« Tradable vs. nontradable jobs | Main | Wordsum vs. the Bible, education held constant »

July 23, 2011


If the data showed the reverse, HS would be saying that obviously verbal ability leads to lucrative "value transference" skills. Behold, a man who can explain any data equally well!

Did you control for gender? It could be that women have higher wordsums and lower incomes, since they often choose teaching or other less lucrative career tracks.

"we are looking only at respondents whose highest educational credential is a bachelor’s degree."

English Lit BA's don't pay. High paying verbal jobs like lawyer require post grad.

Is it possible the correlation becomes distorted above (or even by) a certain level of education? I would expect someone with an English/Lit degree to have a larger vocabulary (at least WRT the type of question in the wordsum test) than an engineer or scientist; I would not expect them to be more intelligent. Also, are there general distortions in the Wordsum test that could account for this? For example, IIRC, elderly women with average IQs and low levels of education tend to have very high Wordsum scores. Also, the research giving Wordsum a .71 correlation is over 30 years old. Is it still valid given the changes in education since then?

Would you agree that higher verbal ability is needed for value-transferring? Or maybe only a certain level of verbal-ability after which you are too nerdy to persuade people into transactions that benefit only you?

I used to think of high math ability vs. high verbal as a rough approximation of value creating vs. value transference, but this is evidence against that view.

Wordsum has a low ceiling.

So, if true, this measn that the correlation between IQ and income is mediated entirely by level of formal education.

As I've said before the effectiveness of this mediation is far less in the US and Canada than it is everywhere else.

This is pretty easily explained in that higher-IQ people who reject formal education are by and large relegated to lower-paying jobs/careers because playing the credentials game is beneath their intelligence.

I would venture that almost all of these people are from middle or lower class backgrounds, for whom the costs of acquiring degrees is much more problematic than for a lower-IQ child of wealthy parents.

Also, this likely includes people who are willing to take jobs doing what they want to do rather than what pays them the most money.

Who says that superior verbal abilities lead to superior job performance?

[HS: Schmidt and Hunter say so.]

Maybe this has something to do with the fact that employers value mathematical ability over verbal ability and people with higher verbal ability may have lower mathematical ability and vice versa.

A better explanation would probably be that it isn't terribly logical to waste your life working or striving once basic necessities are obtained. Why waste energy striving for that $70,000 position if you live relatively comfortably at $40,000?

HS, your explanation would seem to explain why a high IQ person with only a high school or associate degree makes less money than a lower IQ college grad.

Intelligence is a liability in a lot of situations. Successful politicians, with a few exceptions, seem to have an IQ of about 120. For most jobs, dealing with most people, you need to be smart enough to do certain things, but not so smart that you can't relate to average people. On top of this people of average intelligence often hate intelligent people.

HS Your explanation fails to explain why people with a lower wordsum have a higher income. You would have to go further and say that employers not only don't value the supposed higher performance but actually find it detrimental. There must be something detrimental about the people with a higher wordsum (or something beneficial to lower worsum) which causes them not too keep up with their lower wordsum peers. Perhaps smarter people don't always have superior performance as you assume, perhaps their intelligence is a detriment in some way. Or perhaps people who talk like the average person have a social benefit when it comes to receiving promotions. In many fields that only require a bachelors degree social intelligence and relatability to customers and managers is more important than straight intelligence.

[HS: General intelligence has been shown over and over again to have a good correlation with all aspects of job performance. There have been many academic research papers demonstrating this. Some guy by the name of Hunter has done a lot of research in this area. But doing a better job apparently doesn't correlate with better careers.]

Truly low IQ people can't get into college, so perhaps this is selecting for the sub-set of people with high-IQ, but mostly due to their mathematical ability. You can do pretty well with in engineering, accounting, finance and computer science with just a bachelor's and you don't need high verbal ability to do any of those.

You guys might now this. If I needed a go to link that discussed how much of the IQ/g gap could be attributed to genetics versus environment where would I go?

I’ll assume (despite my doubts) that the GSS data is representative. However, even assuming that I think Sigma has reached the wrong conclusion that higher verbal ability leads to lower incomes. Firstly if people get hired on the basis of their academic degree alone as suggested and then get into an automatic track in their careers irrespective of their ability – the ability to get into a career path (interviews, written assessments, project management skills etc etc) is positively correlated with higher verbal ability (along with other ‘intelligent’ qualities). The question becomes why do higher wordsum scores do no not reflect the income data.
I may be incorrect, but my speculative hypothesis is that the anal ability to get a perfect 9 or 10 on a wordsum score is more likely to be distributed among people with higher intelligence but also with a less social personality – which may lead to general propensity not to take supervisory/management positions. I can’t quite find a link but I think there are studies out there showing that beyond a point higher IQ is positively correlated to qualities that maybe defined as socially awkward – the ‘nerd syndrome’.

For e.g. – As I have previously commented my experience has been that in general PhD hires turn out to be bad managers but more importantly – they seem to be unhappy in management positions.

i.e. People with a perfect wordsum score may be less good at middle\upper management work (generalists skills) in less intelligence intensive professions - like HR, office administration etc.

Actually equating verbal facility with professional ability and career track selection is highly biased and unreliable, since we all know that professors, writers, librarians, historians and think tank research fellows will have higher word-sum scores, yet have lower income than engineers, doctors and/or lawyers.

It would make more sense to organize the scores based on occupation and see that the higher scorers tend to dominate in their field against lower scorers in the same field.

I would have to agree that those with super high IQ over 150 tend to be socially awkward and it hinders their ability to self-promote and advance in their career. Also in many cases high IQ individuals don't place a high importance on income.

This is interesting. I'd think a Bachelor’s degree is high enough that you aren't simply sorting for lazy, under-achievers and mal-contents. That is to say, a bachelor's degree is enough of a credential for most people (at least the 6, 7, & 8-ers) to attain the full income potential of their IQ. After that, it's not IQ, but conscientiousness and the other 5 personality factors that become important.

On the other hand, a case can be made that the WS 10's with only a bachelor’s are either lazy under-achievers or can't get along jumping through life's hoops. How do 10's do at the Masters level vs. PhD?

An Aug 2009 issue of the Economist highlights how taking more maths in school can make you richer and how reading classes has a negative or no effect on earnings.It varies with race,though.

[HS: That article is about blacks and what courses they take in middle school and high school, which I don't think is relevant to a group of people of all races with bachelor's degree.]]

So maybe high IQ people who don't want to go to university should aim to be self-employed?

Employers do not want to hire people who they perceive as being smarter than they are. If you have a very high verbal fluency, you will intimidate a lot of people and will unlikely get hired.

"Employers do not want to hire people who they perceive as being smarter than they are. If you have a very high verbal fluency, you will intimidate a lot of people and will unlikely get hired."

I recently had an interview where I think this was the case. I've noticed it a lot more with female managers. I was speaking with an older women about it and she said that many women feel threatened by a male fast tracker directly under them because they might take their job. Apparently it's more common with older women who dealt with the glass ceiling in earlier decades.

Huh? That makes no sense. HS is implicitly saying that higher intelligence means less prestigious degrees. The whole post relies on that fallacy – not sure why no one has picked that up.

I have a high IQ and excellent people skills. I have performed very well in management and have been asked to take higher leadership positions with higher I come. I have turned them down because I much prefer research.
Robert Hume

It was once believed that civil engineering companies preferred straight B students to straight A students because the employers believed (rightly or wrongly) that straight A students were nerds who could not effectively interact with the companies' clients and staff.

In fact, a survey of civil engineering employers in the early 1970s showed that they cared only that the potential employee have a BS degree from an ABET-accredited school. They did not care what school (MIT = Akron State), what discipline (BSCE = any BS engineering discipline) or what grade average (C = A). Their opinion was that the BS Eng'g degree was a sort of psychological test that indicated a significant level of intelligence and good work habits. They universally agreed that they would teach the new hires what they had to know.

Very consistent with the surveys of Thomas J. Stanley.

The average millionaire is smart, but aware that he is not the smartest and is a striver.

I am waiting patiently for some posts about global warming...

"A better explanation would probably be that it isn't terribly logical to waste your life working or striving once basic necessities are obtained. Why waste energy striving for that $70,000 position if you live relatively comfortably at $40,000?"

Because the $70,000 position is higher in the company and therefore carries with it more prestige. Additionally, the extra money could be saved and put into the children's college fund, which would also prepare them for similar positions.

"For e.g."

For exempli gratia = for for example.

Half Sigma, I'd advise adding in people with professional degrees and phDs.

What about the age of these people? Some of the younger ones may not have yet reached their full earning potential. At age 35 - 40, you could still be making a crappy post-doc's salary or something. Does the trend hold for those 50+?

OK, what I said above wouldn't apply to those with only a Bachelor's and in non-academic positions obviously...

Actually, I think this set of people might be selecting for the smart but maladjusted, or the smart but those who go into unstable careers. If you have a bachelor's in the sciences, there isn't much you can do without a PhD. Same with some other degrees where you really need a PhD to do higher level work (history, anthropology).On the other hand, if you're a smart artist, let's say, you may not have a guaranteed income above a certain level but, presumably, you're OK with that. So, the people who have a bachelor's but are less smart might be going into careers that are more pedestrian but actually better compensated. Personality factors might be playing a big role here.

Jay, you are obviously correct that 'the $70,000 position carries more prestige', but at what cost? To rise in business usually takes long hours and the studying of very boring subject matter in your spare time. These same hours can be spent studying more interesting and difficult topics that often have little monetary reward. Take for instance political ideology (why we are all at this site): there's no real money in it but there is some prestige in being able to develop and argue a personal political philosophy. Smarter people tend to enjoy the prestige of being able to effectively argue difficult subject matter more than the average Joe. But this takes time. Something that becomes more and more limited if it's all devoted to rising in the rat race.

'the extra money can be put in children's college fund'. Again, the time spent to rise in business can likewise be spent studying the overall contours of how our education system works. Look at Obama. His mother was relatively poor but she probably knew far better than most how important a prestigious education was to getting a position of real power in today's world and how to get her son that education. It's doubtful that if she would have devoted her intelligence to rising in business, she would have learned as deeply about the mechanics of our education system and how to game it as I'm assuming she did.

"For exempli gratia = for for example."

An example of verbal ability that will not ever translate to higher earnings ;-)

> An example of verbal ability that will not ever translate to higher earnings ;-)

Wrong. There is no way someone who can point out that tautology is poor. It is a reflection of IQ, attention to detail, and a curious nature. Jay M, can you share with us your income?

That said, I'm not sure Jay is entirely correct, if you take it literally.
exempli = example gratia = free
"for e.g." could be read for "for free example".
I never took Latin

My Latin is rusty, but "exempli gratia" literally corresponds to "for the sake of example". It's a standard Latin construction composed of gratia paired with a noun in genitive case.

I think you just cherry picking (data mining) to get whatever result you want. With a test as unreliable as wordsum and a statistic as unreliable as self reported income, all kinds of strange anomalies are possible. Is this negative correlation between wordsum and income even statistically significant? Does it only appear when you control for education? Does it appear at all levels of education or just among BAs? Is it confounded by social class? (dumb rich kids getting a free ride through school and then life?) You claim Dale and Krueger found the same result but you provided few details. You also dismissed better studies showing the opposite conclusion because they were not specific enough or are too old (terman study) so is this just a recent phenomenon? How recent?

I think gratia is the ablative of gratia so its the Latin equivalent of por favor.

Exempli is the genitive of exemplum.

So exempli gratia means what?

But Jay M is right.

2nd Try has made another retarded comment. I guess he grew up poor or his gradparents were poor or he's ony twelve.

"So exempli gratia means what?"

Literally "for the sake of example".

Just for example would be exemplo.

If "pro exempli gratia" is acceptable Latin then so is "for e.g.", but it's not standard usage.

A chart on this blog shows a VERY high correlation between income and wordsum among people with graduate degrees. The author downplays it but if you look at the slope on his graph, it's very steep, suggesting IQ has the most economic value at the highest levels of education:


[HS: The chart I looked at shows that as wordsum increased from 7 to 10, there was barely any increase in average income.

There was a drop in income for people with wordsum 6, but you have to wonder how someone with a wordsum 6 (IQ less than 100) managed to get a graduate degree in the first place!]

The graph shows that among folks with grad degrees:

IQ 111 $45,000

IQ 116 $57,000

IQ 123 $61,000


Half sigma, there are two charts on the page. I'm referring to the bottom one, which converted the word sum scores to IQ equivalents.

Looking at the second chart I linked to, it seems that high IQ people get richer within every educational category except perhaps high school grads and BAs. This is because people who get only high school or only a BA are probably only getting those credentials to please their parents, and their future earnings are largely determined by family connections or luck because they have no real goals.

By contrast people who go to junior college or graduate school are driven self-made people with specific career goals and their ability to reach those goals is a function of IQ, thus the strong correlation between IQ and money in these categories.

I agree with half sigma about career tracks being important, but surely IQ predicts how far and fast you can progress within your career track. Otherwise why do we care about IQ if it only measures book smarts, and has no real world practical value beyond getting credentials

I agree with half sigma

This is just one more reason why going to grad school is such a bad idea. It prevents you from getting on a career track for years, putting you well behind your peers. It's a trap for just the kind of person who would score well on a vocabulary test.

Here are a hundred more reasons:

It is another way in which we channel relatively high-IQ people into non-productive occupations, but grad school/academia is a much less rewarding route for individuals than higher-paying government jobs.

[HS: You missed the previous post in this series showing a clear income advantage for respondents with graduate degrees.]

"you have to wonder how someone with a wordsum 6 (IQ less than 100) managed to get a graduate degree in the first place!"

Actually, I think this issue has been discussed here before:

1. Which academic field of study produces more graduate degrees than any other?

2. Which academic field of study has the lowest SAT-derived IQ for entering freshmen?

Hint: The answer to both questions is the same.

The comments to this entry are closed.