4 Comments
User's avatar
Sufeitzy's avatar

You might look into supersectors from BLS which aggregates industries in a much more balanced way:

Mining & Logging

Construction

Manufacturing

Trade-Transport-Utilities

Information

Financial Activities

Professional & Business Services

Education & Health Services

Leisure & Hospitality

Several of these were blips in 1950, and today far outsize Construction or Manufacturing - Professional Services, Leisure, Education - I don’t even know how to classify Professional & Business Services in RIASEC but I can sure say that they attract and retain a lot of very highly paid women.

Expand full comment
Steve Stewart-Williams's avatar

Interesting - will check it out!

Expand full comment
Sufeitzy's avatar

The thing that makes me the most curious is who generated the vocational classification system. I routinely use NAICS in industrial benchmarking, it generally works unless industries are clustered, for instance computers and software are often part of the same company, and more and more equipment has computer controllers. In this index programming and computer systems are at opposite ends… “office management” appears aligned with computing. Curious!

I looked at the RIASEC codes, they were generated in 1958, people were to be one of six “types”, and so on. I’ll go out on a limb and say vocational analysis frameworks generated in 1958 by a man born in 1919 may not be totally relevant to 2023. Men’s labor participation rate was almost three times that of women in this period, there was the concept of a “secretarial pool”, a human-usable computer keyboard was being developed.

I’d be much more interested in a classification system updated to the 21st century, and clustered based using less biased semantic cluster analysis of work/job description.

All this seems to do is reproduce an anachronistic set of job biases; I’d be fascinated to have a set of 20 job types generated by interest women starting with their “dream” jobs on descending list of frequencies, and the same for men.

It mainly tells me what women didn’t have entry to in 1958.

Our world is more services than manufacturing.

Expand full comment
Steve Stewart-Williams's avatar

True, the RIASEC categories were generated a long time ago - more than half a century. (!) It may well be possible to come up with new categories that fit better with modern occupations. But let me give a partial defense of the RIASEC categories. Despite any faults, they do still do a pretty good job of predicting which jobs people will choose and thrive in. Also, by continuing to use the same categories, we can monitor changes and stability in people's preferences over time.

Expand full comment