Ask for 2010 but then end up getting ACS estimates from 2006-10

Hi there - I’m trying to get a number of housing and population characteristics for 2010 at the block group level. Using IPUMS NHGIS, I select geographic level = block group; year = 2010, and then I enter my source tables of interest using the topics filter. When I get my downloaded files I see that some of them come from the ACS 5 year estimates for 2006-2010. My understanding was that if I’m interested in 2010, then I should be using the ACS 5 year estimates for 2008-2012 (so that 2010 is the “middle year.” Do I need to specify that dataset rather than filtering by “year?” Thank you!

Currently, the NHGIS year filter pertains to both the period covered by the data (e.g., 2006-2010) and the release year and geographic vintage of the data. ACS 2006-2010 data appear when “2010” is selected because they’re part of the 2010 ACS data release and correspond to 2010 geographic units (as represented in the 2010 GIS files).

Yes, filtering by dataset is one way to ensure that you’re selecting data only from your desired dataset. Another way is to check the “Year - Dataset” column in the selection grid before selecting a table and make sure the table is from your dataset of interest. Sorting the table by “Popularity” can also help to bring the 2010 census data to the top of your search.

You’re right that 2008-2012 is the 5-year period that should generally correspond best to the year 2010, but the filter system doesn’t take that into account. (I’ve thought about changing our system to do that, but it would also be problematic in other ways!)

1 Like