This dataset contains information derived from a random sample of deidentified, consumer-level records from a major credit bureau. The credit bureau data are from August 2023 and contain more than 10 million records. We also incorporate estimates from summary tables of the US Census Bureau’s American Community Survey (ACS). We use ACS one-year estimates (2022) where possible, but for areas with smaller populations and for metrics that incorporate zip code–level information, we use the ACS five-year estimates (2018–22). Finally, we also use estimates from the 2020 Decennial Census.

The credit bureau data do not include information about race, so the white and people of color metrics are based on the racial makeup of zip codes within the geographic area (nation, state, and county). Specifically, the majority-white communities are based on credit records for people who live in zip codes where most residents are white (at least 50 percent of the population is white), and communities of color values are based on credit records for people who live in zip codes where most residents are people of color (at least 50 percent of the population is African American, Hispanic, Asian or Pacific Islander, American Indian or Alaska Native, another race other than white, or multiracial). In prior editions of Debt in America, the threshold for defining communities of color and majority-white communities was 60 percent. As such, the 2024 update is not directly comparable with previously published breakdowns of credit health indictors for communities of color and majority-white communities. The ACS data include information on people’s race, so the white and people of color values for ACS metrics are calculated directly for those populations.

Data are reported at the national, state, and county levels for the 50 states and Washington, DC. Data on young adults are only reported at the national and state levels. In the dataset, missing and unavailable values are noted with “NA.” Some values are missing because no data are available. Others are missing because credit bureau metrics are not reported when they are based on fewer than 50 people. In other cases, values for white communities and communities of color are not reported because there are no zip codes with predominantly white populations or populations of color in the county or state. The map breaks are determined using the Jenks Natural Breaks method.

In 2022, the census implemented significant changes to the definition of county-equivalent boundaries in the state of Connecticut, which our credit bureau data provider has not yet updated. As a result, we continue to report county-level credit statistics for Connecticut using the definition of 2021 county-equivalent boundaries. Further, county-level data on household incomes, educational attainment, insurance coverage, and racial demographics from the 2022 or 2018–22 ACS are not reported for Connecticut counties. Zip code–level data on the racial makeup of communities from the 2022 or 2018–22 ACS are still used to identify communities of color and majority-white communities at the state and county levels for Connecticut.

For more information, please see the technical appendix.