# Texas Cannabis Ledger data-integrity audit

Audit date: September 3, 2026

Status: Passed after corrections described below.

## Scope

This audit independently recomputed the 2025 dataset from the original Texas DPS county/offense workbook and the original U.S. Census Bureau county population workbook. It did not validate transformed Ledger data against itself.

The checks covered the annual statewide summary, all three ranking calculations, and a stratified sample of rendered county pages: the largest and smallest counties, counties with no matching row, counties with a reported zero, the highest eligible rates, and Leon, Bexar, Harris, and Travis counties.

## Results

- 254 Texas counties appear exactly once for 2025.
- 250 counties have at least one matching marijuana-coded DPS row.
- Four counties have no matching row: Jeff Davis, Loving, Motley, and Stonewall. They are marked missing, not zero.
- Seven counties have matching rows whose sum is zero: Briscoe, Glasscock, Hemphill, Irion, King, Roberts, and Sterling.
- Statewide marijuana-coded records: 41,701.
- Classified as possession: 41,049.
- Classified as other marijuana offenses: 652.
- Possession plus other equals the total.
- Every normalized county count, category, population value, and calculated rate matched the independent recomputation.
- All three rendered ranking orders matched an independent recomputation, with county name as the deterministic tie-breaker.
- Rate equals records divided by the Census population estimate, multiplied by 100,000.
- Comparative rate rankings include 163 counties with populations of at least 10,000. Smaller counties retain a calculated rate but receive no rank.

## Corrections made during the audit

The ingestion process previously converted all county names to title case. It now preserves official Census capitalization for DeWitt, McCulloch, McLennan, and McMullen counties. Ranking ties now use county name as an explicit secondary sort.

These corrections affected display names and deterministic ordering. They did not alter the underlying arrest-record counts.

## Independent published comparison

The Marijuana Policy Project reports 29,496 Texas cannabis arrests for 2025, including 28,753 possession arrests, using FBI-derived data. The Ledger's Texas DPS criminal-history recomputation is higher by 12,205 total records and 12,296 possession records.

The figures are not forced into agreement. Texas DPS criminal-history records and FBI/UCR-derived figures come from different reporting systems and can differ in definitions, coverage, agency participation, submission timing, and counting units. The Ledger uses the Texas DPS county/offense workbook because it supports the stated county-level analysis and discloses the separate comparison.

## Reproducibility

Run `pnpm audit:data` to download the original source workbooks, independently recompute the values, and compare them with the normalized dataset and rendered pages. Run `pnpm verify` for the full build, SEO, and data-integrity gate.

## Sources

- Texas Department of Public Safety, Crime Records Reports and Statistical Information: https://www.dps.texas.gov/section/crime-records/crime-records-reports-and-statistical-information
- U.S. Census Bureau, Vintage 2025 county population estimates: https://www2.census.gov/programs-surveys/popest/tables/2020-2025/counties/totals/co-est2025-pop-48.xlsx
- Marijuana Policy Project, The State of Enforcement: https://www.mpp.org/issues/criminal-justice/the-state-of-enforcement-progress-and-shortcomings-in-cannabis-arrest-rates/