in such a short amount of information you can find a pattern to explain any of the answers, it just depends about how complicated you want to get.

very similar to the fact that you can write down a string of any random numbers you like, and still find a pattern linking them all up

EDIT: feel free to disagree, but at least bloody quote me and have a discussion about it

Surely you couldn't suddenly decide that the circle has changed paths though? I'd say a sequence of 8 - in this case - is enough to suggest that there is only one course/ path for each object. For instance, the triangle would have no reason to suddenly end up in the centre. I could understand if it was shorter and a little more ambiguous perhaps but the paths are pretty strict in this case.

Reply 23

Daniel Freedman

7

Original post by shorty.loves.angels

Surely you couldn't suddenly decide that the circle has changed paths though? I'd say a sequence of 8 - in this case - is enough to suggest that there is only one course/ path for each object. For instance, the triangle would have no reason to suddenly end up in the centre. I could understand if it was shorter and a little more ambiguous perhaps but the paths are pretty strict in this case.

The thing is, as didgeridoo12uk is saying, it's feasible to ascribe to the sequence of diagrams any formula for a pattern that doesn't break what we can see. We could say that the diagrams progress as they do for the first 8, but then the 9th is always fixed and looks like A, or something. If this were the pattern (defined across 9 diagrams), there'd be no contradictions. Provided there's more than 1 diagram in the sequence, the number of diagrams doesn't make a difference.

BUT, this doesn't mean there isn't a right answer. Given the 5 possible answers, there is only one answer that fits the internal pattern defined by the first 8 diagrams (and this is so designed by the question maker. It's feasible to have this question as non-multiple choice). In other words, there is only one choice we can make without adding any of our own information about the sequence that has not already been shown by the question. Taking into account all the information from the first 8 diagrams, and adding no more characteristics to the pattern that we haven't already seen, it is possible only to arrive at one answer: C. If, for example, we instead chose D to be our 'correct answer', we'd see that this choice doesn't break the pattern that we've seen for the triangle and the star, but it DOES break the pattern we've so far seen for the circle (though as a sequence of nine terms it makes perfect sense, if we look at it is a continuation of the sequence already defined by the first 8 [which is the only sensible way to look at it], we see that 'D' is incorrect).

As for having questions like this as multiple choice, the reasons I see include: it's easier to actually make the test from a practical viewpoint, it allows the designer to eliminate the possibility of there being two different and non-incorrect continuations of the sequence, and it allows the test-sitter to use the process of elimination to help them.

EDIT: Note that when I'm talking about "these questions" in general, I am referring to the ones published as test questions. I don't mean that the above is true of any sequence of diagrams: it is possible to design a question where there are two possible patterns that a shape is following, and these patterns overlap for the entirety of the sample but then differ on the term that the test-sitter has to choose. The test makers are careful to avoid this happening (either by eliminating ambiguity through the multiple choice answers, or displaying a sufficient number of terms to unambiguously define the patterns).

(edited 13 years ago)

Reply 24

shorty.loves.angels

16

Original post by Daniel Freedman

The thing is, as didgeridoo12uk is saying, it's feasible to ascribe to the sequence of diagrams any formula for a pattern that doesn't break what we can see. We could say that the diagrams progress as they do for the first 8, but then the 9th is always fixed and looks like A, or something. If this were the pattern (defined across 9 diagrams), there'd be no contradictions. Provided there's more than 1 diagram in the sequence, the number of diagrams doesn't make a difference.

BUT, this doesn't mean there isn't a right answer. Given the 5 possible answers, there is only one answer that fits the internal pattern defined by the first 8 diagrams (and this is so designed by the question maker. It's feasible to have this question as non-multiple choice). In other words, there is only one choice we can make without adding any of our own information about the sequence that has not already been shown by the question. Taking into account all the information from the first 8 diagrams, and adding no more characteristics to the pattern that we haven't already seen, it is possible only to arrive at one answer: C. If, for example, we instead chose D to be our 'correct answer', we'd see that this choice doesn't break the pattern that we've seen for the triangle and the star, but it DOES break the pattern we've so far seen for the circle (though as a sequence of nine terms it makes perfect sense, if we look at it is a continuation of the sequence already defined by the first 8 [which is the only sensible way to look at it], we see that 'D' is incorrect).

Surely using only the information given is the whole point of these tests though? There isn't the scope for interpreting it in any other way due to the fact that there are a selection of answers given. Had it been left blank, then that would suggest more room for argument? Just an idea...

Surely being given answers to choose from actually adds to the information given. I don't think I'm explaining what I mean very well. But my point is, if we just use the information given (paths of shapes, continuous paths - not repetitions), then there is only one answer in this case. Whereas 'if we choose D'... well, if we weren't given a 'D' to choose from then we would not make any assumption that this answer could be possible. Terrible explanations. I'll go before I confuse myself :tongue:

TO ADD: Just saw your edit. I think it is to make marking easy to be honest. I think it also hinders people who are able to interpre things differently and justify other patterns.

(edited 13 years ago)

Reply 25

Daniel Freedman

7

Original post by shorty.loves.angels

Surely using only the information given is the whole point of these tests though? There isn't the scope for interpreting it in any other way due to the fact that there are a selection of answers given. Had it been left blank, then that would suggest more room for argument? Just an idea...

Surely being given answers to choose from actually adds to the information given. I don't think I'm explaining what I mean very well. But my point is, if we just use the information given (paths of shapes, continuous paths - not repetitions), then there is only one answer in this case. Whereas 'if we choose D'... well, if we weren't given a 'D' to choose from then we would not make any assumption that this answer could be possible. Terrible explanations. I'll go before I confuse myself :tongue:

TO ADD: Just saw your edit. I think it is to make marking easy to be honest. I think it also hinders people who are able to interpre things differently and justify other patterns.

I think I must have miscommunicated the point of my post.

All that's left for me to say, really, is: find me an example of a published question where you think there are two non-incorrect answers, and I'll show you why you're mistaken.