Tarrant County’s August 20, 2025 vote to cut polling places from 331 to 216 (a reduction of more than 100 sites) is a live example of policy change that produces rich qualitative data, public testimony, commissioners’ remarks, and community response. This post shows researchers and policy analysts how to run a reproducible qualitative analysis of polling site cuts using public records and testimony, and how Evidano (www.evidano.com) speeds transcription, coding, cross‑segment comparison, and visual reporting so you can move from raw comments to stakeholder-ready findings in days, not weeks.
Fast take + source
What happened: On August 20, 2025 Tarrant County commissioners voted 3–2 to reduce Election Day polling sites to 216 (down from 331 in 2023) and cut early voting locations; officials cited cost savings and low turnout. Read the original reporting at www.propublica.org/article/tarrant-county-texas-polling-sites-early-voting-cuts.
- Why analysts care: site reductions shift travel/time costs and often depress turnout among Black, Hispanic, and young voters, a pattern cited repeatedly by speakers at the county meeting.
- What you can learn: who is impacted (by neighborhood, age, language), what reasons officials give, and how narratives map to policy levers (hours, site proximity, transportation).
Snapshot: numbers & facts
| Date | Metric | Value | Source | Implication |
|---|---|---|---|---|
| Aug 20, 2025 | Commissioners vote | 3–2 to reduce polling sites | www.propublica.org/article/tarrant-county-texas-polling-sites-early-voting-cuts | Decision enacted at county level |
| 2023 vs 2025 | Polling sites | 331 → 216 (↓115) | ProPublica reporting | Significant local access reduction |
| 2023 | Turnout (countywide nonpresidential) | ≈12.5% of registered voters | County election admin (reported) | Used to justify cuts; low baseline turnout |
| 2025 (law) | State minimum Election Day sites | Reduced to 212 (from 347 in 2023 req.) | Texas legislative change | Gives counties legal floor for cuts |
| Budget note | Estimated savings | ≈$1, 000, 000 | County official statement | Fiscal rationale cited by commissioners |
What happened (plain English)
Tarrant County Judge Tim O’Hare and two Republican commissioners argued that countywide voting (since 2019) plus historically low midterm turnout made many sites redundant. The two Democratic commissioners opposed the move, warning it would disproportionately affect Black, Hispanic, and college‑age voters. Officials said the cuts were based on 2023 turnout and site accessibility; some popular early voting sites were later restored after pushback.
- Political context: O’Hare launched an electoral integrity unit in 2022 and supported redistricting earlier in 2025; the vote follows national rhetoric about mail‑in ballots and ongoing Texas policy changes.
- Community response: Dozens of residents spoke against the cuts; organizers argue reductions are effectively voter suppression regardless of stated budget motives.
How to run a qualitative analysis of polling site cuts
Goal: turn meeting transcripts, public comments, press releases, and social posts into evidence that shows who’s affected, where access changes, and which arguments drive policy.
- Assemble sources: meeting minutes/transcripts, speakers’ list, local media (e.g., ProPublica), council statements, and demographic maps.
- Preprocess: standardize timestamps, redact PII, and apply a custom dictionary for local place names and officials (O’Hare, Ramirez, Ludwig).
- Code: mix inductive themes (access, cost, fraud claims) with deductive codes (race, age, transportation, turnout).
- Cross‑segment: compare themes by precinct, race/ethnicity, and speaker type (resident vs. official) to surface differential impact.
- Visualize: site‑level heatmaps, co‑occurrence networks of themes (e.g., 'cost' + 'efficiency' vs 'access' + 'community'), and quote buckets for stakeholders.
Implications for researchers and policy teams
For UX / Civic researchers
Map friction: use coded testimony to estimate increased travel/time costs for identified cohorts and prioritize mitigation (mobile sites, extended hours).
Design studies: run quick intercept surveys in neighborhoods where sites were removed to validate qualitative claims.
For policy analysts & advocates
Evidence for hearings: build an indexed dossier of quotes tied to precincts and turnout impact to support requests for additional sites or legal challenges.
Monitor narratives: track the recurrence of 'fraud' vs 'efficiency' frames across statements and media to anticipate policy shifts.
For data teams
Triangulate: combine qualitative themes with voter files and spatial data to quantify likely turnout changes and model who loses access.
Do more, faster with Evidano
Ingest & clean
Import commissioner transcripts, public hearing audio, PDF reports, and social posts into Evidano. Use automated transcription with a custom dictionary for local names and PII redaction to prepare a research‑ready corpus in hours.
Code at scale
Apply an initial codebook (access, cost, fraud, transportation, demographic impact) and let Evidano propose subcodes from recurrent language. Review, merge, and freeze codes, reproducibly.
Cross‑segment analysis & visuals
Run thematic frequency by precinct or speaker type, generate co‑occurrence networks, and export hierarchical theme → subcode visualizations for briefings.
Narrative & stakeholder outputs
Produce an evidence pack: top quotes by precinct, timeline of decisions, and an executive summary. Share clickable reports with policymakers and legal teams.
Security note: Evidano encrypts data and does not use your inputs to train third‑party models.
Two‑week workflow: checklist to reproduce this analysis
Week 1: ingest, transcribe, and clean
- Day 1–2: collect transcripts (commissioners, public speakers), media stories (e.g., ProPublica), and precinct maps.
- Day 3: run Evidano transcription and apply custom dictionary; redact PII.
- Day 4–7: develop initial codebook and run automated coding; review and adjust.
Week 2: analyze & report
- Day 8–10: run cross‑segment frequency and co‑occurrence analyses; generate maps and quote sets.
- Day 11–12: draft executive brief and stakeholder slide deck.
- Day 13–14: finalize deliverables and circulate an interactive report for review.
FAQ: common questions about qualitative analysis of voting access changes
Q: How do I compare impact across precincts?
A: Link coded testimony to precinct boundaries and run frequency + normalized rates (themes per 1, 000 residents) so you control for population differences.
Q: Can qualitative findings influence legal or policy action?
A: Yes, well‑documented testimony tied to spatial data and turnout models strengthens filings, public comments, and advocacy memos.
Q: Is this approach reproducible?
A: With a frozen codebook, clear data provenance, and exported analysis artifacts (code logs, visuals), the workflow is fully reproducible and audit‑ready.
Conclusion, next steps
If you’re tracking access changes like Tarrant County’s August 20, 2025 cuts, a focused qualitative pipeline converts testimony and transcripts into targeted, actionable evidence for advocates, analysts, and officials.
- Start small: run a precinct‑level pilot (one hearing + 2 weeks of work) to validate themes and estimate impact.
- Ready to try it? Sign up or request a demo at www.evidano.com to import transcripts, run thematic and cross‑segment analyses, and produce stakeholder-ready reports in days.
Keep reading
- Commentary on NewsTwo Definitions: Climate Change Acceptance for UndergradsHow a PLoS One Delphi study (Aug 25, 2026) defined climate change acceptance for undergraduate science students, and how AI-enabled qualitative analysis applies it.
- Commentary on NewsResearcher-in-the-loop: AI-enabled UX researchHow the researcher-in-the-loop model governs AI-enabled UX research. Learn practical governance, stats from the August 2026 piece, and how Evidano supports this workflow.
- Commentary on NewsResearcher-in-the-Loop: Governance for AI UX ResearchGovern AI in qualitative UX research with the researcher-in-the-loop model from Jennifer L. Bowie (Aug 25, 2026): practical rules, risks, and tool mappings.