No Clocks

"Geography is the science of our world. And it’s an organizing principle for all other sciences. You can overlay biology on top of geography, overlay human patterns like land uses or where different people live on top of geography. You can see it all together* as an integrated whole." - Jack Dangermond, Esri President*

Demographics for the United States are available for 27 geographic layers. At the core of the geography stack is the standard Census Bureau hierarchy: Nation, States, Counties, Tracts, Block Groups, and Blocks. The lowest level of Census geography (blocks) is used within Esri software for data apportionment to produce summary estimates for custom areas. Additionally, four other geographic layers are shown in a separate hierarchy chart later in this tutorial, focusing specifically on tribal areas. The chart below illustrates the relationship between the non-tribal boundaries.

The Nation is the broadest geographic level for the United States covering the 48 contiguous states, Alaska, Hawaii, District of Columbia, and Puerto Rico (not shown).

All 50 states and the District of Columbia are identified by a two-digit Federal Information Processing Series (FIPS) Code and assigned alphabetically based on the state name. The Census Bureau treats Puerto Rico (FIPS 72) as a state for the purpose of presenting data.

States are subdivided into legal/administrative areas called Counties (or equivalent areas). Texas is comprised of 254 counties.

In Alaska these equivalent geographic divisions are referred to as boroughs, or census areas (which are statistical). In Louisiana, these areas are called parishes. The District of Columbia, a state equivalent, is not further divided by county but also treated as a county equivalent for statistical purposes.

Maryland, Missouri, Nevada and Virginia have incorporated places that are not part of a county and are treated as county equivalents.

In Puerto Rico, county equivalents are called Municipios.

Every county (or equivalent area) is assigned a unique five-digit code where the first two digits represent the** **state FIPS followed by a three-digit county code that is assigned alphabetically within each state.

Counties do not cross state boundaries.

The next level of geography, Census Tracts, are statistical subdivisions of counties or county equivalents as defined by the Census Bureau and its local stakeholders.

The size of a tract is defined by settlement density and partitioned using the physical landscape or legal boundaries. From Census 2020, a typical tract will contain around 4,000 persons but generally vary between 1,500 to 7,000 persons, as per the 5th and 95th percentiles of the distribution.

Census tracts are identified by an eleven-digit code. Sometimes this code may include a decimal separator between the ninth and tenth digit. The first five digits of a tract code begins with the two-digit state FIPS and three-digit county code**  **followed by six digits identifying the tract portion of the full code (ex: SSCCCTTTT.TT, where S=state, C=county, T=tract).

Tracts are non-overlapping and wholly contained within county and state boundaries.

Census tracts are further subdivided into areas called block groups, which are comprised of a collection of blocks. As with tracts, these areas are defined by the Census Bureau along with local stakeholders. From Census 2020, a typical block group will contain around 1,300 persons but generally vary between 600 to 2,600 persons, as per the 5th and 95th percentiles of the distribution.

The first five digits of a block group code begins with the two-digit state FIPS, the three-digit county code, the six-digit tract code, followed by one digit identifying the block group (ex: SSCCCTTTT.TTB, where S=state, C=county, T=tract, B=block group). Block groups with a "0" code represent all-water block groups.

Block groups are non-overlapping and wholly contained within tract, county, and state boundaries.

The last component of the nested layers are Census Blocks. This is the Census Bureau’s most granular geographic area. Blocks are generally small in area, particularly in urban areas. However, blocks in rural or remote areas may cover hundreds of square miles.

The Census Bureau establishes these statistical areas that are typically bounded by visible features that are incorporated in this process such as streets, rail lines, hydrological features, and so on. A group of blocks forms the higher level of block group. There are 8,132,968 blocks in the Census 2020 geography, compared to 8.2 million in Census 2000 and slightly more than 11 million in 2010. Block codes follow the format of SSCCCTTTT.TTBLLL, where S=state, C=county, T=tract, B=block group, L=block.

Blocks are foundational layer of the geographic boundaries in this nested stack

Esri summarizes demographic data for 12 other area types that are not part of the above nested hierarchy:

County subdivisions (CSDs) are the primary divisions of counties and include census county divisions (CCDs), minor civil divisions (MCDs), census subareas, and unorganized territories. CCDs exist in 20 states, MCDs are in 29 states plus D.C., while census subareas (CSD equivalents) exist only in Alaska. In Puerto Rico, County Subdivisions are referred to as Barrio, Barrio-Pueblo, or Subbarrio. Some CSDs represent legal/administrative entities (MCDs) while others are established by the Census Bureau and local stakeholders for statistical purposes (CCDs). County Subdivision codes follow the format of SSCCCDDDDD, where S=state, C=county, D=Subdivision.

Places include incorporated places (usually cities, towns, villages, or boroughs), census designated places (CDPs), and balance portions of consolidated cities. Places are always within a single state, but they can and often do cross county boundaries. In Puerto Rico, places are referred to as Zona Urbana and Comunidad. Places are uniquely identified using a two-digit state FIPS code and a five-digit place FIPS code, following the format of SSPPPPP, where S=state, P=Place.

Core-Based Statistical Areas (CBSAs), which include metropolitan and micropolitan statistical areas, are composed of one or more counties and are defined by the U.S. Office of Management and Budget (OMB). A metropolitan statistical area is associated with at least one urbanized area of 50,000 or more inhabitants. A micropolitan statistical area is associated with at least one urban area of at least 10,000 people, but less than 50,000. The central counties of a CBSA are included based on population in urban areas while outlying counties are included in the CBSA based on a worker commuting threshold. Principal cities in the name of CBSAs are determined by population and employment thresholds.

Designated Market Areas (DMAs) are television markets defined by The Nielsen Company and are revised on an annual basis. Most DMAs are composed of one or more counties, although a few DMAs include county parts. For the DMAs that split counties, Esri finalizes the DMA boundaries to align with boundaries of block groups. While the vast majority of the United States is covered by a DMA, there is an area in Alaska that is not part of any DMA. The Nielsen Company does not produce DMAs for Puerto Rico.

Public Use Microdata Areas (PUMAs) are statistical regions created in collaboration with State Data Centers (SDCs) with each area containing at least 100,000 people. They are established from counties and/or census tracts and remain stable throughout the decade. PUMAs cover all of the United States, Puerto Rico, and certain island areas that meet this population requirement. They are significant because they serve as the most detailed geographic areas referenced in publicly available microdata files, such as the Public Use Microdata Sample (PUMS) data from the American Community Survey. PUMA codes are 7 characters long, consisting of a 2-character state identifier (S) followed by a unique 5-character code (P). They are formatted as SSPPPPP, with each PUMA code being unique within its state. Codes are assigned sequentially from the northwest corner of the state. If the 4th and 5th characters are "00," the PUMA covers an entire county or multiple counties. If they increment from "01," the PUMA includes parts of one or more counties.

H3 is an open-source global hierarchical grid system developed at Uber for indexing geographies into a hexagonal grid that is often used to understand patterns in large geospatial datasets. H3 hexagons keep their shape and indexing consistent over time, making it easier to track data changes. Each hexagon is equidistant from its neighbors, reducing sampling biases. Esri creates data for Hex resolutions for two through seven. Note that hexagons cannot be perfectly divided, so smaller hexagons are only roughly contained within larger ones. H3 index values (IDs) are in a 15-character hexadecimal format, with the second character representing the resolution.

Congressional districts (CDs) are the areas from which individuals are elected to the U.S. House of Representatives. Once the apportionment of congressional seats is made based on census population counts in a state, each state establishes CDs to elect representatives. A congressional district is uniquely identified using a two-digit state FIPS code and a two-digit CD FIPS code: SSDD, where S=state, D=District, where congressional district codes range from “01” to “53”, the code for “At Large” (single district for state) is “00”, the nonvoting delegate code is “98”, and “ZZ” is used for any area not assigned to any congressional district. The District of Columbia and territory of Puerto Rico are assigned a code of 98 to identify their single area, nonvoting delegate status in Congress.

School districts are geographical areas designated by local or state officials to organize and manage public education within a specific region. The Census Bureau tabulates data products for these areas, largely for use by the U.S. Department of Education.  Esri provides three types of school district boundaries: unified, secondary, and elementary . School districts can vary significantly in size, with some serving only a single school and others encompassing multiple schools across a large geographic area. In general, where there is a unified school district, no elementary or secondary school district exists; and where there is an elementary school district, a secondary school district may or may not exist. A school district boundary is uniquely identified using a two-digit state FIPS code and a five-digit school district code and are always within a state.

State Legislative districts are geographical areas designated by areas from which voters elect a person to be their representative in state or equivalent legislatures. Each district sends one or more representatives to the state house (lower chamber) and senate (upper chamber). Members of the smaller upper chamber typically represent more people and usually serve longer terms than members of the larger lower chamber.  Through the Redistricting Data Program, the Census Bureau updates state legislative districts every two years. Nearly all state legislatures include an upper chamber (senate) and a lower chamber (house) with districts for each represented by separate upper and lower layers. Exceptions to this are Nebraska which has a unicameral legislature and the District of Columbia which has a single council. In these situations, areas are treated as upper chamber by the Census Bureau which means that the lower chamber state legislative district layer does not completely cover the US.

State Legislative Districts are represented by a unique 3-character code, identified by state or equivalent. Codes follow the format of SSDDD, where S=State and D=District. Listed below is the geographic code breakdown for the Upper and Lower Legislative districts for State Senate District 14 and State House District 53.

Residential ZIP codes are defined solely by the U.S. Postal Service to expedite mail delivery not associated with other geographic layers. Because ZIP code boundaries are not contiguous census geographic areas, or stable over time, data estimated for ZIP codes can also change; sometimes monthly or whenever the U.S. Postal Service revises delivery routes. Residential ZIP code data is estimated from block data established from block group estimates, using a correspondence file created by Esri to assign Census block points to ZIP code boundaries. Esri updates ZIP codes each year based on the latest boundaries and vintage obtained from TomTom. These boundaries completely cover the U.S. and Puerto Rico, including areas where there is no mail delivery (very remote/rural areas).

No official source for ZIP code boundaries exists. As a result, depending upon the source that created them, differences will exist due to the selection of data inputs and methodological approach used to draw these postal polygons. Moreover, since this inventory is subject to monthly updates by USPS, there can be vintage differences in the boundaries as well. So, it is important to use caution when combining ZIP code summary data from sources outside of Esri Demographics.