"IQ measures how good you are at taking IQ tests"

Great Debates
1

In the interminable debates about various intelligence-related topics that constantly take place on the SDMB, a frequently-expressed opinion is something along the lines of “IQ measures how good you are at taking IQ tests”, or something along those lines. IE, it has no real meaning outside itself. I am very skeptical of this.

So here’s a thought experiment: Bob and Jane are both starting small companies. They are going to design and manufacture widgets. They need 50-person working forces which will include inventors, engineers, technicians, salespeople, marketing people, webmasters, receptionists, etc.

They have pool of one million people who live in their region who are all eager to work at these companies, but all of whom have exactly zero training in all of the above fields. In fact, let’s say they’re all recent high school graduates. So each of those one million people is given an IQ test. Bob is allowed to interview and hire whoever he wants among people with IQs between 95 and 105. Jane is allowed to interview and hire whoever she wants among people with IQs between 135 and 145. Then the 50 people they have each chosen to hire are given the necessary training (including years of college and grad school for the inventors and engineers… heck, this is a thought experiment), and then Bob and Jane start their companies and see how successful they are.
I believe that in this thought experiment, Jane’s company is likely to do far better than Bob’s.
Does anyone disagree?

2

I don’t know. I’m by no means a genius, but I’m intelligent enough that if I apply myself I can do pretty much any task I set out to do. I say “pretty much” because I’ve got a learning disability with regards to spatial relations. I was tested extensively at age 14 by some neurologists and other experts, and while I scored well on verbal/math/etc. tests, I came out retarded (their term) when it came to spatial tests. So if I take a standardized IQ test, that’s going to drag my numbers down. Sure, if I was looking for a job that was mechanical in nature it would be an issue, but if I was applying to be an editor or a marketing executive or something I’d be perfectly suited to it, and possibly better at it than someone with a higher IQ. I guess what it boils down to is what most people object to in the first place: there are different types of intelligence, and the tests don’t necessarily take that into account.

3

Yeah, but there’s a big difference between saying “a single number can not possibly fully represent the full spectrum of human intellectual capabilities, so we should not assume that IQ is the be-all and end-all of intelligence classification, nor should we assume that someone with a higher IQ (even assuming all testing was done accurately and in a magically unbiased fashion) is automatically better at any or all brain-related tasks than someone with a lower IQ”, which is pretty inarguably true; and saying that IQ means nothing.

4

Jane’s company is likely to do better, yes. The thing is that such tests can be taught to some extent, and to some extent a person who is good at such tests might only be good at taking tests – particularly if he’s someone who was simply taught how to take them. But, in the modern day US, there isn’t a terribly lot of “teaching to the test”. Or, at least there wasn’t before No Child Left Behind. So, in general, it is going to be a fairly decent measure of intelligence.

5

It’s not that IQ tests don’t measure anything, but that what they measure is a really narrow slice of human mental capability.

For example, if I’m trying to hire a salesman for my imaginary company I’m going to care a lot more about his people skills than his ability to solve word problems. Can he meet a group of strangers and immediately have them like and trust him? That sort of thing is really important in a sales position, but is completely not reflected in an IQ test.

Sure, all things being equal, you’d rather hiring someone with a high IQ than a low IQ for almost any position. But, all things being equal, you’d rather someone who is a friendly extrovert over a sullen introvert. Or hire someone who is strongly self-motivated over someone who is passive and lazy.

6

Jane’s company may have a slight edge, but in sacrcely matters since your thought experiment is so highly unrealistic. I get the impression that back in the 20’s through about the 50’s, when dinosaurs roamed the earth, IQ tests were frequently used by employers, as well as schools, the military, and anyone else who wanted to measure anyone’s intelligence. Time went on and people started to notice that a person’s IQ score didn’t have much relationship to his or her performance on real-world tasks. More over, IQ score wasn’t even consistent in any given individual. Give a person multiple IQ tests and he or she would get widely varying scores. So, consequently, we’ve now see very few employers bothering with IQ tests, and I’d wager that my experience of never taking an IQ test is common to most of my generation. Or put another way, Alfred Binet, the French psychologist who invented the first IQ test, was right when he said that test results should not be interpreted as numerical measures of intelligence. It just took us about a century to listen to him.

Of course the one place where IQ tests still are used is in academic labs whenever somebody is trying to prove a result about intelligence being correlated with something or other. I can only assume that this is because the academic world, as we all know, lags decades behind the real world in terms of accepting reality.

7

You’d have to consider boredom and retention as an issue if you gave every job in the factory to people in that IQ range. Some jobs require a lot more thought and creativity than others.

Wiki article on IQ covers predicted job performance:

“for hiring employees without previous experience in the job the most valid predictor of future performance is general mental ability.”

Otara

8

Not I, mine (3 tests) are within 6 points of each other. I’m also apparently very good at taking those tests… yet I probably would end up completely destroying any company that was stupid enough to hire me. Frankly, I wouldn’t hire me… and I’ve hired some really shady people… probably because I hire people who are best for the job regardless of IQ, formal education… or understanding of basic personal hygiene.

9

I think few would seriously argue that IQ tests performance has no correlation with success in any other activity. Or that two or three S.D.s difference on that test is not significant of some aspect of intellectual functioning.

And assuming that other factors are equally distributed and then the high IQ test group would likely produce better engineers and possibly better designers (not likely on the artistic design side though) … not likely much better line workers.

But let us change the thought experiment some: Bob and Jane both understand the skill sets they need and can read people and their skill sets well. Bob is able to hire 100 of whoever he wants of the group that is IQ of 95 to 105 taking any other factors he wants into account. Jane must hire the top 100 scorers of IQ 135 to 145 and is not allowed to take any other factors into account. Is IQ such a strong predictor of broad functioning that Jane’s company is likely to do better? Or is it little enough of a factor that Bob’s being able to select for other prospective employee attributes will more than compensate for the far superior IQ of Jane’s employees?

10

How would you know Bob can really do this unless you’ve already empirically verified it?

The literature suggest we generally tend to overestimate our ability to judge on other factors. Otherwise IQ wouldnt be the best predictor in this kind of scenario where theres no previous job experience, some other factor would be.

Otara

11

IQ is at best a very roundabout way of measuring aptitude. You can’t directly measure, for example, someone’s spatial ability. You could administer a test measuring how long it takes the subject to arrange a bunch of irregularly shaped blocks into a particular pattern, and that might give you a clue about the subject’s spatial ability (which is in turn a clue about the subjects aptitude for tasks like widget-making). Unfortunately, such a test is time-consuming and expensive to administer on a large scale.

Most people who talk about IQ tests in the general sense refer to written tests which are much cheaper and easier to administer, but are restricted by their medium (“How many triangles do you see in this diagram?”). So they only directly measure your ability to perform well at answering written questions that are thought to correlate with hypothetical abstract abilities that should translate into real-world aptitude.

Now, a properly designed written test can be helpful, but you can’t summarize intelligence as a single number. In one of my intro psych classes they said that it’s more useful to divide intelligence into about a dozen subcategories, each with three or four subcategories of its own.

12

IQ and creativity only correlate up to 120 or so. Also it seems to be linked with madness. You’d want creative problems solvers yea?

If so Jane’s advantage just got knocked down to 120

Also I’m not convinced IQ test scores are so inflexible.

13

Actually I take my last post back.

Reading my cite a little farther:

Suddenly this seems a lot more interesting.

14

Well, IQ tests by definition measure how well you take IQ tests. The question is whether ability to take IQ tests well is correlated with other things. It clearly is at the extremes, and it clearly isn’t precise enough to measure tiny differences.

BTW I agree that Jane’s company will do better - so long as she selects the best talent she needs from the pool she has. High IQ doesn’t mean one can do anything well, so if she chooses randomly she could still lose.

15

In relation to the inventors, engineers, technicians etc, yes - Jane’s group will do much better.

16

Not line workers the way they are used in most factories. However smart line workers can look for trends in defective material they get and look for ways of improving the flow - and they can often do better than the smart process engineers because they are closer to the problem. That is what giving them the power to stop the line is all about. But I do agree that it is probably not the best use of a high IQ person, though I’d higher the smartest line applicant who I find - and try to take advantage of, not ignore, suggestions.

17

Right, but Jane, while interviewing her pool of candidates for someone to hire as a salesperson, is going to try to find someone with people skills. As will Bob, while interviewing his pool of candidates. So they’ll both end up hiring people with people skills. And Jane’s will have a much higher IQ. Will that be relevant to a job like salesperson, one which does not necessarily immediately involve solving puzzles and doing other “IQ-test-like” things?

My suspicion is that it will still be relevant, and that if both sales people have equal people skills, the one with the higher IQ will, all other things being equal, be a better salesperson. But I have nothing other than a hunch to support that.

18

It’s very easy to disagree with basic proposition.

Let’s assume that IQ test scores are good predictors of academic achievement. There is some evidence to show this. The problem states that all of the employees will receive the necessary training, including schooling. So Jane’s employees may have higher grades in the education phase, but that does not have a strong correlation to productivity.

The overall success of the companies is most likely tied to management and marketing, and may be more dependent on ‘people skills’ that are not reflected in IQ tests.

Let’s assume that there is some correspondence between IQ scores and intelligence. Only a few of the employees in the companies would be working in areas related to areas requiring high intelligence. Most would be running machines, doing paperwork, selling, driving trucks, and other mundane tasks that could easily bore the more intelligent employees. Jane may end up with a crew of grumbling dissatisfied employees who feel they are working at jobs below their status. These people may never even make it through the educational process for a machine operator or truck driver, leaving Jane with a shortage of crucial employees.

Even the tasks of engineering and inventing may be better performed by people who are motivated to meet the practical needs of the company, limited by time and cost, while the higher performers may waste time and effort on more scientifically significant problems, that do not have practical application. Higher IQs may also correlate to NIH a productivity killer in many circumstances.

If you look at investors in small companies, it is likely that Bob and Janes experience and their ability to demonstrate their organizational and decision making skills. These factors would not correlate at all to the IQ of their employees. Jane’s smarter group may not have the capital to operate if she herself is lacking in those areas.

The problem also states that the IQ tests are issued just after high school graduation. In the real world, the correspondence between high IQ and college admissions reinforces an inherent bias by stopping the educational process for low scorers. But in the case cited, all of the people selected will get the education required. The correlation between IQ and educational achievement may disappear in those circumstances.

Back to the start of the OP, if Jane and Bob were not widget manufacturers, but instead in the business of hiring people with high IQs, Jane would have the definite advantage. The further the actual business diverges from the results of IQ tests, the less influence those test scores will have. That is why it so often said that IQ tests measure the ability to take IQ tests, because that is only stromg correlation that can be proven.

19

I don’t think Jane’s company would necessarily have much of an advantage. Work ethic is far more important than IQ. I happen to have a high IQ but I have a crap work ethic. I’m also short on common sense and am easily bored. My ability to whizz through a verbal reasoning test has no bearing on whether I have the drive or the wherewithall to slog through the 1001 humdrum chores which comprise the bulk of the typical day at the office. If Jane hired me solely on the basis of my IQ, she’d have to deal with my tardiness, my daydreaming, and occasions (like today) when I fake a stomach bug so I can spend the day lounging around in my boxer shorts drinking Dr. Pepper, watching 30 Rock and playing my guitar.

At the other end of the spectrum, if she’d hired (my about average IQ) housemate, she’d be taking on someone who (to the best of my knowledge) hasn’t had a sick day in about three years and is happy to take his work home with him. Outside of this hypothetical, guess which of us makes more money and has the better employment history?

20

You’re rather ‘rigged’ the experiment by choosing such a wide gap between the IQ scores of permitted candidates and, at the same time, specifying a workforce consisting of several jobs that are more normally occupied by people with higher IQ’s.