I would like to create some descriptive statistics on US congressional district level using 1984/1985/1986 CPS or 1980 Census data. I am new using these datasets and I cant really understand which variable identifies the districts. Any help will be greatly appreciated.
Best,
Fatma
In the 1984-1986 CPS microdata, the lowest level of geography is metro area for those households located in a metro area or state for all other households. Thus, it is not possible to identify respondents’ congressional districts.
In the 1980 Census microdata, the lowest level of geography is the county group, which is a combination of counties or portions of counties that total at least 100,000 in population. Unfortunately, there is no guarantee that county groups will be sufficient for recreating Congressional districts. For example, Pennsylvania’s 12 th district in 1980 consisted of Westmoreland, Somerset, and Bedford counties; however, the county group variable combines Somerset and Bedford with Fulton (11 th ) and Huntingdon (20 th ) counties. Indeed, I was unable to accurately recreate any of Pennsylvania’s 23 Congressional districts in 1980 due to overlapping boundaries.
While it is therefore highly unlikely to identify Congressional districts in the available microdata, there is a solution if you are interested only in descriptive statistics. The NHGIS project, which, like IPUMS, is housed in the Minnesota Population Center, offers many of the Census summary tables at the Congressional district level as far back as 1967. While NHGIS is primarily focused on mapping data, the summary tables can be extracted independent of shape files as .csv files with optional descriptive headers. If it is not necessary to alter your district-level descriptive statistics, these summary tables should suffice.
Hope this helps.