Census tracts are the building blocks of local demographic analysis — but what exactly are they, how are they drawn, and what can their data tell you about a neighborhood?
If you've ever looked at a detailed demographic map of your city and wondered what those small polygons represent, they're almost certainly census tracts. Census tracts are small, relatively permanent statistical subdivisions of a county, designed by the Census Bureau to have roughly homogeneous population characteristics and between 1,200 and 8,000 residents, with an optimum size of about 4,000.
Tracts are drawn — and redrawn each decade after the Decennial Census — to contain relatively similar populations and to respect visible boundaries like streets, rivers, and other geographic features. The 4,000-person target means a dense urban neighborhood might fit in a single city block, while a rural tract could cover hundreds of square miles. The US has approximately 84,000 census tracts.
Every tract has a unique 11-digit identifier: a 2-digit state code, a 3-digit county code, and a 6-digit tract number. Tract 06037101110 is in California (06), Los Angeles County (037), tract 1011.10 (101110). CensusDepth uses these GEOIDs directly in tract profile URLs — you can look up any tract at /tracts/{geoid}/.
The ACS 5-year estimates are the main data source for tracts. Because 5 years of data are pooled, even small-population tracts have reliable estimates for most variables. However, some variables with small cell counts — like specific occupations or detailed racial/ethnic categories — may be suppressed (shown as null) to protect respondent confidentiality.
Tract-level data includes: population, age distribution, race/ethnicity, income, poverty rate, educational attainment, housing values, rent, homeownership rate, and commute patterns. All are available in CensusDepth's tract profiles.
Census tracts are not neighborhoods, though they often approximate them. A named neighborhood like "Lincoln Park" in Chicago might encompass several tracts, or a single tract might span parts of two adjacent named neighborhoods. Tracts are defined by the Census Bureau for statistical purposes — they don't have names, signs, or community organizations. But because they're stable over time (modified only every 10 years) and consistently sized, they're the best available unit for longitudinal neighborhood-level analysis.
You can look up any address's tract GEOID through the Census Bureau's geocoder. Once you have the GEOID, you can browse the full demographic, income, and housing profile directly on CensusDepth. Tract data is especially useful for understanding within-city inequality — the difference between a high-income tract and a poverty-concentrated tract in the same city can be dramatic even when they're a mile apart.
One important caveat: the ACS is a sample, and for small tracts, margins of error can be large relative to the estimate. A tract-level poverty rate of "22% ± 15%" means the true value could plausibly be anywhere from 7% to 37%. Always check the margin of error on tract-level data before drawing strong conclusions. CensusDepth displays these margins prominently for variables where they're large enough to matter.
The American Community Survey 5-year estimates are the backbone of modern demographic research. Here's what the data covers, how it's collected, and why it matters for every community in America.
Using a composite of median home values, median rents, and household income, we identified the ten US cities with the best affordability profiles in the latest Census data.
From median home values doubling in a decade to record rent-cost-burden rates, the Census Bureau's ACS data paints a vivid picture of housing stress across America.