The accuracy of IQ tests
Robert Calvert
accuracy of an IQ test, how would I do it. This is what I was asking
myself earlier today. Then it hit me.
I know many people who studied a public school curricula, went on to
college and lived out their entire lives, but yet they still believed
that the earth had to have been created in six literal days as taught
in the Bible. I, on the other hand, was brought up in a fundamentalist
household, and went to a private Seventh-Day-Adventist church school.
But, in spite of all this, I was able to figure out on my own that the
Theory of Evolution had to be a much more logical explanation for how
we all came into being. This would suggest to me that any question
that requires pure <g> to answer is not time dependent at all (i.e.
theoretically, you would either get it right away or you would never
get it at all). Conversely, any question (that you already have the
knowledge to solve) that doesn't require any <g> to answer is going to
be totally time dependent.
So, based on the above premise, here is the way I decided that an IQ
test should be judged. If the average person can increase his/her IQ
score by 0.66 SD merely by doubling the time spent taking the test,
then the test in question probably doesn't measure intelligence at
all. If, on the other hand, the score doesn't change by doubling the
time spent taking it then, chances are, it measures intelligence with
100% accuracy. I'm sure that real world IQ tests fall somewhere in
between these two extremes. The extent to which various kinds of IQ
tests are time dependent would seem to be a good way to compare their
accuracy. All this assumes, of course, that the tests in question are
properly normed and use questions with only one objectively definable
answer. If the tests in question are inherently bogus (i.e. use
questions with more than one correct answer, require the wrong answers
to questions, require knowledge that's not made available to the test
taker, etc.), then this technique probably won't work vary well. But
if the score on any IQ test (bogus or otherwise) increases 0.66 SD
when the average person doubles his/her time taking the test, then I
can say with reasonable certainty that the test in question is
worthless.
Would the critics of IQ tests be satisfied with this definition of
accuracy?
Robert
Bigfoot
theory? Well, one is born every minute I guess. I have no solid opine on
the matter, other than that people who think they DO know are fooling
themselves.
Ever heard of Piltsdown man? It was taught as "fact" in public schools for
30 years..until it was revealed as a complete fraud. I bet some people here
would like to sweep that episode under the rug, huh? But I digress from the
main thrust of your posting.
Lastly, it's virtually impossible to "study" for a legitimate IQ test. SAT
isn't a legitimate IQ test. Can you study for the Rubiks Cube? Yeah..but
you'd better be a smart SOB to remember the tricks.
"Robert Calvert" <Herc...@pcstarnet.com> wrote in message
news:a3cd6a89.03021...@posting.google.com...
Alan
The other night, on CBS News, there was a story about some scientists in
England who were testing crows. The scientists put a piece of meat into a
little basket. The basket was about the size of a pebble (small enough for a
crow to lift) and put in the bottom of a long glass vial. The vial was about
two feet long. Within the same room the scientists located a couple of
wires. The wires were straight and also about two feet long.
One by one crows were put in the room. Almost all of them walked around and
around the vial pecking at the glass trying to get at the meat. One of the
crows grabbed a wire in its beak and hoisted it up. Soon after it just
dropped the wire. You could tell it was trying to figure out how to use the
wire to get at the meat. Who knows, perhaps with time this bird could figure
out something with the wire. However, one particular crow was put in the
room that knew what to do after only a brief period of thinking. (It was SO
interesting watching the crow ponder his predicament!) This crow not only
grabbed the wire, but bent a little crook in it. With the wire, he reached
down the vial and grabbed the little basket, pulling the whole thing out of
the tube.
There was no time restraint on the experiment as far as I could tell.
The only arguments I can see for time restraints are 1)practicality. Who
wants to wait around forever while someone stares at a test. 2) Pressure. In
our everyday lives we must meet deadlines. Granted some don't do well with a
deadline hanging over their head, but deadlines are part of our lives. 3) A
time constraint can be looked at like a picture frame, (limitation) and most
IQ tests are like snap shots. That said, perhaps there are IQ tests that
don't have time constraints.
Thanks again,
Alan
--
Windsurfing Club: http://www.ibscc.org
"Robert Calvert" <Herc...@pcstarnet.com> wrote in message
news:a3cd6a89.03021...@posting.google.com...
Bill Vajk
Cool story!
There's another side to it, in nature anyway.
The crow that gives up on the problem quickly has a pretty good
chance of stumbling onto an even better opportunity elsewhere.
They're primarily opportunists. That's what they do best.
In human parlance what the crow did is sometimes called tripping
over a dollar bill to pick up a penny. That's why test taking
involves doing all the questions/problems that come easily first,
then returning to the more difficult ones (provided there's any
time remaining.)
In fact in life we often pigeonhole stuff we decide we'll return
to later when it is more convenient or appropriate. I have been
declared odd because a lot of the time I do, while for most
people 'out of sight--out of mind' rules.
Cleverness stopping a person from following their normally successful
decision tree is probably generally (how's that for a hedge?) not a
good thing.
SPAWN OF A JEWISH CARPENTER: CINDY SMITH
Herc...@pcstarnet.com (Robert Calvert) writes:
No, I wouldn't. People with ADD or other learning disabilities are
often given additional time on IQ tests, and their scores increase by
a high percentage. In experiments, students without ADD or any
learning disability were given additional time on standardized tests,
but their scores did not increase measurably. How fast you can answer
questions on a test is not as important as whether you can answer the
question. Giving ADD and other learning disabled students additional
time simply levels out the playing field.
> Robert
--
Cindy Smith Unless the LORD build the house,
c...@dragon.com they labor in vain who build.
c...@5sc.net Unless the LORD guard the city,
c...@romancatholic.org in vain does the guard keep watch.
Me transmitte sursum, -- Psalm 127:1
Caledoni! All your base are belong to us.
A Real Live Catholic You are on the way to destruction.
in Georgia! What you say.
>->> <<-< Go against the flow! You have no chance to survive make your time.
Alan
ADD. That's done by a time constraint on one of the subtests. My
girlfriend's son is on Aderal now because of his performance (or lack of) on
one of the subtests. We weren't even that interested in his final results
btw. Had he been given an extra amount of time we wouldn't have found his
specific problem
In 'real life' we can't always give sufferers of ADD additional time.
Additional time is what we want with the Iraq situation, but apparently your
prez has a time constraint.
Time is money in the business world. I'd live to give some of my ADD
employees time to finish tasks, but is that fair to the ones who don't
demand more time?
If your goal is to find a way someone who suffers from ADD to score high on
an IQ test...that's a different issue
Alan
--
Windsurfing Club: http://www.ibscc.org
"SPAWN OF A JEWISH CARPENTER: CINDY SMITH" <c...@cygnus.dragon.com> wrote in
message news:JCZ58G...@cygnus.dragon.com...
Dante Alighieri
:^) Theory of Evolution had to be a much more logical explanation for how
:^) we all came into being.
In the "beginning" all was "plasma", pure energy theorized to be the
smallest particles we know and energy such as the radio spectrum
that has a "particle theory" as one of it's theories and energy such
as magnetism and gravity. . During a length of time incomprehensible
to man the particles randomly collided, bounced off each other, hit again
and "stuck together" with whatever adhesive force that we still theorize.
The new, larger random particles grew in size in similar fashion. In time
the elements were formed and themselves randomly "connected" forming
molecules and so forth. Much of this random debris still inhabits most of
the universe. The molecules developed a "survival" instinct and began
to replicate themselves. The "single cell creatures " combined on occasions
with different molecular clusters each doing single tasks. So forth and so on
for time incomprehensible to man this process continues throughout the "space"
we call the universe.
This is logical. Anything else is not. Humans are capable of theorizing much
that other life forms cannot. We are still a random collection of space debris.
We have "learned" to use other randomly formed clumps of space debris to
reproduce our particular random collection of particles and energy. That's all.
michelangeloĊ½ merisi da caravaggio, a Henotheist, whatever that may be.
Dante Alighieri
:^) time restraints
Our theorization that if you instantly see a giraffe in a refrigerator minus one "f",
no, bad example, ...
Point being the test has many questions developed. Then the test is given
to as large a base as possible. The "good" questions are used, the "bad" are
discarded. A very large representative group oh high school students were
given a test during their junior year for several years that evolved into the ACT.
How quickly an answer is seen IS NOT always indicative of a higher intelligence
quotient - BUT - sometimes it is an accurate indication. The accurate indicators are
among the questions kept.
We deal in generalities, indicators, because for thousands of years every attempt
to cram humans into a mold has failed. AND WE"RE STILL TRYING ! ! ! ! ! ! !
Not me as an individual, not everyone, but some of our fellow humans are still trying
to just that.
Robert Calvert
This would suggest to me that, for what ever reason, IQ tests may not
be accurately measuring the intelligence of students with ADD even
though it may still be accurately measuring the intelligence of normal
students.
Robert
Robert Calvert
test of the people taking the test.
Imagine an extreme example in which the general population is given an
IQ test that requires them to do 100 first grade arithmetic problems
in 20 seconds. We would inevitably find that some people could answer
more of the questions than others. And if we tested enough people, we
would find a few who could score in the 99.9999999987th percentile
(i.e. an IQ of 200). But I'm sure that most reasonable people would
agree that starting with a simple test like this and extrapolating
such an extreme conclusion from it is a stupid way to try to measure
intelligence. And, if we didn't know anything else about this test, we
could easily reveal the absurdity of it by simply doubling the time
that people are allowed to spend on it. If we did this, we would find
that the vast majority of people would double the number of problems
they could solve correctly. But now let's say that (for example)
0.0429% of the population was unable to significantly improve their
scores by doubling the time spent on this test. From this, we could
deduce that this test is only accurate up to an IQ score of 50. In
fact, even if we're not smart enough to answer any of the more
complicated questions in a given IQ test, we could still determine the
point at which it's accuracy begins to drop off by using this
technique.
Robert
winter
> In 'real life' we can't always give sufferers of
> ADD additional time.
in real life, those who exhibit add behavior
have survived till now; in real life, they'll
continue to survive, as they have till now.
we've gone through a time of exhibiting reasons
for a kind of behavior that made demands of us,
that behavior appears to have been rejected by
a body of our population sufficently large and
organized to make its opposition determinative.
(they alone in this country did that, but they
are far from alone in the world.)
now we've no time, and we're hustling with all
the muscle that detroit can reasonably bolt on,
now its compassionate conservativism.
(conservation takes effort and energy; have you
ever noticed what "conservatives" give no real
though to spending that they may conserve what?
there are always those that a name does not en-
compass.)
one of the most interesting things about add
is that it now has a name; the effort that gave
it a name will continue, and where that will lead...
as far as "the accuracy of iq tests" are concerned
iq tests will continue to highlight, but they'll
(probably) evolve (to include measures of accomp-
lishment, and/or some measure of consistancy in-
volved with idea/behavior tables for examples;
indeed it seems the call (aside from vanity) to
engage them uses some such unstructured assess-
ments now.)
(then there's always 'accuracy in measuring ,,, what?'
(there's a story about how the apple of discord came
to be, and our conservatives would wish to conserve
what?))
these things seem before us.
we seem in titan time.
rgrds,
David Ray
"Robert Calvert" <Herc...@pcstarnet.com> wrote in message
news:a3cd6a89.03021...@posting.google.com...
> accuracy of an IQ test, how would I do it. This is what I was asking
> myself earlier today. Then it hit me.
>
You have to consider that the number 1 criticism of IQ tests, at least that
I have heard in my academic experience, is that IQ tests are socially and
culturally biased. Time given to take the test, whether increased or
decreased, would probably create no significant difference, especially not
the .66 SD that you speak of.
It is my opinion, therefore, that we need to re-valuate education before we
re-evaluate IQ tests and the like (i.e. GRE, SAT, etc).
Regards,
Dave
Dante Alighieri
:--)Point being the test has many questions developed. Then the test is given
:--)to as large a base as possible. The "good" questions are used, the "bad" are
:--)discarded.
In some tests a few "bad" indicators are kept.
The test, therefore, will contain "dummy" questions that are not
included in the evaluation.. The test taker does not know if any
question is relevant or will be ignored in the evaluation process.
Time is a factor. If the testee takes too long answering the placebo
and does not answer "counted" questions a lower evaluation will result.
Just getting to work?
IP: 64.198.49.221
DNS: zzz-064198049221.splitrock.net
SPAWN OF A JEWISH CARPENTER: CINDY SMITH
> The best function I've seen of honest-to-goodness IQ tests is to diagnose
> ADD. That's done by a time constraint on one of the subtests. My
> girlfriend's son is on Aderal now because of his performance (or lack of) on
> one of the subtests. We weren't even that interested in his final results
> btw. Had he been given an extra amount of time we wouldn't have found his
> specific problem
> In 'real life' we can't always give sufferers of ADD additional time.
> Additional time is what we want with the Iraq situation, but apparently your
> prez has a time constraint.
> Time is money in the business world. I'd live to give some of my ADD
> employees time to finish tasks, but is that fair to the ones who don't
> demand more time?
> If your goal is to find a way someone who suffers from ADD to score high on
> an IQ test...that's a different issue
Are you suggesting that Mensa should not accept standardized test
scores from those who have been given additional time on the tests?
I see nothing wrong with giving certain students with documentation
additional time on tests. Many jobs allow people to take additional
time on their work. One can have time constraints and still allow
additional time. Many businesses would be the poorer if they rejected
excellent work performance simply because it was done more slowly than
others.
Bigfoot
less..decided to create a "culturally" blind IQ test. There was no written
question on the exam...all the questions were visual. They received test
results that virtually confirmed every IQ test given in the USA in the last
30 years.
Funny , isn't it? Despite the fact that a completely 'culturally blind' test
was given with similar results..how those results are ignored? I don't
think they CAN be ignored....I think it's a FACT that differences DO exist
between the different groups. What amazes me...is how so many american
"intellectuals" can continue to claim "everybody is equal in
everyway"...when 70% of the NFL and NBA are black. OBVIOUSLY blacks are
superior athletically in the USA ( at the risk of being labelled
racist..it's due to selective breeding of blacks when slavery existed in the
USA), which I completely agree with...but why are so many people eager to
endorse that feeling while denying the obvious mental differences?
Intellectual cowardice.
"David Ray" <dave....@earthlink.net> wrote in message
news:DdS3a.5113$_c6.4...@newsread2.prod.itd.earthlink.net...
Robert Calvert
> LOL...and Mary...to measure intelligence what would you use?? 50 yard dash
> times? I think you KNOW the facts but are too "politically correct" to state
> them, which I find reprehensible.
Yes, I agree. She's being way too nice to you. :-)
Robert
> I can understand someone saying something
> to save their lives..but saying something just to "fit in" is akin to
> "selling your soul".
>
> There comes a point when being "nice" can violate FACT. Apparently you
> prefer to err on the side of "nice", which is a common failing...you are
> afraid of offending someone. I could go into a detailed "psychological
> profile" here, with a great degree of accuracy..but I won't embarrass you in
> that regard. What I WILL say..is that you feel a need to be "accepted". The
> only reason I say this with such conviction is because I KNOW you are an
> intellectual......if you weren't I wouldn't bother responding.
>
> Do you feel it's more important to be "accepted", or to be RIGHT?? If you
> only answer one point of my posting..answer this.
>
> "Mary" <dev...@mensa.org> wrote in message
> news:rl5t4v8v47h08npdl...@4ax.com...
> > On 13 Feb 2003 17:06:16 -0800, Herc...@pcstarnet.com (Robert
> > Calvert) wrote:
> >
> > >So, based on the above premise, here is the way I decided that an IQ
> > >test should be judged. If the average person can increase his/her IQ
> > >score by 0.66 SD merely by doubling the time spent taking the test,
> > >then the test in question probably doesn't measure intelligence at
> > >all.
> >
> > If you are looking at tests that measure intelligence why are you
> > looking at IQ tests. IQ tests measure only IQ -- Yes IQ is related
> > in some ways, but intelligence is not really measurable.
> >
> > Mary
> >
Robert Calvert
> David..you may be interested in the following. The JAPANESE, no
> less..decided to create a "culturally" blind IQ test. There was no written
> question on the exam...all the questions were visual. They received test
> results that virtually confirmed every IQ test given in the USA in the last
> 30 years.
>
> Funny , isn't it? Despite the fact that a completely 'culturally blind' test
> was given with similar results..how those results are ignored? I don't
> think they CAN be ignored....I think it's a FACT that differences DO exist
> between the different groups.
They certainly do:
"So you believe a 1 in 9 billion chance is more "logical" than the
creation theory? Well, one is born every minute I guess."
Now, would you be willing to admit that you have a below average IQ
yourself?
Robert
Robert Calvert
> news:a3cd6a89.03021...@posting.google.com...
> > If I had to conjure up a scientifically objective way to judge the
> > accuracy of an IQ test, how would I do it. This is what I was asking
> > myself earlier today. Then it hit me.
> >
> (Rest Deleted to save bandwidth)
>
> You have to consider that the number 1 criticism of IQ tests, at least that
> I have heard in my academic experience, is that IQ tests are socially and
> culturally biased.
Unfortunately, the number 1 criticisms are also the most politically
correct. If the test is culturally biased, then it would be what I
described earlier as "bogus". (i.e. use questions with more than one
correct answer, require the wrong answers to questions, require
knowledge that's not made available to the test taker, etc.) Barring
this, the only real problem that an IQ test might have is it's ability
to measure <g> and nothing else. A good example of what <g> is not can
be found in the unusual talents of an idiot savant. Even though such a
person can accomplish incredible mental feats very rapidly, I don't
know of anything an idiot savant can do that a normal person can't do
given enough time. To make a long store short, true <g> is the ability
to push into new frontiers of thought, not rehash the old ones
rapidly. I've heard many people compare a contest of intellect to a
"race". Personally, I don't think that this is a very good analogy.
True <g> is more analogous to a threshold condition than anything else
(i.e. you either get it or you don't). And ascertaining the time
dependency of an IQ test would be the best way to determine this.
> Time given to take the test, whether increased or
> decreased, would probably create no significant difference, especially not
> the .66 SD that you speak of.
>
Then this would suggest to me that most IQ tests currently in use
measure intelligence fairly well. Assuming that what you say is true
of course.
Robert
Robert Calvert
> Mary <dev...@mensa.org> wrote:
>
> >IQ is only a glimpse at some of the outward signs of intelligence as
> >they pertain to academic skills, no more and no less.
>
> Mary,
>
> the problem with that statement is that it is based on your
> personal definition of intelligence, but everybody has a
> different one.
>
> Non-scientists have a very strong tendency to define
> intelligence in an extremely wide sense. (A child that can climb
> well has climbing intelligence, etc.) The reason for that is
> often that people fear the concept of a clearly defined and
> measurable intelligence.
>
> If you want to use the word intelligence in a scientific
> context, you have to step back and use a definition from the
> field of psychology or psychometry, which is much narrower.
>
> Or you just drop the word altogether and use IQ instead, which
> is much better defined. And the IQ turns out to be quite
> meaningful on its own. Please read
> http://www.michna.com/intelligence.htm for some scientific
> consensus information on the predictive value of the IQ.
>
Here's an interesting excerpt from that webpage:
"Most standard tests of intelligence have been constructed so that
there are no overall score differences between females and males."
I wonder how they go about accomplishing this. I'd be willing to bet
that the tests women score higher on are more time dependent than the
ones that men score higher on. :-)
Robert
> We've been close to this subject before, but it's been a while,
> so some readers might not mind the refresher course.
>
> Hans-Georg
Hans-Georg Michna
>Hans-Georg Michna <hans-georgN...@michna.com> wrote in message news:<bda55vg03ir8r4gdp...@4ax.com>...
>>... Please read
>> http://www.michna.com/intelligence.htm for some scientific
>> consensus information on the predictive value of the IQ.
>Here's an interesting excerpt from that webpage:
>
>"Most standard tests of intelligence have been constructed so that
>there are no overall score differences between females and males."
>
>I wonder how they go about accomplishing this. I'd be willing to bet
>that the tests women score higher on are more time dependent than the
>ones that men score higher on. :-)
Robert,
the article already has the answer. women score better in some
tests (verbal) while men score better in others (like 3D
rotations). You only need to adjust the weights of the tests.
Since the differences obviously cancel each other out to some
extent, the overall difference isn't that big anyway, so this
doesn't devalue the concept of IQ much. It does show though that
intelligence is structured, and the IQ captures something like a
total. Indeed some intelligence tests give subscores.
Hans-Georg
--
No mail, please.
Bigfoot
"Polly Wog" <Poll...@hotmail.inv> wrote in message
news:nrc4a.8707$_k4....@nwrddc03.gnilink.net...
>
> "Bigfoot" <to...@zianet.net> wrote in message
> news:3e514881$1...@news.zianet.com...
> > David..you may be interested in the following. The JAPANESE, no
> > less..decided to create a "culturally" blind IQ test. There was no
written
> > question on the exam...all the questions were visual. They received test
> > results that virtually confirmed every IQ test given in the USA in the
> last
> > 30 years.
>
> not surprising, actually
>
> > Funny , isn't it? Despite the fact that a completely 'culturally blind'
> test
> > was given with similar results..how those results are ignored? I don't
> > think they CAN be ignored....I think it's a FACT that differences DO
exist
> > between the different groups. What amazes me...is how so many american
> > "intellectuals" can continue to claim "everybody is equal in
> > everyway"
>
> okay, I'm coming in late, and it's not in the thread on my server. who
> exactly are you quoting here? the Coeur d'Alene Welcome Wagon? most
> intellectuals I know know they have a higher score on a type of test.
that
> doesn't relate to any real-world ability to perform. it also (more
> importantly) doesn't automatically grant someone extra privleges compared
to
> another human being; we are all equal in what rights we should have (you
> could ignore the word "should" and rip me to shreds for about half a
second.
> if you wish to redefine what most of civilised society chooses to believe,
> maybe you should go start your own 3rd world country and hoard weapons of
> mass destruction. I'd like a reason to blow you to hell, as the same
> constitiution that grants human equality to everyone, also grants you free
> speech. in case I'm not being clear, I'll say it in your own degenerate
> language, consisting of small easy-to-pronounce words: do not spill your
> bile here; I'll feed you your gall bladder after I rip it from your body.
> now, back to the discussion at hand).
>
> blacks, by virtue of their shorter muscle structures (compared to whites)
> look bigger and more muscular. selective breeding? maybe. actually
> stronger? no. maybe it has something to do with thousands of years of
> selective breeding, like the color of their skin, or maybe it's just a
short
> while as slaves. even an intellectual coward will realize statistically,
> the odds are against this muscle structure being a result of slavery.
now:
> mental differences? here, I feel compelled to turn your theory of
selective
> breeding back on you, only modified slightly. it's very arguable that
this
> is a learned behavior, taught by generations of circumstance rather than
any
> realistic difference. call it selective training. this proves a people
can
> be taught they are unliked by anonymous faceless persons representing a
> small percentage of the over-all population who will use this people as
> scapegoats to explain their own felt inadequacies and persecution
complexes
> (hint hint), and use random violence to instill terror in a people who
> appear different, but are guaranteed protection.