Study: IQ test results are higher among those who really want high scores | Hacker News Reader

Study: IQ test results are higher among those who really want high scores

pnas.org

40 pointscwan29 comments

Is it particularly surprising? I've never looked into any research on the subject, but it would seem fairly common sense to me that in a huge number of tests, ranging from IQ tests to History exams at school, people who care will try harder and therefore do better.

Many psychological findings are remarkably un-surprising in retrospect. And I agree with you, this seems stupidly obvious once I know the answer. But I took an undergraduate course basically dedicated to the history of IQ testing and this gem never came up. A lot of things seem obvious if you know the answer. But I guarantee you, this never occurred to many, many people who have devoted their lives to studying IQ. Also, a lot of things that seem fairly common sense turn out to be wrong when you subject them to empirical testing, so be careful about the applying the 'common sense' filter.

Many people think that IQ tests will measure accurately even if the participant doesn't try hard, kind of like how psych tests work.

It's not true, though. The questions are -hard- and if you don't try, you won't get the right answer and you'll do worse.

So while it shouldn't be a surprise to anyone who's taken one, not everyone understands how the test works.

> A lot of things seem obvious if you know the answer. [...] Also, a lot of things that seem fairly common sense turn out to be wrong when you subject them to empirical testing, so be careful about the applying the 'common sense' filter.

A Yahoo researcher wrote a book about this very thing:

http://www.amazon.com/Everything-Obvious-Once-Know-Answer/dp...

The first part of their finding, that motivated people or those with an incentive will do better on IQ tests, does seem common sense.

The second part of their finding is more interesting to me, although I'm still trying to fully understand it. I think they are suggesting that innate heritable intelligence isn't as predictive of life outcomes as previously thought because part of what was being measured was motivation to do well on IQ tests.

In other words, two people could be born with the same innate intelligence, but one of the two could be more motivated and do better in both IQ tests and life outcomes. Previously people would've attributed the difference in life outcomes to a difference in innate intelligence. It still seems like IQ would be an interesting number to look at, but it isn't quite measuring what people believed it was measuring.

  Previously people would've attributed the difference in life outcomes to a difference in innate intelligence

What? I mean, sure there are some people who think that, but surely clever people realise it's not the case? Even as a ~14 year old at school I could tell just from my class mates that some hadn't been born amazingly bright but worked their asses off and would end up doing better than the lazy people who found stuff easy instead.

Surprising today, or 100 years ago when people thought IQ tests measured some immutable property of the human brain :) today of course we know it's possible to change your IQ in all sorts of ways.

Apart from underfeeding children or hitting people on the head, I'm not aware of any ways to significantly change IQ. Could you elaborate?

Spatial navigation and logical reasoning improved by higher levels of testosterone, for example.

I don't understand - injecting a person with testosterone increases their logical reasoning ability? I hate to be that guy, but [citation needed].

http://books.google.com/books?id=IHGdmrKlVl0C&pg=PA57...

http://www.ncbi.nlm.nih.gov/pubmed/3067252

This relates to women, but this applicable to mens too:

http://en.wikipedia.org/wiki/Hypogonadism#In_men

   Difficulty concentrating
   Memory Loss-difficulty in choosing words in language

How about a caffeinated coffee? Or a poor sleep?

Indeed.

http://cscs.umich.edu/~crshalizi/weblog/520.html

http://scholar.google.com/scholar?as_q=raise+IQ&as_epq=&...

I don't have academic access to the actual article. A commonsensical result may not be surprising but I want to know how they did their metastudy. Anyone care to elaborate?

http://lesswrong.com/lw/im/hindsight_devalues_science/

I think 50 years from now we will look back on IQ tests as a terrible way of measuring intelligence.

There are quite a few who feel that way today.

it's one of those things that's a terrible option that's better than any other option that we can think of

Well, there's Howard Gardner's theory of multiple intelligences, however it doesn't really give you a single number the way IQ does (how could it!). I'm reminded somewhat of the scene on the spaceship to Battle School from Ender's Shadow.

There is next to no evidence in favour of the multiple intelligences theory.

Multiple Intelligences, the Mozart Effect, and Emotional Intelligence: A Critical Review

* This article reviews evidence for multiple intelligences theory, the Mozart effect theory, and emotional intelligence theory and argues that despite their wide currency in education these theories lack adequate empirical support and should not be the basis for educational practice. Each theory is compared to theory counterparts in cognitive psychology and cognitive neuroscience that have better empirical support. The article considers possible reasons for the appeal of these 3 theories and concludes with a brief rationale for examining theories of cognition in the light of cognitive neuroscience research findings*

PDF link

http://lynnwaterhouse.intrasun.tcnj.edu/Multiple%20Intellige...

sure, and many types of intelligence aren't well covered. however, most modern tests have a variety of subtests that are aggregated to produce the final value. People just talk about IQ because single #s are easy to talk about http://www.assess.nelson.com/test-ind/stan-b5.html

IQ is not a measure of intelligence, just as measurements of facial features is not a measure of beauty. It's qualitative, not quantitative.

If IQ is such a poor measure of general intelligence, it should be simple to establish a better method.

IQ is imperfect, but it is the best we have. IQ strongly correlates with performance on almost every other test of mental ability. There is no evidence to support the 'multiple intelligences' hypothesis. All of Gardner's 'intelligences' correlate across a population.

If IQ is a poor measure, devise a better test; All else is ideology, not science.

The main reason iq tests aren't so big anymore is because of the pc movement. They are the best way of testing for intelligence that we have so far.

I wonder what MENSA thinks of it.

Uh, doesn't any test that requires conscious engagement measure the desire of the test taker?

One could say that the scale in a doctor's office showed results commensurate with the desire of the person being weighed if some people had no desire to step on and thus weighed 0 lbs.

Reading list about IQ testing for those of you who would like more background on the subject:

http://en.wikipedia.org/wiki/User:WeijiBaikeBianji/Intellige...

So it seems like it helps measure desire too.

Is that so bad?

Yes, but desire to do well on a canned test, which AFAIK does little more than flatter your ego if you score high.

Personally, I always got really excited by these tests as a kid, but looking back I sort of admire anyone who blew them off because they would rather be playing outside.

Reply on news.ycombinator.com