| Please note that the LTDB is based on complex programming and file construction. Send messages about problems you may encounter in using these materials to John Logan (firstname.lastname@example.org). But read and think about all of the documentation first! We will do our best to ensure that the files are accurate and as user-friendly as possible, and this page will be updated periodically.
From this page you may download data files and program code used by the Longitudinal Tract Data Base. Some of these are large files and may require considerable time to download. Cite the US2010 Project webpage as the source of data (
http://www.s4.brown.edu/us2010/Researcher/Bridging.htm). The reference for technical documentation is John R. Logan, Zengwang Xu, and Brian Stults. 2012. “Interpolating US Decennial Census Tract Data from as Early as 1970 to 2010: A Longitudinal Tract Database”
Professional Geographer, 66,3: 412-420.
Coverage: Some census tracts will be missing in the file in various years. Click here for an explanation.
LTDB Standard data set: We now offer two standard data sets. The first is for full count variables. It includes the key variables from Census 2010 (race and Hispanic origin, age by race and Hispanic origin, household composition, and housing occupancy and tenure), with comparable variables from the censuses in 1970-2000 and adjusted to 2010 boundaries. The file only includes the population counts for every variable; we leave it to the user to compute percentages or other constructed variables. The second is for sample count variables, characteristics such as median income and occupation. These sample variables for 1970 through 2000 are from the one-in-six decennial censuses, called "long form data." For 2010 we rely on the American Community Survey (ACS 2008-2012), which pools ACS samples across five years, with 2010 as the midpoint.
The standard data set for each decade is a comma-delimited .txt file with the variable names in the first row. We also provide a codebook that identifies the original tables in each census year from which data have been taken.
Click here to download the codebook.
Select a file type:
Select a year:
The original data with comparable variables for each year (1970-2010) in the original boundaries and with original tract ID codes are also available for download. Full count and sample count data are provided in separate files. Note that most “full count” variables are also found in the “sample count” file, but their values are different because they are based only on the sample. The Census Bureau did not adjust them to match the full count numbers. The "sample" data for 2010 are from the 2008-2012 American Community Survey.
Select a file type:
Select a format type:
Select a year :
User tools: These tools are for persons who wish to examine more closely the geographic relationship files or to use them to harmonize a user-provided data set.
1. Select a crosswalk file for a specific census year in either STATA or Excel format. The 2000-2010 crosswalk identifies what kind of boundary change occurred for each tract (the column heading is "changetype"). 1 = no change. 2 = multiple tracts merged to one. 3 = one tract split into more than one. 4 = 2+ tracts reconfigured into different 2+ tracts.
Select a format type: Select a year for the crosswalk:
2. Choose a program: The LTDB also offers code in Access and STATA that can be used, for any given year, in conjunction with the crosswalk file and an input data file prepared by the researcher.
Select a code type to download:
3. Backwards LTDB: This is for 2000-2010 only and is only available in STATA. Click HERE to download the backwards LTDB crosswalk file and STATA code.