3 ms·
Whenever you aggregate data, particularly about human behaviors, the grouping is always going to be somewhat arbitrary. An underlying assumption, which we can
by jrumbut 11y ago
Whenever you aggregate data, particularly about human behaviors, the grouping is always going to be somewhat arbitrary.
An underlying assumption, which we can make because the consequences of being wrong are so low (god forbid we be entertained by an inaccurate infographic!), is that all occupations will lose some people that should be counted with them, and gained others they don't deserve, and that this will balance out.
It would be interesting to learn if there is some sort of systemic bias, say jobs commonly held by women were more or less likely to be lumped together, or jobs with more itinerant workforces weren't counted well.
- Retric 11y agoIMO, there is a usage problem. If you define an arbitrary worker type 1 as X, and then graph the % of the workforce, average hours, or change in pay you can see meaningful trends. However, if you create arbitrary category's and then just look for the largest group or average salary etc. then how you define each group becomes extremely important. IE: Heart Surgeon with GP ends up an average that nobody really fits.
- jrumbut 11y agoIt does appear, if the citation of IPUMS at the bottom (a cool source of data, for those who haven't seen it: https://usa.ipums.org/usa/ https://usa.ipums.org/usa/ ) means anything, that they are at least using standard classifications. That doesn't mean their distinctions make sense, but at least they're openly available and documented.