In the last few months, we've been busy working on a dataset that will help our users compare all the historical data gathered over time on the most recent geography. We're proud to announce our latest release: U.S. Decennial Censuses on 2010 Geographies. As of right now, we published 1990 Census data and 1980 Census data on the 2010 geography, and very soon we will release data for other reallocated census years. Read on to find out more about this latest addition and what it means for you.
While US country and state borders rarely change, smaller geographical units, such as census tracts, often do change based on population shifts, which can be a real obstacle when trying to compare data over time. These changes take place after each new census, and they represent something researchers and government officials must take into serious consideration.
As a result of population growth or decline, the Census Bureau will split or combine census tracts. The average population of a census tract is 4,000 people on average (ranging from 2,500 to 8,000 inhabitants) and approximately 1,500 housing units.
If local census statistical areas committees (CSACs) find that a census tract exceeds these limitations when conducting reviews of existing census tracts, as it happened in tract 317.01 (shown in the map below), the Census Bureau might split the tract into two or more smaller tracts (317.03 and 317.04).
Map 1: 1990 Census Tract splitting into two new tracts in 2010
For researchers doing neighborhood-level analysis, this can make comparing tracts across time more complicated. According to the 1990 census, the total population of Tract 317.01 was 5,574. Fast forward to 2010, tract 317.01 has been replaced by tracts 317.03 and 317.04. But what if you wanted to see how many people lived in 1990 in the area that’s covered by Census Tract 317.03?
The latest Social Explorer data release solves this problem. We took the 1990 data and presented it on the 2010 geographies. In other words, even though Census Tract 317.03 didn’t actually exist back in 1990, we used boundary change information and demographic files to estimate how many people lived in that particular area to make cross-decade comparisons possible.
There are also examples of two or more tracts merging into a single new tract. For instance, Sweetwater County, Wyoming went through a major tract change.
Map 2: Tract changes in Sweetwater County, Wyoming in 2010
As you can see on the map, areas BNA 9703, 9702, 9701, 9714, and 9713 were merged, creating a new census tract, 9716. Block numbering areas (BNAs) are geographic entities similar to census tracts, and delineated in counties. (Bear in mind that for the 1990 census, the only difference between census tracts and BNAs was generally the type of organization doing the delineation.) Using Social Explorer’s latest data, you can find out how many people lived in the area that made up census tract back in 1990, even though it didn’t technically exist back then.
To research a specific area at the tract or county level over time, you need to be able to use the same set of geographical boundaries. We took care of all the data reallocation and boundary comparison for you, so you can trace a neighborhood or county over the decades with just a few clicks.
Our new comparative dataset includes over 300 tables, ranging from population, age, sex, employment status, and school dropout rates, to poverty, veteran status, and ancestry for all geographies from census tracts to the whole country.
Get started with: