Is IPUMS CPS a good choice doing county level analysis

I have some specific questions about Current Population Survey, and wonder anyone could provide some insight on them?

  1. I know the CPS is not designed to provide sub-state estimates but IPUMS CPS does provide county variable and they do have the five counties in NYC . How reliable are the results if I am analyzing monthly employment trend at the NYC level? Particularly, if I am just looking at several occupations/industries in NYC, do you have any sense of how reliable they are?
  2. What about borough level analysis (particularly at each county in NYC) using CPS?


The CPS sampling treats New York City (consisting of Bronx, Kings, New York, Queens, and Richmond Counties) the same as states, so statistics about NYC should be as reliable as state-level estimates. You can read more about the CPS sample design in Technical Paper 77 (specifically see page 52).

The CPS is not designed to produce reliable statistics at the county level. However the counties of NYC are all identified in the CPS and statistics from these counties will be representative. Also, since the counties of NYC are large, these will be more reliable than most county estimates.

In both cases, how reliable your analysis is will depend on the size of the sample, the size of the industry or occupation and the particular variables you are using. I would suggest calculating confidence intervals for your statistics to get a clearer picture of their reliability.