Awhile ago we used data from the National Election Study to estimate the partisanship of different occupation categories:

As we wrote at the time:

Professionals (doctors, lawyers, and so forth) and routine white collar workers (clerks, etc.) used to support the Republicans more than the national average, but over the past half-century they have gradually moved through the center and now strongly support the Democrats. Business owners have moved in the opposite direction, from close to the national average to being staunch Republicans; and skilled and unskilled workers have moved from strong Democratic support to near the middle.
These shifts are consistent with the oft-noted cultural differences between Red and Blue America. Doctors, nurses, lawyers, teachers, and office workers seem today like prototypical liberal Democrats, while businessmen and hardhats seem like good representatives of the Republican party. The dividing points were different 50 years ago. The Republicans still have the support of most of the high-income voters, but these are conservatives of a different sort. As E. J. Dionne noted in analyzing poll data from 2004, the Democrats’ strength among well-educated voters is strongest among those with household incomes under $75,000—“the incomes of teachers, social workers, nurses, and skilled technicians, not of Hollywood stars, bestselling authors, or television producers, let alone corporate executives.”

But these comparisons are pretty crude: NES has only a couple thousand respondents for each election, and so we can’t really make any find distinctions.

But then through Dan Schiffer I learned of this attractive visualization of partisanship of occupations, as measured by campaign contributions. The number crunching and graphics were done by Verdant Labs, and they had the cute idea of lining up vaguely similar pairs of occupations that differed in their political donations. Here are a bunch of them:

“Sculptor” vs. “Plastic surgeon”: that’s cute. Just remember: the data represent campaign contributors, not voters.