I've seen this chart before http://www.ssc.wisc.edu/cde/cdewp/98-07.pdf), but if you examine the data a little more closely, you'll find that the stats are not to be taken too seriously, for the following reasons:
1) Sample size: data are collected for sample sizes of "30 cases or more" for each occupation; hmmm, that really seems like a broad and effective sampling to me...not.
2) Sample population: "Based on [male/female] Wisconsin [University] graduates and siblings who reported a job during 1992-94". Yeah, we all know that Wisconsin is a hub of science, engineering and technology.... right.... maybe that's why MDs are the smartest ones there... because it's Wisconsin for godssake! I'm pretty sure the engineers who designed the Mars rover have an IQ higher than 130.
1) Sample size: data are collected for sample sizes of "30 cases or more" for each occupation; hmmm, that really seems like a broad and effective sampling to me...not.
2) Sample population: "Based on [male/female] Wisconsin [University] graduates and siblings who reported a job during 1992-94". Yeah, we all know that Wisconsin is a hub of science, engineering and technology.... right.... maybe that's why MDs are the smartest ones there... because it's Wisconsin for godssake! I'm pretty sure the engineers who designed the Mars rover have an IQ higher than 130.