I am learning to use CPS data from IPUMS as a new user. I calculated the seasonally unadjusted unemployment rate for March 2025 and I get 3.9% using IPUMS CPS. However, BLS data shows the official NSA figure as 4.2%. I am not sure why there is discrepancy? I would appreciate any insights on why there is a difference? I am using Stata 16.
Thanks.
In order to replicate BLS topside labor force estimates, observations should be weighted using the composite weight COMPWT. When I apply this weight in the IPUMS CPS BMS online analysis tool (see screenshot of input fields below), I find that 3.8% of the labor force in March 2025 were unemployed experienced workers (i.e., people who had been previously employed) and 0.4% were unemployed new workers, totaling the 4.2% figure in the official data.
For more information, I recommend reviewing our blog post on replicating alternative measures of unemployment in the CPS.
Thank you so much for showing me how to replicate BLS data using composite weight. But I noticed that my IPUMS CPS data download did not include a composite weight as a default download variable. So I have been using the final basic weight with my calculations. This get me within 0.1 percentage point of official estimates.
I will redownload the data and check the composite weight.
Thank you again for your time. This is very helpful.
Basel Saleh
