Research libraryMethodology & data
Methodology & Research Brief · v1.0.0
What the index measures, how it is built and validated, what it does not say, and how to become a founding research partner.
ERI Methodology & Research BriefAug 27, 2026United StatesPublished
The weights
Dimension weights, set by the authors and disclosed. Users who prefer different weights can recompute from the published results table.
Inputs
17
indicators, each with its source variable ID
3
federal sources: Census ACS, BLS LAUS, BEA
3,222
counties scored on one method
Normalization and grades
Each indicator becomes a national percentile rank (0–100); dimensions average their indicators; the composite is the weighted mean.
A ≥ 80
grade cut point
B ≥ 65
grade cut point
C ≥ 45
grade cut point
D ≥ 30
grade cut point
Validation and sensitivity
84%
of grades hold under ±25% weight perturbation
2020–22
out-of-time validation window (COVID)
3,080
counties in the empirical weight check
Limitations stated in the brief: one validation shock; county-government-only fiscal data; ACS margins of error in small counties; cross-sectional associations in the cascade model overstate short-run causal effects.
Founding research partners are invited to validate and refine the methodology. Sponsors support the research; they do not influence the methodology or the findings.
Dig deeper
See the report's methodology section.
Sources carry different vintages and lags by design and are never combined inside one figure. The write-up's caveats section says what the index does and does not claim.
Alt-30 (2026). ERI Methodology & Research Brief, v1.0.0 (v1.0.0). Alt-30. https://portal.alt-30.com/reports/eri-methodology
More in methodology & data