An online IQ test that works?

[TooNice]

iq tests are mostly pattern matching - "spot the odd one out", "what's next in this sequence", "foo is to bar as baz is to ___" and so on

that kind of thing is extremely easy to learn [...]

iq is meant to test only "cognitive reasoning", not "knowledge". you're not meant to be able to revise for them, or practice for them other than the basic test structure.

That implies that pattern matching is more knowledge than cognitive reasoning. Something I find questionable.

[...]i tend to score about 20 points higher on crappy online iq tests than i did on professional tests taken as a fresh-faced youth, despite not having changed in intrinsic "smartness" over the years. if anything, booze has killed brain cells

I don't know what are the 'professional tests' you speak of, but when I speak of IQ tests, I refers to the ones recognised by high IQ societies (Mensa and the likes). And you can bet that all online tests are void. This does include the BBC one I've linked, but I do find it more credible than others.

@TheAnimus: I do not think that IQ tests is a test of whether you can think the same way as the people who've designed the test. There is nothing subjective to the answers. However, yes, the tests will emphasise on aspects the testers consider to be related to intelligence. You may disagree with the way certain aspects is weighted, or even what should be in the test. But there should be no problem as long as it is understood that the score is based on how well the person did in comparison to others for that type of tests.

@badass: There -is- a correlation between IQ and academic performance. But are you saying there should be absolutely no correlation between the two? That said, it is possible for someone who slacks in education school to have a very high IQ score (if I argue that 'knowledge' is not entirely separate from intelligence, I do think that there is more to intelligence than the knowledge of geography/history or even the ability to solve original maths/computing problems), and said person will not necessarily be top of the class.

[boxrick]

always find iq tests to be so random, never consistent and a bit naff... the only " accurate " ones are which asks loads and loads of questions ...

[directhex]

That implies that pattern matching is more knowledge than cognitive reasoning. Something I find questionable.

you're suggesting that you'd learn every possible pattern before a test and regurgitate (i.e. "knowledge"), rather than see a pattern for the first time, apply intelligence, and come up with an answer (i.e. "cognitive reasoning"), which is false

I don't know what are the 'professional tests' you speak of, but when I speak of IQ tests, I refers to the ones recognised by high IQ societies (Mensa and the likes). And you can bet that all online tests are void. This does include the BBC one I've linked, but I do find it more credible than others.

standard maths & verbal reasoning tests as conducted under test conditions by a child psychologist

and to be brutal, mensa tests are a joke. i got bored halfway through at age 11.

[BlindMelon7]

Oh heavy. What have I done!

[TheAnimus]

@TheAnimus: I do not think that IQ tests is a test of whether you can think the same way as the people who've designed the test. There is nothing subjective to the answers. However, yes, the tests will emphasise on aspects the testers consider to be related to intelligence. You may disagree with the way certain aspects is weighted, or even what should be in the test. But there should be no problem as long as it is understood that the score is based on how well the person did in comparison to others for that type of tests.

its mostly the word asociation ones that annoy me, rock is to stone, as seagul is to pigeon.

There are some which you can not fault, such as the one with the cubes that are half one colour, half another. You are then asked to use them to re-create patterns.

Too many test mearly aquired understanding.

[badass]

so how do you define intelligence, if not by the ability to solve problems?

Thats how I define intelligence, however people with a better vocabulary get higher scores as some of the questions are simply testing vocabulary amonsgt other things. Dont some of them also ask questions that require a knowledge of history.
If you take one person of a certain level of intelligence, and no education, get them to take an IQ test, then give them 10 years of edudcation, the same person educated will score a lot higher, yet their intelligence is the same.
Yes, there is a correlation between acedemic ability and intelligence, however you can have one without the other.
I have to admit, I did get some cruel pleasure at college with a particular group that had the acedemic ability but weren't that intelligent - they found that working hard being well behaved and looking down on others weren't all that was required to get A's at A-Level, and to rub salt in the wounds, me and a few mates did without doing much work at all

[MSIC]

Different strains of psychology, never mind biology, physiology etc, recognise that there are many different aspects to 'intelligence'.

The main reason for this is that intelligence is not a 'thing' in fact, but a concept, in a similar way to the concept of 'the mind' - useful for discussion purposes, useful to ascribe values to a person (or sometimes groups of people), but not technically correct.

Psychologists therefore will consider (amongst a great many other traits)...
Adaptability
Emotional maturity
Learnt knowledge & recall / memory
Capacity for further learning / memory capacity

And then subdivide within others etc etc....

In many ways i see parallels with computers.
Speed of thinking, capacity of storage, ability to reliably recall, strengths in certain fields....
CPUs, RAM, hard drives, graphics etc.

Asking "how good is my computer" is too vague and open ended a question unless you give it a context (ie "for what purpose") and intelligence is a similar concept - some people are simply better suited for certain tasks.

There is also an interesting discussion to be had i think with the idea that people can learn to answer tests, the basic premise being if a person learns, via repitition and over time, to answer IQ tests that have some minor variability to the questions, then by the time they get a high score have they improved their intelligence or just their ability to answer those sorts of questions?

[directhex]

Thats how I define intelligence, however people with a better vocabulary get higher scores as some of the questions are simply testing vocabulary amonsgt other things. Dont some of them also ask questions that require a knowledge of history.

no iq test will ask you for knowledge of history. ones which require manipulation of words are unfortunate - but years of psychological profiling have shown that people who are "smarter" have more varied vocabularies than others, so in some senses it's a logical thing to include on a paper

[directhex]

There is also an interesting discussion to be had i think with the idea that people can learn to answer tests, the basic premise being if a person learns, via repitition and over time, to answer IQ tests that have some minor variability to the questions, then by the time they get a high score have they improved their intelligence or just their ability to answer those sorts of questions?

which is why you only ever IQ test kids. testing adults is a fruitless exercise

[TooNice]

you're suggesting that you'd learn every possible pattern before a test and regurgitate (i.e. "knowledge"), rather than see a pattern for the first time, apply intelligence, and come up with an answer (i.e. "cognitive reasoning"), which is false

I did not suggest that (you skipped the 'I find questionable part'). I am merely interpreting what you previously said:

"iq tests are mostly pattern matching - "spot the odd one out", "what's next in this sequence", "foo is to bar as baz is to ___" and so on - that kind of thing is extremely easy to learn"

That statement suggests:
1. IQ Tests are mainly pattern matching.
2. Pattern matching is 'extremely' easy to learn.
Hence "IQ Tests are extremely easy to learn" (I disagree, but I see no other interpretation*).

I also stand by that an IQ test is about as reliable an adult as it is for a child. There is simply nothing stopping a child to be 'trained'.

---------------------------

Going back, no, I am not suggesting to learn every pattern and regurgitate. That's memory. And if memory plays some part in intelligence and knowledge acquisition, knowledge to me is not exclusively limited to that. You mentioned a correlation between vocabulary and 'smartness'. Well, I classify vocabulary as knowledge. But when you learn a word, you learn not just the word, but also the ability to derrive the meaning of the word preceeded/followed by certain prefixes/suffixes. And those become knowledge too.

Basically, I consider knowledge to be more than hard facts you've learnt, but anything that is indirectly generated from it. And in practice, the more you things learn, the more you 'generate' at the same time. That may explain why someone who spend a lot of time reading would better at verbal reasoning tests than if s/he didn't. And I suspect that a child who's been given a Lego as first his/first toy would have an advantage over a child who's not given anything in spacial visualisation tests.

---------------------------

@MSIC: I do agree that intelligence is a vague concept. Personally, I think that intelligence is a combination of most thing listed, and probably some unlisted. I don't know how I would 'weight' those elements. But I think most of those elements are interrelated to some extent (I would say less for emotional maturity). I think, and accept that IQ tests focus certain aspects more than others. People can think that what it measures is tosh, and that's fine. It has limitations, just like any other test. Incidentally, I do put a lil more faith in Mensa approved organisation a group of unnmed child psychologist.

[finlay666]

i think you mean over

no i mean under

because a childs brain is not yet fully devoloped and has still the capicity to learn so much more generally speaking

and if I did mean over 18 then surely all the test the nation tests, mesa IQ tests etc are completely pointless....

[directhex]

and if I did mean over 18 then surely all the test the nation tests, mesa IQ tests etc are completely pointless....

bingo

mensa's a club for bus drivers to feel better about themselves

[pollaxe]

bingo

mensa's a club for bus drivers to feel better about themselves

LOL! Well said!!

[chicken]

always find iq tests to be so random, never consistent and a bit naff... the only " accurate " ones are which asks loads and loads of questions ...

I've always found them very consistent, I have done several over the years and scored within 5 points of 135 every time. What's funny is I thought I'd give it a go when I was drunk once and got 165, tested it again about a year later and got the same, so I'm 30 points more intelligent after alcohol!

[badass]

bingo

mensa's a club for bus drivers to feel better about themselves

Roffle. Its funny because its true

I always called mensa "a club where academics pat each other on the back about how academically able they all are"
Personally, I'd rather have a job that allows me to get what I want out of life.

[TooNice]

So, someone can't be in Mensa and get a job that get what s/he want in life?

Some of the comments are really not above statements such as 'Computer enthusiasts are all geeks with no lives. I'd rather have a life than spend wasting my time squeezing that last % of performance on my PC". Or, "All models are stupid with no personality". "Lawyers are [...] who [...], professors are [...] who [...], people who fly first class are [...]" . Etc. *Assuming* you know anyone more than one individual who is part of the club, do they even act as you describe?

I am skeptical that there is a larger number of bus drivers in the club, or professors than there are successful businessmen/women.