What Does IQ Measure? [video] | Hacker News Reader

What Does IQ Measure? [video]

youtube.com

6 pointstomlin18 comments

Can IQ be considered a scientific measurement, when a coffee or a bad sleep can affect the outcome? What do people here think?

I don't see why this should prevent it from being considered "scientific". These things limit how precise it can be but that doesn't really matter much. A few points don't matter.

Do the test when you have whatever is a normal amount of caffeine for you and after a normal nights sleep.

Ouch... The psychology is not science, though. It is kind of shamanism that is trying to be science. Raven progressive matrices are measuring your statistical biases in the first hand. The better you solve the patterns in this test, the more often you'll jump to conclusions on a base of insufficient data. And that is what they call an IQ. I have passed that test with a very high score, and I was surprised this test was as easy, and my results were as high. Then I had to think about this test for a while. The problem that I fight in myself (jumping to conclusions with little statistical evidence and insufficient data) for decades is regarded as high IQ. Bullshit.Disclaimer: I probably jumping to conclusion right now based on my own test, which is not a statistical evidence.

I cannot understand you position. You seem to assume that exactly one of these is true:

1) IQ is a measure of "true" intelligence

2) IQ completely fails as a measure and is meaningless

and since 1) is too nebulous to be true then 2) must be true.

I am not an expert on IQ, but if it was highly correlated with being able to recognize and guess patterns from incomplete data it would not be a surprise to me.

Even just the ability to formulate possible solutions to ambiguous problems could credibly have high correlation with IQ.

I am not saying that these are true, I am just wondering why they sound so impossible to you.

My point is: IQ tests fail to provide a measure of intelligence. The linked video at the end discusses some reasons why, such as environmental, educational, and cultural differences. The primary problem with these tests is that they are created by groups of people who have their own specific view on what intelligence is.

Yes, these tests assess certain traits of an individual, but intelligence is likely a more complex topic. Consider the language part of some tests, for example. These tests include vocabulary that the test subjects must know in order not to be discriminated against due to cultural and environmental factors. Instead, vocabulary capacity is measured. Language tests should include logical tasks of varying complexity, which must be solved without providing a variety of answers. If answers are provided, this will test for pattern recognition traits.

I agree that IQ is not a true measure for intelligence, but it does not need to be one to be a meaningful measure

What exactly for? All quantitative measures have a purpose. What good does IQ do as a quantitative measure if it can't measure what it stands for?

As I previously said You seem to assume that exactly one of these is true:

1) IQ is a measure of "true" intelligence

2) IQ completely fails as a measure and is meaningless

There are other things worth measuring.

What was the point of repeating yourself? You didn't specify what those measurable things were. My point is, the term "IQ" should not be used if the test is unable to measure what its name implies. That's about it. The cultural bias in IQ tests has been widely documented. However, I must admit the usefulness of these tests by the U.S. Army and Marine Corps in decreasing fatality rates. It might be better to rename these tests though.

One thing it apparently measures decently well is how good of a beginner you are on a new random thing.

Honestly I too would appreciate a different name because too many people[0] see it as "Smart Elite Classifier Test".

wrt cultural biases it is the same issue, you would need to develop a different test for each new cultural context. I agree that biased test have been used wrongly.

Overral IQ has a serious public image problem as some of its most vocal supporters are also its worst enemies.

[0] among both supporter and detractors

I don't get it. Is a tape measure invalid because height varies over the course of a day?

lol! Good one!!!!

Does a tape measure change?

The measurement does.

But the instrument for measuring doesn’t change. That’s what makes it dependable, predictable and something to measure against. You can’t measure against a moving target.

This is only evidence that we should not (tw: strawman argument) create a classist society where your lifetime status is algorithmically decided by a single test taken once at 18.

If you assume that IQ measures $SOMETHING this just mean that both sleep quality and caffeine influence $SOMETHING

IQ measures how well you can do an IQ Test.

It has no other usefulness. A bit like HN karma points: It looks useful, but signifies nothing.

how dare you

Reply on news.ycombinator.com