CPS and NBER merge

I want to merge the IPUMS CPS basic monthly extracts to the NBER basic monthlies. In the forums here, this blog post is often pointed to as offering the answer: IPUMS FAQs: How can I merge my IPUMS CPS file with an NBER CPS file? – Use It for Good

However, that post references the variable HUHHNUM which is not available in the most recent data.

If I want to merge using recent years, like 2019 to 2023, what should I use to merge to NBER?

For samples after May 2004, you can merge IPUMS CPS observations with the original Census data using only the variables HRHHID, HRHHID2, and PULINENO (LINENO in IPUMS) as linking keys. HUHHNUM is only necessary for linking samples prior to this period. I did encounter difficulty using this method with NBER files since these variables appear to have been edited there, but I was able to use and verify this linking method with the original Census data. I have reached out to the NBER for more details and can share if I find anything new regarding those files.

1 Like

Hi Adam,

I saw your email asking about linking and wanted to share some additional info. Ivan reached out to us at the NBER about the issue with not being able to link NBER’s CPS files to IPUMS. I did some digging and discovered an error in our source code that causes incorrect values for HRHHID, HRHHID2, and two other variables to be written to the NBER data files.

We think these problems could extend back to the 2019 files and are currently working on correcting all relevant files. I will post an update here when that is complete along with info on which files were updated.

I’ve uploaded a corrected version of the January 2023 monthly basic Stata file for testing purposes in the meantime. Please let me know if you have any issues with linking or notice any other issues.

Thanks again for your question and thanks to Ivan for reaching out!


1 Like

Oh great, thanks Chris. Do I get an ipums mug for that?

Haha! That would be up to IPUMS. I unfortunately don’t think we have any NBER mugs but that’s a cool idea!

Edit: I probably should have mentioned that I work at the NBER not IPUMS. Sorry!

To be clear, I have been able to merge ipums to census files using year month hrhhid hrhhid2 lineno, so I question whether the ipums data is the issue.

Ah okay that makes sense! So its not ipums that is the problem here but NBER data, thank you that clarifies. But you should definitely get mugs.

1 Like

Hi again,

@econuser1 Just wanted to give an update that we are now working on a solution that makes it easier for users to merge NBER CPS with IPUMS CPS, at least for certain years, and provide informational text for years where users will have to convert NBER identifiers to be able to merge with IPUMS. I’ll let you know more details when I have them.

1 Like

@econuser1 and others: I just finished updating all of the programs, raw files, and processed files for the CPS monthly basic to correct the issues discussed above. I also reorganized and renamed all of the files to be more consistent. If you have any scripts referencing https://data.nber.org/cps-basic2 you will need to update these to reflect the new naming convention and directory structure found at https://data.nber.org/cps-basic3. I wrote some documentation on the new structure and naming convention and of course would greatly appreciate any feedback on the documentation or anything else.

Edit: Adding link to the NBER’s CPS monthly basic data for future viewers.

1 Like