Methodology
How this works
- Every number has a source. Each value carries its publisher and source URL. If a research vendor or economic model produced it, we name those too. We also record the reference year, the release date, when we retrieved it, and the sha256 of the exact file we read. Tap or hover any value to see all of it.
- Missing numbers stay blank. If a source doesn't publish a value, the cell says so and gives the reason. We don't fill gaps by hand or by imputation. We never quietly swap in another year, either. A value from a different year keeps its own year and gets a flag.
- No score and no ranking. We never add the sections up into one economic total. States are always listed in the same order, and no page sorts them by a value.
- Derived values show their inputs. When we compute a value, it names every input. It also takes the worst grade among them.
- Validation never edits a value. Errors stop a publish. Warnings ship with the data in the validation report.
Comparability grades
Every value gets a letter for how safely it compares across states.
| Grade | Meaning |
|---|---|
| A | Federal series. Same definition, method and period in every state. |
| B | State agency records. Same concept, though definitions vary a little. |
| C | State-commissioned impact study. Vendor and model differ by state. |
| D | Descriptive only. Never compared across states. |
We also record a source tier for the publisher. A is federal and B is a state agency. C is a study a state paid for, and D is a third party.
The federal core
Every state's economy section uses the same federal series:
- BEA GDP by industry, and regional price parities
- Census population and household estimates
- BLS employment and wages (QCEW)
- U.S. Fish and Wildlife Service license and apportionment data
All of them are grade A.
Two tourism measures that are easy to mix up
Tourism spending intensity is visitor spending divided by GDP. It shows scale. Visitor spending counts gross sales and GDP counts value added, so the ratio compares sizes only.
Visitor-exposed GDP share (narrow) is the value-added version. We build it from two BEA industry lines. One is arts, entertainment and recreation. The other is accommodation and food services.
Neither one measures a return on promotion spending.
Known blanks
- State wildlife survey estimates. The 2022 national Survey of Fishing, Hunting and Wildlife-Associated Recreation didn't publish state estimates. Arkansas paid for its own, but it hasn't been released. Maine didn't commission one. We don't fall back to the 2011 state reports. They predate the survey's redesign and measure a different population. National values carry the label Latest available survey estimate: 2022.
- Arkansas state park revenue and spending. Every public source we found reports State Parks together with Tourism and Heritage. None of them reports State Parks alone.
We don't compare harvest counts or grade D values. They're descriptive. The gaps page lists every blank and its reason.