The Census Bureau tabulates population by census geography ( ie., census blocks, tracts, counties), not by the precincts and wards that election offices actually run. Without a way to bridge that gap, states would receive decennial population counts that could not be readily matched to their own election geography. The Voting District Project (VTDP) is that bridge: once each decade, as Phase 2 of the Redistricting Data Program, the Census Bureau invites each state to submit its voting districts – its generic term for precincts, election districts, and wards – so that population data can be tabulated for those areas and delivered as part of the GIS redistricting data that enables states and local governments to redistrict their political maps. The program has run every decade since the 1990 Census, when the Bureau first adopted the generic term “voting district (VTD)” and began identifying which submitted areas matched real precincts and which only approximated them. Participation in the VTDP is voluntary, and the value of what comes back depends on local accuracy.
“Voting district,” or VTD, refers to the small areas that state and local governments create to administer elections, such as precincts, election districts, and wards. Through the Voting District Project, states provide the boundaries, codes, and optional names for these areas so that the Bureau can tabulate population data for redistricting political boundaries in accordance with Public Law 94-171.
According to the U.S. Census Bureau’s Geographic Areas Reference Manual, a census VTD is not always identical to the precinct an election office actually administers, because the Bureau requires VTDs to be built from whole census blocks. Where a local precinct’s boundaries do not follow census block boundaries, the submitted voting district is adjusted to fit them, so that the census version approximates rather than exactly reproduces the precinct on the ground. For anyone working with the data, the practical takeaway is to treat a VTD as the census stand-in for a precinct, usually close, but not a guaranteed match to the boundaries an election office maintains.
The Bureau records the pseudo and actual status of a VTD in a voting-district indicator field: an “A” marks an “actual” VTD that matches the precinct on the ground, while a “P” marks a “pseudo” VTD that only approximates it. GIS analysts who match/join precinct-level election results to VTD-level census counts should treat pseudo-VTDs with caution, since the geography is approximate and aggregating data from the census block level is generally more reliable.
The extent to which local precinct boundaries align with census VTDs can be shaped early in the redistricting data program cycle by ensuring that the physical features defining local precinct boundaries match census block boundaries during the Block Boundary Suggestion Project. This allows more precincts to align with whole census blocks, which increases the accuracy of the Census Bureau redistricting data file by reducing “pseudo” VTDs. The practical payoff is straightforward: fewer pseudo VTDs mean census precinct geography that more closely matches what election offices actually administer.
For more on the program and where voting districts fit within it, see the Census Bureau’s Redistricting Data Program Management page.
For the technical definition of the actual/pseudo indicator and guidance on working with VTD data, see the Census Bureau’s 2020 P.L. 94-171 technical documentation and the Redistricting Data Hub.
Find us on:
