<aside> 💡 We're hoping to make these scores as useful as possible to other researchers, so would warmly welcome any feedback on what kind of file formats, additional occupational categorization merges, additional documentation, etc., would be helpful. Thanks!
</aside>
<aside> 💡 Strong caveat: This is still very much a work in progress. While the format of the data is unlikely to change (so you can write code using it), the actual numbers may well change as we improve the measure. If you ask, we would be happy to notify you of any quality improvements to the data as they become available.
</aside>
occ1990dd levelexposure_by_occ1990dd_lswt2010.csv
exposure_by_occ1990dd_lswt2010.dta
Exposure score percentiles for occupations at the occ1990dd level. The same data are stored in both a csv file and a Stata .dta file.
Columns:
occ1990dd: Census-based occupation classification code (using David Dorn's occupational schema)occ1990dd_title: Title corresponding to occ1990dd.lswt2010: Labor-supply weight (in terms of FTEs) from ACS 2010.pct_software: Exposure score percentile for software, based on global software patents to date.pct_robot: Exposure score percentile for robots, based on global robots patents to date.pct_ai: Exposure score percentile for AI, based on global AI patents to date.