The single methods chapter for The California Report. Every rule that governs the other chapters is stated here once, so the area and history chapters can carry a one-line footer instead of repeating boilerplate. This chapter is hypothesis-neutral by design: the grading and counter-evidence rules exist to keep any single hypothesis from being silently favored.
Every record carries exactly one grade. A lower grade is never silently promoted to a higher one.
| Grade | Definition |
|---|---|
| A1 | Official machine-readable government dataset; peer-reviewed primary research; official registry publication; official agency report. |
| A2 | Official government webpage; regulatory filing; official meeting document; official GIS service. |
| B1 | University / research-institution report; systematic review; nonprofit technical report with transparent methodology. |
| B2 | Reputable news outlet quoting named sources or documents; public statements from identifiable stakeholders. |
| C | Advocacy materials; law-firm summaries; community petitions; social-media statements; unverified case counts; anonymous claims. |
| D | Speculation; unsourced reposts; unsupported online claims. |
Worked examples from this report: the USDA circulars and the 1912 Johnson proclamation are A1; the DTSC 2008 Interim Guidance and NSW dip-site register are A2; the statewide screening layer and the Pulling 1965 corroboration are B1; the 1908 Los Angeles Herald dipping items are B2 (press quoting a named state officer).
No map object, database row, chemical assertion, case report, chart, or finding exists without a
source_id linking to the source registry. Every inferred record additionally records:
inference method, input sources, assumptions, confidence, and date created. OCR-corrected spellings
(e.g. "Joplln" → Joplin) are flagged as this project's readings, not the source's text.
Every finding of substance carries an explicit counter-evidence and limits note. A finding is not allowed to stand on supporting evidence alone. Two absence-vs-proof cautions apply throughout:
Inferred or "likely" data is never displayed as an actual measurement. Every claim shows its confidence. Language is always disciplined to distinguish: Verified fact · Credible report · Official statement · Scientific finding · Public allegation · Unverified case report · Model-based estimate · Inference · Hypothesis · Missing evidence. Model figures (chapter 72) are labelled MODEL ESTIMATE and carry the "could be off by ~10×" caveat wherever they appear. Confidence badges in the platform: Verified Official · Primary Scientific · Official Public Record · Credible Secondary · Public Allegation · Model Estimate · Unknown.
Land-use and program history reach as far back as reliable public records allow (the tick program is 1907-1912 in Orange County). The health/investigation window is primary Jan 2005-present, extended Jan 2000-present. Every dataset records: publication date, observation/coverage period, retrieval date, temporal precision, and whether the record is current / historical / inferred / archived.
The full graded source list, 109 rows, is in research/source_registry/sources.csv, one row
per source with: id, title, publisher, author, url, publication date, retrieval date, source type,
geographic and time coverage, official/primary/peer-reviewed flags, data format, reliability grade,
known limitations, and notes. Key IDs used across chapters 70-72: S-USDA-C174 and
S-USDA-C183 (USDA BAI circulars, A1); the 1912 Johnson proclamation (California State
Library scan, A1/A2); the State Veterinarian Biennial Reports (A1/A2); S-NSW-DIP (A2),
S-UF-VATS (B2/A1), S-ITRC-CDV (B1) for the jurisdiction comparison; S-STATEWIDE (B1)
for the eleven-community screening layer; and the CDNC press items (B2).
This platform is an independent research and data-organization project. It does not provide medical advice and does not establish that any pesticide, property, organization, employer, school, water provider, government agency, or other party caused any illness. Publicly reported health events may not have been independently medically verified. Geographic and temporal overlap does not establish exposure or causation. Formal conclusions require authorized epidemiological analysis, verified medical information, exposure assessment, toxicological review, and independent scientific evaluation.
This is the standard disclaimer for The California Report. It appears once here; other chapters carry a one-line footer pointing to it.