In the United States, 51 percent of all residents live in a child care desert. Child care supply is especially low among certain populations, with 58 percent of Hispanic/Latino families, 60 percent of rural families, and 55 percent of low-income families living in areas without enough licensed child care providers. In the U.S., 69 percent of mothers of young children participate in the labor force. CAP finds that child care deserts are associated with fewer mothers in the workforce.

CAP collected data on the location and licensed capacity of nearly 235,000 licensed or registered child care providers from every state plus the District of Columbia. The authors included all child care centers, family child care providers, Head Start providers, and public and private preschools in these states in order to get a full picture of the supply of licensed child care options available to nearby communities.

Each child care provider was geocoded based on the physical address provided on their license, giving the authors a precise latitude and longitude for every provider. In cases where the precise address was not available, the authors used the ZIP code to approximate the address. The authors then grouped providers by census tract and added up the total licensed capacity of the locations in each tract. This information was merged with a variety of census estimates on population demographics, population density, family income, poverty, and maternal labor force participation, among other variables. All data are publicly available, and a list of data sources and census variables can be found in the accompanying report’s Appendix.

Using this original dataset, the authors applied the child care deserts definition to the more than 73,000 census tracts across the country. This geographic unit is sometimes used by city planners to approximate neighborhoods. Census tracts generally have a population size of between 1,200 and 8,000 people, with most tracts housing around 4,000 people. These geographic units usually cover a contiguous area, and their boundaries commonly follow identifiable features.