Methodology
United States market • Applies to every published report on this site
1. What We Measure (And What We Do Not)
Car Issue Check summarizes reporting activity in public records — never mechanical condition.
- What we measure: owner-complaint filing activity, safety-recall campaigns, ODI investigations, federal crash-test ratings, and manufacturer service guidance.
- What we do not measure: individual vehicle accident history, salvage/title records, odometer accuracy, maintenance quality, or lien status. Those require a separate vehicle-history report.
2. No Score, No Signal, No Rating
Every published report presents official records directly — it does not compute or imply a reliability score, a "reported issue signal," a risk band, or a buy/avoid recommendation of any kind. This is a deliberate, tested product decision, not a missing feature: public complaint data cannot support the precision a numerical score would imply, and this project's own validation work found that even classifying a single piece of evidence can flip a report's internal usefulness assessment — a risk we chose to disclose rather than hide behind a single number.
- We use Manufacturer Service Guidance for technical service bulletins. A guidance entry means the manufacturer issued service guidance related to a component or condition — most entries are diagnostics, software updates, or routine procedures, not defects.
- We state plainly that complaint volume does not prove failure prevalence. Filing counts are heavily influenced by sales volume, vehicle age, and publicity.
3. How a Report Is Built
Each published report is built from official public sources only (NHTSA complaints, recalls, investigations, and safety ratings; EPA fuel-economy data; manufacturer service-bulletin bulk files) for one specific year/make/model/configuration, resolved by VIN decoding rather than assumed from a model name. Individual complaint records are grouped into reported issue patterns by affected system, with plain-language symptoms and the number of indexed filings behind each pattern. Grouping is descriptive; it does not verify any individual claim.
4. Evidence-Strength Labels
Instead of a severity grade, each reported issue pattern carries a neutral label describing how corroborated it is by official records — never how common, how dangerous, or how likely:
A recall, investigation, or manufacturer bulletin describes the same pattern as the complaints.
Multiple independent owner complaints describe the same pattern, without official corroboration.
Based on complaint filings alone, with no official record addressing the pattern.
A small number of reports — enough to disclose, not enough to call a clear pattern.
5. Missing Data & Neutral States
When a data category is unavailable or too sparse, the report shows an explicit state ("Insufficient data", "Not rated", "None found in dataset") instead of estimating.
6. Limitations
- Complaint volume does not prove failure prevalence.
- This report does not predict whether an individual vehicle will experience a problem.
- This report does not replace a professional pre-purchase inspection.
- Public databases are living records — our indexes may lag official systems.
- Newer model years have accumulated less time on the road and therefore fewer reports.
We refresh our indexes periodically but do not guarantee real-time synchronization with government systems. Manufacturers add and amend service data continuously — always confirm recalls with the official VIN-specific lookup.