§00
Side by side
Comparisons
Comparison pages are the easiest thing on a site like this to generate and the easiest to get wrong. 5,419 hospitals give 14.7 million possible pairs and 102,268 schools give five billion; almost all of them would be filler. A pair gets a page only when it passes the gate below — the rest render an explanation and are marked noindex.
§01
The rule a pair must pass
A pair only gets an indexable page when all of these hold:
- Same entity type. A hospital and a school do not compare, and the URL says which type it is before it says which two.
- Same state. Measures are collected and funded state by state, so a cross-state pair compares two systems as much as two facilities.
- At least five measures reported by both, so the table has content rather than dashes.
- Both records above the completeness threshold already used for indexing.
Where the publishing agency states no direction for a measure, neither column wins. A bigger number is not automatically a better one, and deciding otherwise would be this site inventing a finding rather than reporting one.
Comparison URLs are deliberately absent from the sitemap. They are reachable from the two records they compare, which is the only place anyone arrives at one from.
§02
Worked examples
Pairs from the live data that pass the gate.
§03
What else answers the same question
Every entity page already carries its peers, chosen by the same rule, with a distance-from-median chart.