Originally Posted by ItDontMeanAThing
The paper is crap. It's self-published (where 'self' is the academic group an U of Wisconsin the author is part of). As far as I'm concerned anything claiming to be SCIENCE that wasn't published in a peer reviewed journal isn't worth the paper it was written on.
The sample graphed is U of Wisc. graduates and their siblings, so it's not representative. I can't find anything about sample size or response rate (of those asked for info, what % responded) Other relations in the study between intelligence and occupation do have large and representative samples.
The IQ test used, the Henmon Nelson test, seems to exist on Google almost entirely in relation to this study or what it's mean and SD are. Furthermore, Mensa doesn't list it as one of the standardized IQ tests. Sounds like a U of Wisconsin in-house project.
The occupation categories are so broad to be meaningless, unless the intent was to show low correlation between intelligence and occupation. The definition of the occupation groups is not stated in the graphs, instead one is referred to an appendix. Seems like another in-house substitution where national standards exist. We'll never know, as the pdf contains no appendices.
A lot of papers are published as working papers or tech reports while they work their way through the journal review process (publicatoin process can take several years). They often end up as journal paper somewhere else and often with a different title -- not sure if this is the case for this paper but given that the lead author is a prof. of sociology at U. Wisconsin (fellow ASA, member NAS) I would tend to give him the benefit of the doubt.
The sample being from U. Wisc. is fine. All we are trying to do is get a sanity check on the ranges of what IQs are possible within a given profession. If you know of a larger study, please post it I'd be interested in seeing it. In the regression they are using n = about 4500.
I'd be more concerned about whether the admissions process to university (some of the data seems to be from the late 70s) is still similar now. For example, med school is highly competitive and it is a top choice for high academically achieving minorities such as asians. There may not have been this much competition 30 years ago.
I don't consider Mensa an authoritative source but a quick search on google shows that mensa does recognize it: American Mensa | Qualifying Test Scores