Re: Race and IQ

On 15 Nov, 18:00, "Daniel T." <danie...@xxxxxxxxxxxxx> wrote:
Vend <ven...@xxxxxxxxxxx> wrote:
- There is no clear definition of "Intelligence", and thus there is
no clear measure of it. Tests measure a value called IQ which may
or may not correlate with what intelligence is intuitively
considered to be.

I don't think that is correct.

Intelligence is a very general mental capability that, among other
things, involves the ability to reason, plan, solve problems,
think abstractly, comprehend complex ideas, learn quickly and
learn from experience. It is not merely book learning, a narrow
academic skill, or test-taking smarts. Rather, it reflects a
broader and deeper capability for comprehending our surroundings
-- "catching on," "making sense" of things, or "figuring out" what
to do.

In over 50 years, Artificial Intelligence researchers managed to
develop lots of interesting and useful things but failed to meet their
original and primary objective: developing an human-like general
artificial intelligence.
One of the causes of this failure is that it was found that there is
no definition of intelligence detailed enough to be applied in
designing a machine.

Of the tasks you mention, for instance, those relative to reasoning
are limited by NP-hard and higher complexity classes and those
relative to learning are limited by the No Free Lunch theorems.

The only ways to get around those limits seems to be the usage of
domain-specific heuristics, approximations and simplifications, which
seems to be what humans do.

Therefore it could be well that a true general intelligence doesn't
exists, what really exists could be an ability to give "good"
responses, on average, in a particular environment, adapting to it
within certain strict boundaries.

Intelligence, so defined, can be measured, and intelligence tests
measure it well.

You can measure the ability of an individual to solve certain
problems, but you can't know whether the problem you chose are good
candidates for determining intelligence.

This applies in particular to the questions in which the subject is
shown some items arranged in a sequence or a table and he or she is
asked to insert an item in a gap or to find an outlier. In those
questions usually there isn't a logically deducible answer, only
pattern-matching can be used.

IQ tests measure how much the thinking processes of the subjects match
those of the test authors.

This doesn't mean that the tests are useless. They can predict
educational performance (which, as far as I know, was the original
purpose they were designed for) and to some extent professional
performance.
But I wouldn't say that they can actually measure intelligence.

They are among the most accurate (in technical
terms, reliable and valid)

What does it mean in technical terms?

of all psychological tests and
assessments. They do not measure creativity, character,
personality, or other important differences among individuals, nor
are they intended to.

[snipped agreeable things]

One other thing I think is important to note. It used to be that a score
of 50 meant you were as intelligent as the average person half your age
(and a score of 150 meant you were as intelligent as the average person
50% older than you.) However, when using such a scale, the scores level
off too much. The bell curve system was instituted so as to accentuate
the differences between individuals from the average.

I presume you are referring to tests for children.
I thought that child development is quite non-linear with time, thus
I'm perplexed of this neat linear relation between IQ and age.

Because of this accentuation, from what I understand (I can't find the
source) an individual's IQ can vary as much a 10-12 points from day to
day, depending on many environmental and emotional factors (e.g., how
much sleep have you gotten lately,) so the 15 point spread found between
US blacks and whites seems to be not that big of a deal.

Doesn't that mean that the tests are not very reliable?
(I don't know the technical definition, though) .

And as you say, experts agree that:

There is no definitive answer to why IQ bell curves differ across
racial-ethnic groups... Most experts believe that environment is
important in pushing the bell curves apart, but that genetics
could be involved too.

(Source for the above quotes:http://www.psychpage.com/learning/library/intell/mainstream.html)

.

Relevant Pages

  • (OT)Racial differences in Intelligence - What Mainstream Science says
    ... blacks or other native-born, ... The BELL CURVES of different groups overlap considerably, ... genetics plays a bigger role than does environment in creating IQ ... Racial-ethnic differences in IQ BELL CURVES are essentially the ...
    (rec.audio.opinion)
  • Racial Differences in Intelligence
    ... bell curves of different groups overlap considerably, ... Other groups (blacks ... Hispanics) ale centered somewhat lower than non-Hispanic whites. ... genetics plays a bigger role than does environment in creating IQ ...
    (sci.cognitive)
  • Re: Edit MS Active Directory
    ... that have an understanding and a large environment. ... Author of O'Reilly Active Directory Third Edition ... permissions on office network and want to remove the admin permissions ... What I need to do is to allow Helpdesk Team ability to logon and have ...
    (microsoft.public.windows.server.active_directory)
  • Re: UNIX SYS ADM HOUSTON
    ... ·Work in a varied environment that includes, but is not limited to, ... platform running Linux, HP-UX, Solaris and IRIX ... ·Demonstrated ability to automate various systems administrative ... THIS IS A TECHNICAL DISCUSSION GROUP, NOT A JOBS GROUP. ...
    (comp.unix.admin)
  • UNIX SYS ADM HOUSTON
    ... Job Description / Duties: Houston TX Opening ... ·Work in a varied environment that includes, but is not limited to, ... platform running Linux, HP-UX, Solaris and IRIX ... ·The ability to be productive with moderate supervision required along ...
    (comp.unix.admin)