meadors@cogsci.ucsd.EDU (Tony Meadors) (05/27/89)
I proposed: >> I would bet that given a "real" cross-section of individuals (think of >> testing everyone that enters the DMV for example) you would find quite >> a tidy correlation between performance on a spelling test and a standard >> IQ test: say, 0.3 to 0.4. Why this is, as well as what it means to >> designers, educators, and we, the poor spellers, is an open question. J. Ellis wrote: >You would lose -- As I posted a couple of weeks ago, research has >discovered that poor spelling is a form of dyslexia that is an >inherited trait -- the problem is that some people have difficulty >visualizing letter and numbers (e.g. phone numbers) in the correct sequences I presume what you mean is that "in addition to the many influences which affect one's development of language skills (and spelling in particular), dyslexia, a heritable condition, is sometimes responsible for perceptual-cognitive difficulties which greatly degrade spelling ability." [and not that poor spelling always signals that dyslexia is about...] >This same research shows that there is no correlation between spelling >and intelligence! >J. Ellis I did miss that posting. Hmm, research on the heritability of dyslexia included a correlational study of spelling and intelligence in the general population? Naturally, the details concerning the subjects and tests are essential before I could comment. So to be stubborn, I still stand by my prediction. Of course, we might both be right. Very simply, I may have a much wider sampling in mind than you: that's why I used the DMV example, to give an idea of the range of IQ's I was taking in: illiterate farmers, politicians, gang members, secretaries, mechanics, artists, gardeners. But rather than just reassert this prediction let me comment on just "how I figure." I figure on two "conditions" underlying this correlation. First, a broad trend toward greater literacy with increasing IQ naturally brings with it a greater command of the language. While errors on the words troublesome to all may differ little with IQ (we all have similar confusability tendencies) the lower frequency words might be a different story. Secondly, the lower end of the IQ spectrum is likely to contain more illiterates and extremely poor spellers than the upper end. Please note that I'm not making any causative, or moral claims here: in particular I'm not claiming that "intelligence" underlies spelling skill, the retention of spellings, etc. So while I still expect a CORRELATION between IQ scores and spelling, I am not proposing that spelling "reflects one's intelligence" or that spelling is a "feat of intellect." happy memorial ! tonyM
mmm@cup.portal.com (Mark Robert Thorson) (05/30/89)
It would be surprising if IQ did not correlate with spelling. Consider this
(quoted from BIAS IN MENTAL TESTING by Arthur Jensen):
"Intelligence test scores have been shown to have significant low to moderate
positive correlations with a variety of other variables, such as honesty
(Mussen, Harris, Rutherford, & Keasey, 1970); non-academic attainment in
extra-curricular activities (Kogan & Pankove, 1974); children's appreciation
of humor as judged from reponse to cartoons of varying subtlety and
sophistication (Zigler, Levine, & Gould, 1966); ability to solve anagrams
(Gavurin, 1967); untrained musical aptitude (Wing, 1941); speed of learning
a number of relatively complex (but not simple) motor skills (Noble, 1974);
susceptibility to certain optical illusions and various perceptual phenomena
(studies reviewed by Honigfeld, 1962); and amount of specific information
retained from viewing a television feature program, especially the incidental,
unemphasized bits of information (Nias & Kay, 1954)."
Jensen goes on to describe in detail a number of other correlates to IQ,
such as brain size, brain waves, height, obesity (negative correlate in
women), and near-sightedness. The latter has a particularly strong positive
correlation.
That high-IQ people are inherently superior to the rabble is clearly
demonstrated by the following quotations from the thoughts of Marilyn
Dos Savant (world's highest IQ person) printed in the May 28, 1989
San Jose Mercury-News (my sarcastic comments in []'s):
Q. Do you believe a person would want to know when he or she will die?
--Anonymous
A. I would. Can you imagine receiving an envelope in the mail that you
knew contained that information and NOT opening it? (Admittedly, you'd
sit and think down first. But just think of the sheer joy of knowing.
Why, you could take up PARACHUTING if you wanted to!)
[Why, you could put a gun to your head and pull the trigger if you wanted to!]
Q. Which is more important: love or freedom?
--Randall Jewell, Silver Spring, MD
A. Freedom. Otherwise, kids would never leave home.
[Love. Otherwise, there wouldn't be any kids in the first place.]
Q. Have you learned anything from being interviewed?
--Joseph Fiallo, Sarasota, FL
A. Don't insult the barber until after you've gotten up from the chair.
[I wonder what interviewer gave you that opinion? Obviously someone who
doesn't have the appropriate amount of awe for the world's highest IQ person
(and former member of the National Security Council during the Nancy Reagan
administration).]hollombe@ttidca.TTI.COM (The Polymath) (06/01/89)
In article <18956@cup.portal.com> mmm@cup.portal.com (Mark Robert Thorson) writes: }It would be surprising if IQ did not correlate with spelling. Consider this }(quoted from BIAS IN MENTAL TESTING by Arthur Jensen): } }"Intelligence test scores have been shown to have significant low to moderate }positive correlations with a variety of other variables, such as ... [ long list deleted ] So, because IQ has been found to correlate with some things, it should also correlate with spelling? Can you say "Hogwash!"? A second point is the difference between a "significant" correlation and a "large" correlation (a difference Jensen, at least implicitly, acknowledges). With a large enough sample size, you can make almost any correlation significant, even if it's only .0001 or less. "Significant but trivial" might be an appropriate description of such a situation. -- The Polymath (aka: Jerry Hollombe, hollombe@ttidca.tti.com) Illegitimati Nil Citicorp(+)TTI Carborundum 3100 Ocean Park Blvd. (213) 452-9191, x2483 Santa Monica, CA 90405 {csun|philabs|psivax}!ttidca!hollombe