Crosswalks of occupation with CPS 1990-2020 sample

[Tariq Abdullah]

I'm trying to create crosswalks of occupation with a sample of 30 years from CPS ( 1990-2020). OCC1990 was used in CPS from 1992-2002 and OCC2010 are being used in CPS since 2002 - if I'm not wrong ( Though the correct information is 2002 census occupation was used since 2002 but the occupation classification of 2002 and 2010 are pretty much similar - so that doesn't make too many difference)

My question is since my data sample is from 1990-2020 do I need to do all the crosswalk from 1990-2000 , 2000-2010 to get a clear cut estimation of all sort of occupation classification. I'm attaching the crosswalk I found from Dorn (2009) and Autor and Dorn (2013 ) here. Here, 1990dd is the most versatile amagamation of all sort ofoccupation.

Anyone with some ideas please enlighten me what I gotta do with CPS data raning from 1990-2020. Do I need to create the crosswalk from 1990-2000 and 2000-2010 with using the attached crosswalk file ?

Eventually, my target is to merge all occ1990 and occ2010 with occ1990dd values. So that, I can merge my CPS data from 1990-2020 with another data file that needs to be merged with exisiting column of occ1990dd

[Dick Campbell]

This can be a really nasty problem. The folks at IPUMS (Integrated Public Use Micro Samples) have done a huge amount of work with the CPS and I would strongly suggest that you take a look at their website (https://cps.ipums.org/cps/about.shtml). I suspect that it will save you a lot of time. The data are available at no charge.