Data Sources

After the 2000 Census, the U.S. Census Bureau stopped using the long form in their decennial census. In its stead, they began collecting data continually in the form of the American Community Survey (ACS). Unlike the decennial census, which provides point in time estimates based on April 1st of the census year, the ACS estimates measure averages over a one-year, three-year, or five-year period. The ACS has approximately 250,000 respondents monthly, totaling 3 million per year. Because of the smaller sample size compared to the decennial census, we need to pay serious attention to standard errors and confidence intervals with the ACS.

Single-year data are only available for larger levels of geographies over 65,000, such as states and large metropolitan areas. Three-year estimates represent 36 month averages, and are available for areas with geographies with populations over 20,000. Finally, five-year estimates represent five-year averages, and are available for all geographies down to the census tract level. If you need special tabulations, ACS also publishes microdata for the Public Use Microdata Area (PUMA) geography.

Tract-level estimates were not available using the ACS until 2010. These estimates represent five year averages of the time period between January 1, 2005 and December 31, 2009.

Indicators on mortality and causes of death shed light on the health, economic, and safety conditions of our communities . Three-quarters of our partners have or make use of cause of death data from vital statistics files.

Every ten years, the Census Bureau conducts a constitutionally-mandated national household survey to count every resident in the United States. The federal government uses Decennial Census data for apportioning congressional seats, for identifying distressed areas, and for many other activities. Beginning with the 2010 Census, the Census Bureau included only the short form (the long form was replaced by the ACS). Short form information is collected on every person and includes basic characteristics, such as age, sex, and race.

The Home Mortgage Disclosure Act (HMDA) requires most lending institutions to report mortgage loan applications, including the outcome of the application, information about the loan and applicant, and location of the property. In 2004, FFIEC expanded the data to include structure type, lien status, and if the loan had high interest rates. FFIEC collects the data in order to determine whether financial institutions are meeting a community’s housing credit needs; to target community development funds to attract private investment; and to identify possible discriminatory lending patterns. The reporting requirements are based on the level of institutional assets and the number of loans originated in metro areas.

You can download the data here.

The Internal Revenue Service (I.R.S.) Statistics of Income publishes zip code level data derived from individual income tax returns, describing characteristics of tax filers, not households or persons. I.R.S. indicators include information about: income level and sources, tax credits and deductions, and tax preparation method. Urban Insitute also as a query system for zip code level data compiled for the Earned Income Tax Credit (EITC) program.

National Center for Education Statistics (NCES) conducts an annual survey of state education agencies to obtain data for every public elementary and secondary school in the United States and its territories, which it then compiles and publishes as the Common Core of Data Public Elementary/Secondary School Universe Survey (CCD). The CCD has two main purposes: 1) to provide an official listing of public elementary and secondary schools and school districts in the nation as a basis for samples for other NCES surveys; and 2) to provide basic descriptive statistics on public elementary and secondary schools. Mostly derived from administrative records, data cover school characteristics such as the school level, grades taught, student-teacher ratio, and federal Title I funding eligibility, and also provide information on enrolled student characteristics, including race/ethnicity, free/reduced price lunch eligibility, migrant status, and gender.

You can download the data here.

Tracking property sales volume and prices enables partners to document how the housing boom and bust is playing out in different types of neighborhoods. Over 70% of the partners have address-level sales data, which are generally obtained from local Assessors’ records. These data are updated more frequently than most other administrative data – ten of those with sales data receive it monthly or quarterly.

HUD has entered into an agreement with the United States Postal Service (USPS) to receive quarterly aggregate data on addresses identified by the USPS as having been "vacant" or "No-Stat" in the previous quarter. HUD is making these data available for researchers and practitioners to explore their potential utility for tracking neighborhood change on a quarterly basis. The potential power of these data is that they represent the universe of all addresses in the United States and are updated every three months. Under the agreement with the USPS, HUD can make the data available publicly at the Census Tract level provided users agree to the terms and conditions of the click-on sublicense.