Interactive publication
Back to the catalogueData qualityCase 04 of 14
Confidence-convention mismatch
A ~67%-confidence radius and a CEP50 radius differ enough that pooling them unconverted yields containment estimates wrong by a factor of two or more.
Class
Case to account for
Affects
2 conversions
01
Detection
Mixed sources with no declared convention; empirical containment far from the nominal rate.
02
Mitigation
01
Require the convention as mandatory metadata
02
Normalize all radii to one confidence level before pooling
03
Prefer vendors that publish a containment calibration
03
Affected conversions
IP lookupH3 (probabilistic crosswalk)
OPENPoint+radiusH3
OPENChapter
IP Lookup To H3More in Data quality
Accuracy-radius attritionAxis-order reversal (lat/lng swap)Coastal / mixed pixelsDuplicated region IDsGeocoding uncertaintyIncomplete geometryMissing CRS declarationMissing polygonsRaster no-data valuesRaster/cell resolution mismatchRounded / truncated coordinatesWrong CRS assumedZero-island coordinates