Analytical or Intuitive?

[elena]

I have a test online that measures cognitive style. It determines
whether the participant has an intuitive or analytical learning style.
It takes about 5 minutes.

It's here: http://www.elena.com

I'm a long-time software engineer who's thinking about changing
careers, hence the psychology course. I ended up wanting to study
people like me (not surprising I guess).

I know the sample will be non-probabilistic. So this will be a
quasi-experiment.

Elena

 

[Chris Uppal]

I have a test online that measures cognitive style. It determines
whether the participant has an intuitive or analytical learning style.
It takes about 5 minutes.

I gave this a whirl, but it won't let me submit my answers.

It wants an email address, and the address validation is tight enough to reject
(e-mail address removed), and while I could (trivially) invent something that did pass the
check, or (only slightly less trivially) bypass the client-side validation
altogether, I'm damned if I can see why I should be put to the bother of doing
so.

Incidentally, the 'submit' button is labelled 'Get My Score' which seems odd
when there are "no right or wrong answers"

-- chris

 

[Joona I Palaste]

I gave this a whirl, but it won't let me submit my answers.

It wants an email address, and the address validation is tight enough to reject
(e-mail address removed), and while I could (trivially) invent something that did pass the
check, or (only slightly less trivially) bypass the client-side validation
altogether, I'm damned if I can see why I should be put to the bother of doing
so.

Incidentally, the 'submit' button is labelled 'Get My Score' which seems odd
when there are "no right or wrong answers"

It's not a "good or bad" score, it's "cognitive style index". 0 to 38
means intuitive, 39 to 76 means analytical. I got 50.
Anyway, I agree about the e-mail address thingy. It's utterly
pointless. I put in "(e-mail address removed)". Apologies to any readers from Niue.

 

[Furious George]

I gave this a whirl, but it won't let me submit my answers.

It wants an email address, and the address validation is tight enough to reject
(e-mail address removed), and while I could (trivially) invent something that did pass the
check, or (only slightly less trivially) bypass the client-side validation
altogether, I'm damned if I can see why I should be put to the bother of doing
so.

Incidentally, the 'submit' button is labelled 'Get My Score' which seems odd
when there are "no right or wrong answers"

BZZZT - wrong answer...you failed - again.

 

[Chris Uppal]

[me:]

Incidentally, the 'submit' button is labelled 'Get My Score' which
seems odd when there are "no right or wrong answers"

It's not a "good or bad" score, it's "cognitive style index".

Then the button should be labelled 'Get My Cognitive Style Index' ;-)

-- chris

 

[Steve Green]

I have a test online that measures cognitive style. It determines
whether the participant has an intuitive or analytical learning style.
It takes about 5 minutes.

It's here: http://www.elena.com

I'm a long-time software engineer who's thinking about changing
careers, hence the psychology course. I ended up wanting to study
people like me (not surprising I guess).

I know the sample will be non-probabilistic. So this will be a
quasi-experiment.

Elena

I am a sucker for self assessment types of sites, so I definitly had to give
it a whirl.

I was a 39. It said that I was analitical, but only barely. It went on to
say that I would be a neutal learner.

Anyone else willing to share thier results?

--Steve

 

[Steve Green]

Anyway, I agree about the e-mail address thingy. It's utterly
pointless. I put in "(e-mail address removed)". Apologies to any readers from Niue.

I guess I am on to many mailing lists already and wasn't that concerned with
people getting it; I just used my normal e-mail, as I do with news groups
too.

I get a lot of mail ... it makes me feel popular ;-)

 

[christopher]

I took it the last time he posted it. The questions are overly vague,
assume a single angle for a given problem, lead or lean toward certain
answers, and generally fail to present a neutral environment for
acquiring non-biased data.

There is a science of survey taking, which in this case needs to
supercede the science of 'cognation' or whatever. I suggest the OP
learn his trade before wasting any more of our time.

 

[Chris Smith]

I was a 39. It said that I was analitical, but only barely. It went on to
say that I would be a neutal learner.

Anyone else willing to share thier results?

Sure, I ended up scoring 20, which puts me firmly in the "intuitive"
camp. However, I could just as honestly have answered the questions to
end up showing up as strongly "analytical".

I agree with the other respondent that the questions were in many cases
ambiguous and unclear. For example, my general approach to an important
problem is to look for intuitive solutions, and then evaluate each of
those carefully and systematically, falling back on a step by step
logical progression when that fails to yield a suitable result. That,
of course, makes it difficult to answer an either/or question about
whether I approach a task with logic or intuition, and most of the
questions were rewordings of this theme.

--
www.designacourse.com
The Easiest Way To Train Anyone... Anywhere.

Chris Smith - Lead Software Developer/Technical Trainer
MindIQ Corporation

 

[Patricia Shanahan]

I took it the last time he posted it. The questions are
overly vague, assume a single angle for a given problem,
lead or lean toward certain answers, and generally fail
to present a neutral environment for acquiring non-biased
data.

There is a science of survey taking, which in this case
needs to supercede the science of 'cognation' or
whatever. I suggest the OP learn his trade before
wasting any more of our time.

In particular, I stopped at the second question, due to lack
of quantification. ("Quantification" is the distinction
between "for all X" and "there exists X").

The first question "In my experience, rational thought is
the only realistic basis for making decisions." seemed to me
to suggest universal quantification. I think there are
issues for which e.g. personal likes and dislikes are a
realistic basis for making a decision, so my answer is "F".

The second question, "To solve a problem, I have to study
each part of it in detail.", does not indicate
quantification. It does not specify a particular problem. It
does not say "to solve any problem". It does not say "to
solve some problems". It says neither "I sometimes have to
study" nor "I always have to study". I can't answer "T"
because I've solved some problems without studying them in
detail. I can't answer "F" because I do have to study other
problems in detail. I'm not uncertain whether it is true or
false, I'm absolutely certain neither is a valid answer to
the question as posed, so "?" doesn't fit.

For most of the questions, my answer should be "It depends
on the circumstances".

The writer of the questionaire really needs to either think
though the quantification of each question, and make it
clear in the wording, or extend "?" to include certainty
that neither "T" nor "F" is always appropriate.

Patricia

 

[Bob]

The second question, "To solve a problem, I have to study
each part of it in detail.", does not indicate
quantification. It does not specify a particular problem. It
does not say "to solve any problem". It does not say "to
solve some problems". It says neither "I sometimes have to
study" nor "I always have to study". I can't answer "T"
because I've solved some problems without studying them in
detail. I can't answer "F" because I do have to study other
problems in detail. I'm not uncertain whether it is true or
false, I'm absolutely certain neither is a valid answer to
the question as posed, so "?" doesn't fit.

I think we can guess your likely result.

 

[Dimitri Maziuk]

Patricia Shanahan sez:

In particular, I stopped at the second question, due to lack
of quantification. ("Quantification" is the distinction
between "for all X" and "there exists X").

Hmm. Following pretty much the same logic I answered '?'
to *every* question, and whaddya know? -- I'm a natural.

There's also a science of interpreting the results: if
someone ticks the same box on every question, their
score should most likely be "you're shitting me, right?"

Dima

 

[Chris Uppal]

The writer of the questionaire really needs to either think
though the quantification of each question, and make it
clear in the wording, or extend "?" to include certainty
that neither "T" nor "F" is always appropriate.

I think you (and the others) are misunderstanding how this test is supposed to
work. Being too "analytical" in fact. Remember that the test is essentially
asking /one/ question in lots of different ways. The "fuzz" in the phrasing is
almost certainly there by design.

Consider what would happen if the questions were phrased as a precisionist
might hope (a precisionist who knew nothing about psychological testing, anyway
;-). The answers would obviously have to include an option, "it depends on the
circumstances", and I imagine that just about everybody would answer that to
just about every question. Which would produce no information.

-- chris

 

[Dimitri Maziuk]

Chris Uppal sez:

I think you (and the others) are misunderstanding how this test is supposed to
work. Being too "analytical" in fact.

There's three angles to each test: the test, its author, and its
intended audience. In this case intended audience is people who
program computers -- occupation that requires above average
analytical skills, attention to detail, and retentive memory.
Patricia's commentis exactly what you should expect to get if
you ask a bunch of programmers a bunch of extremely vague questions.

Dima

 

[Chris Uppal]

There's three angles to each test: the test, its author, and its
intended audience. In this case intended audience is people who
program computers -- occupation that requires above average
analytical skills, attention to detail, and retentive memory.
Patricia's commentis exactly what you should expect to get if
you ask a bunch of programmers a bunch of extremely vague questions.

Agreed entirely that that's the kind of comment that you'd expect to hear
/about/ the test (and I didn't find the questions any more comfortable than
anyone else -- I don't like vagueness either), but that doesn't mean that the
test isn't well-designed. Indeed, I can quite easily imagine the test's author
being just as precise as, say, Patricia or Chris, and finding it even more
painful to pose such mushy questions -- even while knowing that it was
necessary...

I'm not claiming that it /is/ well designed either -- a test is validated by
the its statistical sensitivity and accuracy when applied to a target
population, not by armchair theorising. Not even by /my/ armchair
theorising...

-- chris

 

[Patricia Shanahan]

Dimitri Maziuk wrote:




Agreed entirely that that's the kind of comment that
you'd expect to hear /about/ the test (and I didn't find
the questions any more comfortable than anyone else -- I
don't like vagueness either), but that doesn't mean that
the test isn't well-designed. Indeed, I can quite easily
imagine the test's author being just as precise as, say,
Patricia or Chris, and finding it even more painful to
pose such mushy questions -- even while knowing that it
was necessary...

I'm not claiming that it /is/ well designed either -- a
test is validated by the its statistical sensitivity and
accuracy when applied to a target population, not by
armchair theorising. Not even by /my/ armchair
theorising...

-- chris

There is an additional consideration for testing using
volunteers. It is essential that there be nothing about the
test that will selectively discourage test taking in any way
that correlates with whatever is being measured.

In this case, the structure of the test tends to selectively
discourage people who combine mixed problem solving and
learning strategies with a passion for accurate
communication. The combination might conceivably be commoner
among programmers than in whatever population was used in
validating the test.

Patricia

 

[John B. Matthews]

There is an additional consideration for testing using
volunteers. It is essential that there be nothing about the
test that will selectively discourage test taking in any way
that correlates with whatever is being measured.

In this case, the structure of the test tends to selectively
discourage people who combine mixed problem solving and
learning strategies with a passion for accurate
communication. The combination might conceivably be commoner
among programmers than in whatever population was used in
validating the test.

Patricia

This is quite insightful. My own selection bias: I assumed it was an
email harvesting scam

 

[iamfractal]

I was a 39. It said that I was analitical, but only barely. It went on to
say that I would be a neutal learner.

Anyone else willing to share thier results?

--Steve

Damn it, I got 60.

Then it told me to get a life.

Or a girlfriend.

..ed

www.EdmundKirwan.com - Home of The Fractal Class Composition.

 

[Chris Uppal]

There is an additional consideration for testing using
volunteers. It is essential that there be nothing about the
test that will selectively discourage test taking in any way
that correlates with whatever is being measured.

In this case, [...]

Yes, that's a good point.

I doubt whether the ideal is achievable, though. E.g. the test has presumably
been validated against a population that speaks English as its first language,
but I doubt if that's representative of the people who might follow a link from
a Java newsgroup.

Other skews include:

- the email (mis)handling eliminates both irritable purists (such as
myself) and suspicious buggers (such as John).

- pointing the test at a Java population will (I'd guess) tend to include a
fairly high proportion of would-be programmers who do not, in fact, "combine
mixed problem solving and learning strategies with a passion for accurate
communication". (Yes, that's a value judgement about Java programmers'
demographics compared with the programming community in general. It may be as
wrong as it's unflattering -- it's only a personal opinion anyway -- but there
are legitimate grounds for /questioning/ how representative, statistically,
Java programmers are (or, if you prefer, what they are representative of)).

- A not-uncommon attitude to such tests among programmers is the "hacker"
attitude: "I'm taking this test because I'm interested in how the test is
constructed, not in what it might 'tell' me about myself". That's the attitude
I approached it with, and I suspect that was your attitude too (plus, perhaps,
a willingness to help someone who needed to collect data). The test may
attract disproportionately many people with that attitude.

and so on...

-- chris

 

[elena]

OK, right now I'm trying to determine who's the toughest audience:
Java, C, or C++ programmers. At this point I'm beginning to think that
the response to taking the test may end up to be more interesting than
the results.

Anyway, I apologize for the non-technical nature of the post. I am not
an evil email-gathering spam-bot. Just to let you know, the nature of
cognitive style is being debated. This particular test assumes it is a
unitary construct. Others think context matters.

Please accept my thanks for helping and taking the test.

Elena