Evidence in the inverse crosswalk is counted in cities, never in datasets

A dataset far from every SDG indicator may be a gap in the framework, or may be a dataset the matcher missed — and we cannot tell which from the dataset alone. A positive control proved it: NYC's Housing Maintenance Code Violations is a hand-verified match for 11.1.1 and lands at the 24th percentile, inside the tail.

So a theme counts only through the number of independent city catalogs it appears in. One city publishing forty parking files is a filing habit; twenty cities each publishing one is a category of municipal governance. This is the same discipline as reporting a table across all cases rather than testing one, applied to a corpus instead of a function.