproduce crosstabs and segment comparisons.
Import survey spreadsheets (Qualtrics.csv) and supporting documents in one step; Evidano preserves respondent IDs and metadata so you can recreate code crosstabs by institution, discipline, or OA type.
Create an inductive codebook and scale coding
Evidano supports iterative codebook development and scalable AI-assisted coding with human review.
Start with the eight APC Trap codes from Cantrell et al. or import your own codebook. Use the AI-assisted coding in Evidano to apply codes to long-form responses, then review disagreements with side-by-side context before finalizing consensus.
Run frequency & cross-segment analysis
Evidano produces frequency tables and co-occurrence matrices for cross-segment comparison.
Generate frequency tables, co-occurrence matrices, and compare codes across segments (institution, discipline, Gold vs Hybrid). These are the exact outputs needed to reproduce the study’s finding that HSS authors and Hybrid articles show stronger APC Trap signals.
Visualize and communicate
Evidano creates visuals and quote exports for stakeholder communication.
Create word clouds, co-occurrence networks, and hierarchical code to subcode diagrams for stakeholder briefs. Export clickable quotes tied to codes for rapid inclusion in decision memos.
Security & compliance
Evidano supports encryption, secure sharing, and PII redaction workflows for sensitive qualitative data.
Data is encrypted and never used to train third-party models; Evidano supports PII redaction in transcripts and secure sharing, important when handling identifiable faculty comments.
Checklist: reproduce Cantrell et al. (APC Trap) in 7 steps
Follow this minimal run-book to replicate or extend the APC Trap analysis using your corpus:
- 1) Export Qualtrics responses with metadata (institution, discipline, OA type).
- 2) Import into Evidano and normalize respondent IDs and date fields.
- 3) Draft or import the eight APC Trap codes as a codebook (editable).
- 4) Run AI-assisted coding; review and adjudicate low-consensus items.
- 5) Produce cross-tab frequency tables (codes × institution/discipline) and co-occurrence matrices.
- 6) Generate visuals (co-occurrence network, hierarchical code tree) and assemble a 1–2 page stakeholder brief with representative quotes.
- 7) Share interactive report or export CSV/PNG for governance meetings; track changes over time.
FAQ: qualitative analysis of APCs
Can automated sentiment tools reliably characterize APC comments?
No, automated sentiment tools cannot reliably characterize APC comments in this corpus.
Cantrell et al. found NVivo Auto Sentiment and SentiStrength misclassify mixed or sarcastic responses; use thematic coding plus human adjudication for dependable results.
How many responses do I need to detect the APC Trap?
You can detect emergent APC Trap themes with a modest set of rich responses, but statistical comparisons need larger samples.
The study used 154 substantive responses (from 322 surveys); for emergent themes, 100–200 rich responses with purposive sampling often suffice, while statistical comparisons require larger stratified samples.
Where can I find lists of non-APC journals?
The Directory of Open Access Journals lists many Diamond and APC-free journals.
See DOAJ for guidance on APC-free venues.
Wrapping up: next moves
Cantrell et al. (Jul 1, 2026) show the APC Trap is experienced even at well-resourced R1 institutions, and teams should measure who is affected, where funds leak, and which interventions reduce inequity.
- Reproduce the APC Trap codebook on your corpus and produce cross-segment reports for finance, library, and equity committees.
- Book a demo or pilot at Evidano to import your Qualtrics data, run AI-assisted coding with a shared codebook, and produce publication-ready visuals in days: Try Evidano for free.