Merging NHIS data with Census data


I am interested in using NHIS data and conducting a county level analysis of barriers to care. Is it possible to merge two datasets to conduct this type of analysis? Or would I need to conduct two separate analyses as they are different samples?


I am not certain which two data sets you are referring to in this question. Since, you’ve nested this question under the IPUMS USA category, it seems that you may be asking about merging US Census or ACS data with the NHIS data. Let me know if I am missing something.

With that said, this sort of merge is complicated due to several details. First, the sample of individuals is different between the US Census/ACS and the NHIS data. Therefore, merging at the individual level will not be possible. Second, in the public use NHIS data the lowest level geographical area is the REGION and in the US Census/ACS data the lowest level geographical area is the PUMA. Therefore, merging at a relatively low level geographical area will likely not satisfy your requirements for your research.

Hi Jeff,

Thanks for the quick reply! Those are the two data sets I was referring to, and I imagined that it would not be possible to merge them due to the issues you mentioned, but I just wanted to double check. It sounds like the only geographical analysis we can complete with the NHIS data then is by region. Thanks!