Status:Closed    Asked:Oct 20, 2015 - 04:03 PM

Is it possible to generate the U6 unemployment rate using the data here?

Can the variables available in the ipums extracts be used to generate the U6 unemployment rate that the BLS produces. I would be generating it at the state level.

Is there a variable here tha captures marginally attached?

Do you have the same question? Follow this Question

Staff Answer




It is possible to calculate the U-6 unemployment rate for ASEC samples with the IPUMS-CPS data. To do this, you will need to use EMPSTAT to identify unemployed respondents, WANTJOB to identify the marginally attached respondents, WKSTAT to identify those employed part-time for economic reasons, LABFORCE to identify the total labor force, and STATEFIP to identify state of residency.

Unfortunately, it is not possible to calculate the U-6 unemployment rate for the IPUMS basic monthly samples. Currently, we do not provide the PEDWWNTO CPS variable that is used to identify those marginally attached to the labor force in the basic monthly samples. However, it is possible to merge your IPUMS-CPS data extract with its corresponding NBER dataset. The easiest method is to attempt a sequential merge (i.e. first NBER observation to first IPUMS-CPS observation, second to second, etc.). Since our datasets share the same sort order as the original NBER datasets, this should be successful. You should verify such a merge by confirming that sex, age, and race match between datasets.

Hope this helps.


Oct 21, 2015 - 12:31 PM

Report it


I would like to create an unemployment rate with the CPS 1994-1999 March data...
calculated unemployment rate is different from BLS report
Why are not all the "not in the labour force" categories filled for the EMPST...
calculate U6 from IPUMS CPS basic monthlies?
Login   |   Register

Recently Active Members

View More »

Share |