Join 3,411 readers in helping fund MetaFilter (Hide)


Can anyone answer this IQ test question?
July 20, 2006 4:54 PM   Subscribe

Can anyone answer this IQ test question?

My brother had to take a personnel test for a potential employer and came back frustrated by one question. Does anyone know the answer?

Here it is:

Select the next number in the series:
C A R R
A: S
B: X
C: I
D: Y
posted by somethingclever to Education (47 answers total) 4 users marked this as a favorite
 
"Number?" Is that really the question?
posted by It's Raining Florence Henderson at 4:56 PM on July 20, 2006


yes.
posted by somethingclever at 5:01 PM on July 20, 2006


Ew. 'Cause the verbal side of my brain says that option D would be perfectly acceptable as the next letter in the series, because the letter "Y" is the only option that spells a word (CARRY). If they really specify "number in series," though... that's a whole 'nother question. I'll think on it on the drive home. They aren't Roman Numerals...
posted by It's Raining Florence Henderson at 5:05 PM on July 20, 2006


it's not very difficult, to be honest
posted by unSane at 5:06 PM on July 20, 2006


Keep in mind that some IQ tests work not by having one 'correct' answer, but by having several choices that would reveal different ways of thinking or problem-solving, depending on what the test-taker chooses.

(I would choose Y, because that turns the sequence into a word.)
posted by ferociouskitty at 5:07 PM on July 20, 2006


"D" is 500 in Roman Numerals, and "D" is associated with "Y", so "D" could be a viable answer.

I may be stretching it, but, it's the thought that counts, right?
posted by Raymond Marble at 5:11 PM on July 20, 2006


I would choose "I" because of the consonant, vowel, consonant sequence but the "number" thing is totally throwing me off.
posted by Ugh at 5:11 PM on July 20, 2006


I tried assigning numbers as if A was 1, B-2 etc etc and that makes the numbers

A1, B2, C3, D4, E5, F6, G7, H8, I9, J10, K11 ,L12 ,M13 N14, O15, P16, Q17, R18, S19, T20, U21, V22, W23, X24, Y25, Z26

3,1,18,18

A: 19
B: 24
C: 9
D: 25

as you can see this does not work either... haha I dont know what that proves... other than the fact that this test is crazy hard it was... or maybe it was a mistake?
posted by crewshell at 5:13 PM on July 20, 2006


Telephone keypad? Except who would carry that in their head into an IQ test? Gives 22 77 which might argue S, as it is another 7 in a sequence of 2 the same followed by 3 the same. But it's a hell of a stretch.
posted by A189Nut at 5:16 PM on July 20, 2006


Well, out with it, unSane! Let's have it!
posted by theredpen at 5:25 PM on July 20, 2006


The only answer that makes any sense is D: Y.

The 'Number' thing was either misprinted in the question or misremembered/misread by the candidate. The 'next number in this series' implies that C A R R are numbers, which they aren't. C, I, X and D are all Roman numerals but A, B, R and Y aren't.

Using the cipher A=1, B=2 etc doesn't resuit in any meaningful patterns and anything to do with telephone handsets relies on knowledge rather than intelligence.
posted by unSane at 5:32 PM on July 20, 2006


Er.. so it is a measure of IQ to know how to spell carry? I don't think so. There is something in here.
posted by A189Nut at 5:34 PM on July 20, 2006


According to this guy, the test asks for the next letter in the series, not number.
posted by Roger Dodger at 5:37 PM on July 20, 2006


Right - but his answer is equally based on it being the word Carry?
posted by A189Nut at 5:40 PM on July 20, 2006


Was this a straighforward paper test, where all the other questions had black-and-white correct answers? Or was it a personality "none of the questions have right answers, let's see how the candidate handles that" test?

unSane suggestion makes sense if the latter. But from the fact that somethingclever's brother remembered this question as being different from all the other questions, it seems more likely to be the former.

I don't know the answer, tho. The first thought that came to mind was, the letters might be numbers with some piece removed, for example C might be 0 with a piece missing. I haven't found numbers that work with A and R. (Both are sort of like 8, and to a lesser extent like 4.) This might depend a bit on what font they were printed in.
posted by LobsterMitten at 5:40 PM on July 20, 2006


Also, the test that guy was referencing is more a personality profile than measure of intelligence, so they probably have statistical data which shows whether or not you choose the answer that most other people choose, or if you are unafraid to go with a less popular answer. Even though the answer is meaningless, I bet most people choose Y, but those who think "outside the box" choose another answer.
posted by Roger Dodger at 5:41 PM on July 20, 2006


Then I choose S, to show my knowledge of Northern European Wetlands.
posted by A189Nut at 5:44 PM on July 20, 2006


I don't know either, but it looks like you're not the only one who is wondering...
posted by LimePi at 5:47 PM on July 20, 2006


And he says it is number too!
posted by A189Nut at 5:50 PM on July 20, 2006


A case could be made that C: I is the correct answer, since "I" is the only answer which is "smaller" (earlier in the conventional series of alphabet characters) than the 4 letter of the pattern, R :
Start - Smaller - Bigger - Start Next Group - Smaller - Bigger
C        A           R            R                I                   
Still an unsatisfactory answer in that it requires knowledge of alphabet order to derive, but frankly, most "intelligence" test verbal questions happily suppose such things as "common knowledge." Otherwise making verbal tests is just impossible.
posted by paulsc at 5:52 PM on July 20, 2006


CA = California

RR = Ronald Reagan

Answer is X = Vote for him

Evil Neo-con code question
posted by A189Nut at 5:56 PM on July 20, 2006 [1 favorite]


C -> A: -2, Decreasing by even delta, from odd to odd
A -> R: +17, Increasing by odd delta, from odd to even
R -> R: +0, Same by even delta, from even to even

Sequences to extend:
1: delta direction: -, +, 0 (one of each, so expect repeat, that is, - )
2: delta sign: even, odd, even (expect next to be odd )
3: transition: odd->odd, odd->even, even->even (all but even->odd, so expect that)

Possible answers:
R -> S: +1, Increasing by odd delta, even to odd (matches expected sign and expected transition, but not expected sign)
R -> X: +6, Increasing by even delta, even to even (matches none)
R -> I: -9, Decreasing by odd delta, even to odd (matches expected direction, expected sign, expected transition)
R -> Y +7, Increasing by even delta, even to odd (matches expected transition)

Only one answer (C) is decreasing, that makes it either wrong (if was filler in a the test with a single "right" answer) or interesting, if we don't have an obvious right answer and must eliminate answer that are the same.

Answer is C.
posted by orthogonality at 5:57 PM on July 20, 2006


"3: transition: odd->odd, odd->even, even->even (all but even->odd, so expect that)"

And<>also same sign transition, different sign transition, same sign transition, so expect a different sign transition. Like a composition of two functions, or addition of two since waves.
posted by orthogonality at 6:01 PM on July 20, 2006


The problem with all the proposed 'solutions' is that there is no way of choosing the right one. They are all equally im/plausible.

Having Googled a bit on Caliper, whose tests are NOT simply IQ but seem to measure a whole bunch of variables, I strongly suspect is that there is deliberately no right answer to this question and it may be a situational test to see if you move past it and finish the rest of the test, or get stuck on it -- which could tell the employer quite a lot about the way you might react to an intractable/insoluble problem.
posted by unSane at 6:06 PM on July 20, 2006


More info: on the Caliper site you can view a sample PDF of the kind of report they generate (you have to fill in a form but the info doesn't have to be genuine).

The section relevant to this question (I can't cut and paste as it is protected) strongly suggests that the test measures not simply analytical ability but analytical approach -- for example:

When making difficult decisions [Your Candidate] would tend to take a 'snapshot' view of the situation and formulate a quick fix [THAT'S ME -- UNSANE]. In doing so he is apt to experiment with innovative approaches and take advantage of short lived opportunities. However... he might not take time to consider possible consequences or pay close enough attention to potentially important details, and might make some hasty choices.

I think this question may address exactly this area.
posted by unSane at 6:15 PM on July 20, 2006


Aw, no fun unSane, if this just short form "Thermonuclear War."
posted by paulsc at 6:15 PM on July 20, 2006


Ooooh. Nice one, unSane. It never would have occurred to me that might be in there, and I've *seen* Star Trek II.
posted by baylink at 6:17 PM on July 20, 2006


Are we sure that the letters don't stand for words in some way?

Like "What's next in this sequence: M T W T" where it's "F" for "Friday"?
posted by AmbroseChapel at 6:50 PM on July 20, 2006


Circle (0), Arrow (1), Round (2), Round (3), X-ing (4)?

Was the test in english?
posted by qvantamon at 7:39 PM on July 20, 2006


You just made up this question, right?
posted by cellphone at 7:45 PM on July 20, 2006


Hah! This type of question shows up on this page...Brainteasers Inappropriate for Job Interviews!
posted by sLevi at 7:50 PM on July 20, 2006


I thought it was illegal to have an actual IQ test for employment that was not related specifically to the job.
posted by I Foody at 8:16 PM on July 20, 2006


A, C & B, in that order, which gives you the number SIX. That was the number of years that Bob Carr was Premier of New South Wales (duh).
posted by UbuRoivas at 8:37 PM on July 20, 2006


sorry this response is really late..i just got home and read all of these. LimePl, i saw the page you posted before i posted my question, which is why i went with the assumption that the actual wording was "number" and not just my brother reading/remembering incorrectly.
posted by somethingclever at 9:35 PM on July 20, 2006


I pick answer number D.
posted by flabdablet at 9:37 PM on July 20, 2006


Only one answer (C) is decreasing, that makes it either wrong (if was filler in a the test with a single "right" answer) or interesting, if we don't have an obvious right answer and must eliminate answer that are the same.

Answer is C.
Yes, this makes sense, but now the question is wrong: it should read something like "Select the next letter in this series numerically." By asking for a number, and giving only letters as options, this test gets a C-, with points off for posing as a trick question.
posted by Aquaman at 9:39 PM on July 20, 2006


the test my brother was taking was the caliper test, which i don't think i mentioned before, and he said that it was mostly personality-based but that there were a few questions thrown in that seemed to test IQ. it sounds pretty plausible that the purpose of the question is to test the taker's approach to problem-solving, whether or not she or he is able to move past a difficult problem, etc., as unSafe suggested.

also, the test was in english and my brother is a native english-speaker.
posted by somethingclever at 9:44 PM on July 20, 2006


Was the test computer-based, so that they could note the amount of time he took on that one question?

That makes sense to me. If you give someone twenty minutes to answer twenty questions, and one of them is essentially insoluble, do they get stuck on that question and keep trying to solve it when the other nineteen are easy?

I'm not sure who's testing for what -- someone who keeps on trying to solve one of twenty small problems, ignoring the other nineteen might actually be a good candidate for a coder at Google, but a really bad airline pilot.
posted by AmbroseChapel at 2:29 AM on July 21, 2006


I would choose "I" because only "X" and "I" can also be read as numbers, and if CARR is the beginning of a series, it could be a series of letters that makes up a word. Some of those letters are also roman numerals... In the word CARRION, "I" is the next letter after "C" that is also a roman numeral.
posted by taz at 3:07 AM on July 21, 2006


My guess is that the question is supposed to separate out pragmatists (who will answer "Y" on the assumption the question was probably wrong) from propellerheads who will either formulate an elaborate theory why one of the other answers may be right or just get stuck on that question and proceed no further.

That strikes me as a pretty poor test of personality: a pragmatist might just as easily leave the answers blank or guess one of the other values. Equally it is no doubt possible to come up with a complex numerical reasoning as to why the answer should be C.
posted by rongorongo at 4:28 AM on July 21, 2006


I'd have to see the answers to the rest of the test. If this question came up, and I had no clue what the appropriate answer was, I would come back to it at the end and count all the A's, B's, C's and D's answers I gave and answer this one with the one that I used the least. Not that that is a great answer, but I would try to increase my percentage of getting it right on a guess with a major, but baseless, assumption that the test developer would try to get the correct answers pretty well spread out among the four choices.

If, however, there was one letter that dominated the answers, I would choose that letter with the assumption the test developer had a bias.
posted by JohnnyGunn at 5:53 AM on July 21, 2006 [1 favorite]


My guess is that the question is supposed to separate out pragmatists (who will answer "Y" on the assumption the question was probably wrong) from propellerheads who will either formulate an elaborate theory why one of the other answers may be right or just get stuck on that question and proceed no further.

Wow, that's an objective way to look at it. Here, let me rephrase it for you:

My guess is that the question is supposed to separate out people who don't like to actually think about anything (who will answer "Y" on the assumption the question was probably wrong) from people who take questions seriously and assume the person asking them wants a serious answer.

Gosh, this is fun! In other news, whatever answer was intended, this is a terrible question, and I wouldn't want to work for anyone who thought it was a good think to dump on me at an interview.
posted by languagehat at 7:04 AM on July 21, 2006


Another simple twist:
There are 2 key words: "next" and "select".

"Next" supposes that there is at least 1 previous number. There is none. None = 0. So, if the question were: "What is the next nyumber...?", the answer would be 1, which is the next number after 0.

But the question is not "What" it is "Select". So we have to find the number "1" among the series "S, X, I, Y". I would take answer C: I = 1.
It could work.
Maybe.
posted by bru at 8:17 AM on July 21, 2006


"S"

Because obviously it has nothing to do with the "letters" it's a fully curvy letter, followed by a letter with only straight lines followed by a letter with curvy and straight lines. So to repeat the sequence you go back to the only letter with only curvy lines, the "S" (notice it's the only option that's not straight lines only)
posted by bitdamaged at 8:35 AM on July 21, 2006


I don't know if the test was taken on a computer or not..I'll have to ask him about that.
posted by somethingclever at 9:35 AM on July 21, 2006


JohnnyGunn: I'd have to see the answers to the rest of the test. If this question came up, and I had no clue what the appropriate answer was, I would come back to it at the end and count all the A's, B's, C's and D's answers I gave and answer this one with the one that I used the least. Not that that is a great answer, but I would try to increase my percentage of getting it right on a guess with a major, but baseless, assumption that the test developer would try to get the correct answers pretty well spread out among the four choices.

If, however, there was one letter that dominated the answers, I would choose that letter with the assumption the test developer had a bias.


I don't think there's any reason to assume equal spread (or a robust bias, for that matter), unless the test supervisor or the test material itself informs you so (which is almost never). orthogonality might have some more to say about this, but I can tell you from anecdotal evidence, it's a good strategy to fail a test. ;)

That aside, I'd go with the "red herring / test maker error" hypothesis, and assume D to be the correct answer.
posted by goodnewsfortheinsane at 9:40 AM on July 21, 2006


My guess is that the question is supposed to separate out pragmatists (who will answer "Y" on the assumption the question was probably wrong)

No, wrong is definitely X.
posted by weapons-grade pandemonium at 12:07 PM on July 21, 2006


« Older What recourse do fathers who r...   |  I read a book when I was 12 or... Newer »
This thread is closed to new comments.