Hello,
I’ve been using the original CPS from NBER, both basic monthly files and the ORG, for many of the work I do. One of the reasons why I was using the data from NBER is to match individual records from one month to the following.
However, it seems to me that this matching should also be possible using IPUMS-CPS directly. In general, I would like to know what is the advantage of using the data directly from the NBER (which implies hours of looking at documentation and the risk of errors during the extraction) over using IPUMS-CPS? What is it that IPUMS-CPS does not have?
Many thanks,
The underlying data is the same between NBER and IPUMS, however the IPUMS extract system allows users to take a subset of variables, if desired. Additionally, IPUMS variables are “harmonized” to be consistent over time and have detailed documentation available via a hyperlinked web format. IPUMS also has constructed a number of variables such as family interrelationship variables, that may be useful for some types of analysis.
The main benefit of using NBER would be if a researcher is aiming to replicate a previous analysis done using NBER data or if there is a variable that is currently not offered on IPUMS-CPS. Although, in this latter case, users should check in with ipums@umn.edu so that we can make sure that he variable is indeed not available, and so that we know users are interested in the given variable.