---
title: "American Default Research — Methodology"
slug: methodology
summary: "Methodology overview for the American Distress Index (ADI), State Distress Index (SDI), and County Distress Index (CDI)."
type: methodology
canonical_html: https://americandefault.org/methodology/
---

# American Default Research — Methodology

> American Default Research publishes three composite indices: the American Distress Index (national household distress), SDI (state-level distress), and CDI (county-level distress).

## American Distress Index

Five equal-weighted domains: Delinquency, Default & Legal, Debt Burden, Labor, and Safety Net & Buffer. Each input is converted to a distress-oriented Hazen percentile within its own quarterly history; domain scores are averaged, and the composite is the equal-weight mean of the five domain scores.

National band labels publish only when paired with the composite's rank among its own published quarters. The composite value itself is a mean of input percentiles, not a percentile of quarters; the methodology page explains how to read it.

Full methodology: https://americandefault.org/methodology/

## SDI — State Distress Index

State indicators are converted to distress-oriented state percentile ranks and averaged within the publishable family-v1 domains; a state's score is the share of the other jurisdictions with a lower domain average, rounded down, so it reads 'more distressed than X% of the 50 states and D.C.' States publish ranks and score labels, not the national American Distress Index band labels.

Full SDI methodology: https://americandefault.org/methodology/sdi/

## CDI — County Distress Index

Five equal-weighted domains across county-level distress indicators: Delinquency, Default & Legal, Debt Burden, Labor, and Safety Net & Buffer. Indicator values are converted to distress-oriented county percentile ranks; domain scores are averaged; the five domain scores average into a composite, and the CDI score is the share of the other counties with a lower composite, rounded down: 'more distressed than X% of U.S. counties'. County pages print the national rank only for the 100 most and 100 least distressed counties. Counties publish ranks and score labels, not the national American Distress Index band labels.

Full CDI methodology: https://americandefault.org/methodology/cdi/

## Leading Indicator Research

A five-filter pipeline (FDR-corrected cross-correlation, first-differenced series, multi-crisis validation, Granger testing, out-of-sample validation) tests pairwise indicator relationships and records the relationships that survive every validation stage. Domain-to-domain lead claims are not inferred from the indicator-level scanner.

Research data: https://americandefault.org/api/research/leading-indicators.json

## Data Sources

Originating sources include the U.S. Bureau of Labor Statistics, Federal Reserve Bank of New York Consumer Credit Panel, Urban Institute Debt in America, U.S. Census Bureau (ACS, SAIPE, BFS), U.S. Department of Housing and Urban Development Fair Market Rents, Administrative Office of the U.S. Courts (bankruptcy filings), Federal Housing Finance Agency, and U.S. Bureau of Economic Analysis. Selected agency series are retrieved via Federal Reserve Economic Data (FRED), a distribution service of the Federal Reserve Bank of St. Louis. All public. No proprietary data.

Urban Institute data: Breno Braga, Signe-Mary McKernan, Samantha Atherton, and Kassandra Martinchek, "Debt in America," last updated November 20, 2025, https://datacatalog.urban.org/dataset/debt-america-2025. Licensed under ODC-By 1.0. The County Distress Index is American Default Research's own calculation; the Urban Institute does not produce or endorse it.