Dear IPUMS team,
I was checking documentation regarding the CPS coding break between 1992 and 2003 and came across this section of the BLS website: https://www.bls.gov/cps/cpsoccind.htm.
There, it is said that the BLS created its own Crosswalk with conversion rates connecting Census 1990 to Census 2000 occupations. It is said that:
“BLS created conversion factors between the 1990 and 2002 Census classifications. These factors are based on 3-year average survey microdata (2000–2002) that were dual-coded to both the old and new classification systems. They are tabulated at the major group level as well as by the detailed classifications.”
My questions are the following:
1 - The conversion rates are reported in PDF tables, with one page being allocated to each Census 1990 occupation. Translating this by hand to a unique table with occupation codes is cumbersome, so I was wondering if by any chance you also happen to have an excel version of it (or something of the sort). That would be of great help! The tables I have in mind are tables 5 and 6 from the aforementioned link.
2 - Another question refers to the methodological approach. This table is strikingly different from the Census’ own Conversion rates. For instance, 1990 occupation 04 “Chief executives & general administrators, public administration” displays only 2 links in the Census crosswalk, but more than 10 in the BLS one (https://www.bls.gov/cps/detocc19902002.pdf). What are the sources of those differences? I understand that the process through which both crosswalks were assembled are different (with the Census double-coding a sample from the year of 1990), but the discrepancy is still stark. If possible, could you kindly try to elucidate why this is the case, please?
My apologies for the lengthy message and thanks in advance for your time!
Best,
Pietro