What's the straight dope on IQ and intelligence?

So in another thread (since removed), there was some discussion about IQ. Without wanting to tread too close to that mess, another poster, who seems to know what they’re talking about on the subject, said something along the lines of “IQ represents test-taking ability, not intelligence”.

This confuses me. I was under the impression that there was quite a lot of scientific literature on IQ, and that there was a general consensus among researchers studying intelligence that:

Where am I going wrong here?

I think what he means is that intelligence is intelligence, whereas IQ, as scored on IQ tests, is a measure of how good you are at taking tests.

Maybe someone who is very smart, very intelligent, has never taken an IQ test with pen and paper before in his life, never seen it, has a tendency to seize up in panic and fear in moments of pressure, and has no clue how to do it, and hence performs poorly. He’s not dumb, but he’s just not good at these tests. Maybe his True IQ is 130 but his test scores him as 90.

Then conversely, maybe you have a below-average intelligence person who’s taken such tests all his life and knows how to game them. His real IQ is 90, but due to being extremely well-practiced at test-taking, he tests with a score of 130.

It’s not all bunk. One’s obsession with IQ directly correlates with their high opinion of their own abilities, and willingness to insert it into random discussions.

What is intelligence?

Being able to remember a lot of facts?
Being able to solve novel problems?
Being able to recognize patterns in seemingly random things?
Being about to read people’s faces and body language?
Being able to turn a clever phrase?
Being able to master a musical instrument?
Being able to paint beautiful piece of art?
Being able to cook the most delicious turkey in the world? :slight_smile:
Etcetera, etcetera, etcetera.

Better to have it and not need it than to need it and not have it.

Obviously your score on an IQ test is a measure of how well you do on this particular test. What they don’t talk about is whether this ability correlates with intelligence.
You can say the same thing about the SATs, or any test you take at any grade level.

“Johnny is actually brilliant - he is just flunking out because he doesn’t take tests well.” Sure.

BTW each test question, properly viewed, is a problem, so taking tests well is a measure of how well you problem solve, which is possibly related to intelligence.

Reported. (petecook003)

NM

By definition IQ measures the ability to take IQ tests. As it happens the ability to take IQ tests is correlated with abilities in other areas, but the correlation is not one-to-one. For example with practice and training it is possible to greatly improve ones ability to take IQ tests, but without significantly improving their abilities in other areas. Also someone who is unfamiliar with the cultural norms and thinking styles that are assumed by the test may be very poor at taking the test but may have strong intellectual abilities. As an extreme example (spoilered due to NSFW at 20-25sec)

You mention that IQ tests correlate well with “G”, but G is itself more of a mathematical construct built up from the fact that the ability of various intellectual tasks are correlated. So saying IQ correlates with G is really just saying that IQ correlates a a number of intellectual tasks.

There is a whole field of psychometrics whose practitioners believe that we can identify and measure intelligence. The current standard measure of such intelligence is presented by the scores of a number of tests, the results of which are represented by an Intelligence Quotient. To their credit, they can point to a general correlation between higher IQs and success in either school activities or careers.

Unfortunately, when they try to move across cultures or languages, their results get a lot less consistent,* giving a number of us skeptics the impression that the tests are creating self-fulfilling prophecies. IQ tests rely on the ability of the test taker to be comfortable with sitting for a period of time, choosing from multiple possible written questions. Efforts have been made to create tests that do not require literacy, but it is not clear that such tests actually evaluate the same “intelligence” that is displayed on IQ tests and, again, while tests that are not literate dependant (such as the Raven’s Progressive Matrices) demonstrate somewhat similar results as IQ tests, there is not a direct correlation that would demonstrate that we have actually identified and quantified intelligence.

A frequently sought goal is identified as g, representing a cross referenced set of abilities in language and mathematics. However, g has not really been shown to indicate an actual matrix that includes multiple types of intelligence. For example, one competing theory of intelligence has been proposed by Howard Gardner who lists eight varieties of intelligence, musical-rhythmic, visual-spatial, verbal-linguistic, logical-mathematical, bodily-kinesthetic, interpersonal, intrapersonal, and naturalistic. Gardner’s speculations, however, have their own problems including unclear definitions and other issues.

The reality is that we are still a long way from defining exactly what intelligence is and using the results of our early attempts to creating an accurate definition to describe people are counter productive.

*Another aspect of the cross-cultural issue is that racists pick up (or manufacture) test results and misuse them to make claims about the societies of other places. This can range from the claim that “Africans” have an average IQ of 70, (which, if true, would indicate that the continent would have too few people with sufficient intelligence to fly airplanes, maintain waste water treatment plants, engage in manufacturing, or even manage retail operations), to the outright fraudulent claims made in books such as IQ and the Wealth of Nations in which nations were assigned IQ by averaging the supposed IQs of neighboring nations or by including tests given to fewer than 50 kids in a country.

I think “IQ represents test-taking ability, not intelligence” is a bit of an oversimplification … although it’s true that test-taking is a skill-set perhaps separate from the subject material at hand … as tomndebb says, the matter is far more complicated than that … it helps to be able to take the test as often as needed to get the score we want … military test are infamous for having one choice completely off-base, another being probably off-base, so just randomly selecting from the last two would still score you a 75% …

This is a good book on the topic.

Any defence of the IQ concept that relies on correlations with things associated with social position like lifespan and income is completely unconvincing - of course it correlates, it’s designed to. Not consciously designed, maybe, but the tests might as well be.

Look at the WAIS: “What day of the year is Independence Day” asks the first sample question. If that isn’t an absolute joke of an objective test of intelligence, I don’t know what is.

You ever play DnD or similar games? They usually have two separate stats of “wisdom” and “intelligence”, to reflect the existence of different types of “smarts”.

When people say that “IQ measures how well you do at IQ tests, it’s not exactly a measure of intelligence”, there’s two possible reasons for that. One is people who suffer from a bad case of Them Grapes Are Green. The other is people talking about

  • the fact that there are several types of “smarts”, of which IQ tests focus on one (logical analysis)
  • that even the best-designed of tests will have biases and be affected by factors beyond the designer’s own smarts. A bad translation can really screw up a test question, for example.
  • ok, so you come up with a test that’s all pictures, that will avoid language issues. Ah, but pictures are their own language. Betcherass someone who was navigating for Dad at age 8 and had 7 years of draftsmanship will do better in the same test than their separated-at-birth twin who didn’t get the training.
  • plus, since language intelligence is a thing, going for only pictures means that you’re not considering language intelligence.

That WAIS test? I just found a version online and it didn’t ask me about Independence Day but it did ask me to do basic arithmetic in the form of long-word problems in English. That particular version measures:

  1. the test-taker’s English,
  2. their ability translating worded problems into solvable equations
  3. the ability to call it x
  4. and basic arithmetic

2 and 3 are two things many of my university students had problems with; in fact and despite in theory tutoring Chemistry, I’d say over 90% of the questions I got boiled down to those two. They were not stupid, but nobody had taught them 2 and if anybody had explained 3 they hadn’t explained it very well.

It begs the question: what is the purpose of these tests?

If we look a they are used for, it is often part of an admission process to institutions and thereafter jobs that have high economic and social value. Academic institutions favour skills that are a very narrow definition of intelligence. They concentrate on the traditional Three R’s - reading, 'riting and 'rithmetic. For that they owe much to classical education that goes back to the traditions of Ancient Greece.

Students of our foremost academic institutions are famously unprepared for life outside the delivery of essays in a timely fashion. What skills they learn in analysis and criticism is countered by what they lack in other important sklls. Fighting a war, caring for children, building a home, fixing an engine, growing food, having a relationship, dealing with a bully. There are so many skills that could be associated with intelligence. Why pick such as small subset?

I would suggest IQ tests are, like many other tests. Narrow in focus and constructed in an arbitrary manner. Moreover, the structure of the tests are simplistic paper and pen exercises that can be gamed by repetition and practice.

Why?

I would suggest that the reason is economic. They serve as filter to ration access to expensive academic institutions and the endorsements that lead to - big money jobs, power and influence. In their defence, they are at least provide some consistency. On the other hand they are also something of fig-leaf. They disguise the fact that those with the right training know how to game the system. Those from influential families, who have friends in high places get prefferential access to the pathways to power and wealth. The IQ test is pseudo science, they are convenient for institutions as part of their admissions process because they give the appearance of being fair and rational, but in fact they do not bear up to examination. An academic institution that based its science by the standards of IQ test would have little credibility.

Intelligence is a very broad subject and definitions are often ambiguous and subjective. Little wonder that tests that purport to measure intelligence are themselves as questionable as their underlying assumptions.

All tests are based on an assumption that they are a meaningful and useful measurement of some ability. Some are well constructed and there is a good correlation with that ability, if it can be exactly defined. There are problems with many tests, but intelligence tests are the ‘joker in the pack’ because how can you measure something that cannot be defined?

There is a great deal of pseudo scientific nonesense concerned with measuring human beings and it is at its worst in disciplines like psychology, that make grand claims based on questionable statistical evidence. IQ test arose out of questions regarding the lowest tier of a human ability and attempts to distinguish between the rather unenlighted medical categories of the 19th century that sought to distinguish between Imbeciles, Cretins and Morons. Like Darwinism, it was a idea that was hijacked for political purposes. As the professional and technical class grew in industrialised economies, it needed a way to control access to education and training. The IQ test worked well to to keep out the unwashed and unconnected whilst pretending to be a meritocratic measurement.

The idea of measuring intelligence is an interesting one, it deserves some proper research than all this Victorian IQ test nonesense that carries so much dubious baggage.:dubious:

That’s both backwards and culturally limited.

Results from certain tests which are used in admissions processes and which do not in fact give an IQ value as part of their results, have at different times and been accepted as proof of high IQ. That’s true, but what you said is backwards from it.

Also, those tests and whether they are or are not considered to correlate with high IQ are country-specific. I can assure you nobody has ever claimed that a high result in Selectividad or a low number in the MIR exam indicate high IQ, but the first was used to filter for university access for over 40 years and the second has been used to filter and prioritize access to medical residence programs for more than 50.

If by “institutions,” do you mean colleges and universities? I’m unaware of any that include this as part of the admissions process. Few schools give students IQ tests these days because they’re of so little value, so most applicants wouldn’t have that information.

I agree with most of your post, but this gave me pause.

IQ is an exceptional measure of general intelligence.

This is why the people with higher IQ’s perform more intellectually complex work and more economically successful.

Show me a better predictor of one’s economic success in life. Until you have something even close to the correlation with IQ, you dismiss it out of sheer ideological distaste.

Fight ignorance, even if it makes you uncomfortable.

Just to correct one common misconception: You don’t take an IQ test by sitting down with a pencil and paper and filling in multiple-choice bubbles. An IQ test might have one portion that’s administered that way, but it’ll also have other portions that are administered in other ways, and the whole thing needs to be administered by a trained psychometrician.

Well what would we like to refer to as “intelligence”? If IQ tests correlate well with factors such as “ability to concentrate on tasks for long periods of time” and “ability to solve complex problems” and “ability to hold down intellectually taxing jobs”… Well, whatever it is we’re measuring, it clearly means something, right? We may not have a perfect grasp on what “intelligence” is, but whatever an IQ test is measuring, it seems to have some value.

This makes a lot of sense. So does the other part - 70 IQ is generally considered borderline handicapped; the idea that that could be the average for an entire continent is pretty obviously wrong to anyone whose understanding of Africa has gotten past the level of “Capcom designer”.

Hey, I resent that. I’m obsessed with my IQ, and the only people who know about it are literally everyone who ever read all the way through my profile on any dating app, so a grand total of fucking nobody. :smiley:

Well… Yeah. That seems pretty relevant, wouldn’t you think? If the point of an IQ test is to test your intellectual aptitude, then your ability to perform intellectual tasks would be something you should very well expect to correlate pretty well.