Dear Michael and others
The supposed steady rise in intelligence over generations
(the Flynn effect), the stability of the 15-point gap and
the season of birth effect have all been interpreted in line
with hereditarian and maturational ideas about intellectual
growth. The psychological establishment has not come up with
a plausible explanation for any of these phenomena.
Sternberg's statement that "research by Stephen Ceci and
others has shown that IQ increases as a function of
schooling" is too imprecise to convince. For example an
increase in staying on longer at school could very well
lower raw score levels as students with a wider range of
ability to answer the test questions stay on to higher
levels of education.
(1) At least three published reports from three different
countries based on OTIS type test answers from thousands of
school pupils show that it is grade level and not age which
is connected to the raw scores. There is no rise for those
groups at the same grade level and within the age expected
for the grade. Raw scores rise from grade to grade when the
tests are administered to successive grades at the same time
of the year.
(2) The IQ is derived from a belief in maturation. Test
developers assume that scores rise by increments of 3 or 4
months. This can be seen in conversion tables. The result is
an IQ scale which affects individual children within the
same grade. Younger children have their raw score rank
raised while older children in the same grade have theirs
(3) Over the last century there has been a decrease of age
at grade level. Not because kids are getting smarter but
because of changes in population pressure on schools,
regulations on class size and acceptance of social
promotion. All system effects.
Some years ago I recalculated the scores obtained from a
version of the OTIS which Jim Flynn had claimed showed a
year of intellectual growth over 32 years. Recalculating the
scores according to the demographic annual returns from
schools showed that most if not all of the score change
could be accounted for by demographic change at grade level.
Published in the Oxford Review of Education in 1998.
There may be other factors involved in score levels. For
example, whether a test is timed or untimed or group or
individual, but whether the issue is the stability of a gap,
the effect of the seasons, or score rise, demographic
factors and scoring systems should not be overlooked.
School of Education Studies
Victoria University of Wellington NZ
----- Original Message Follows -----
> My difficulty with Sternberg's response is that as soon as
> he posits that there is such a thing as reified
> intelligence that can be measured through some
> standardized test (whether it is limited or not) he is
> arguing on Murray's ground. I.Q is a construct that we
> developed to prove Murray's thesis, not what Sternberg is
> trying to say. Too much advantage is given to a eugenics
> based perspective by accepting I.Q. tests as legitimate
> ways of measuring some object that we call intelligence.
> From: firstname.lastname@example.org on behalf of David
> Preiss Sent: Wed 1/17/2007 10:45 AM
> To: eXtended Mind, Culture, Activity
> Subject: Re: [xmca] In case you missed it
> Dear colleagues,
> Please see below a letter sent by Robert Sternberg to WSJ
> as an answer to Murray's piece.
> Charles Murray's "Intelligence in the Classroom" is an
> article by a non-scientist filled with serious
> distortions and misunderstandings of the current state of
> scientific theory and research on intelligence.
> First, Murray is roughly correct in the assertion that
> "Half of all children are below average in intelligence."
> This is true in the same sense that half of today's
> children are below the median (not average) in height, or
> below the median age of the population. But median
> heights have risen greatly over the past several
> generations, as have median age spans. Indeed, research
> by James Flynn shows conclusively that median IQs have
> risen as well since 1900. Contrary to the tone of
> Murray's comments, most of the increase is due to the
> so-called general factor, not to other factors. So
> Murray's comments regarding possibilities for educational
> achievement make no sense. A child of today with an IQ
> of 100 would have been scored as having a substantially
> higher IQ 100 years ago. Given that the increase in IQs
> has been about 9 points per generation, that person would
> have had an IQ in excess of 127 at that time, which would
> have led to educational predictions very different from
> Murray's doom-and-gloom predictions. Similarly, a
> 6-footer today is not much above average and would not be
> considered particularly tall, whereas 100 years ago, he or
> she would have been looked at as exceptionally tall.
> Second, IQ is NOT a "ceiling," and I don't know of any
> responsible psychologist who believes it is. IQ gives
> rough prediction of a child's school performance, as does
> socioeconomic status, motivation, and any other number of
> variables. But none of these variables sets a ceiling on
> children's performance. First, they are all highly
> imperfect predictors--success is multi-factorial. Second,
> they are subject to error of
> measurement. Third, they are not etched in stone.
> Research by Stephen Ceci and others has shown that IQ
> increases as a function of schooling, and that it is the
> schooling that is responsible for the increase, not the
> other way around.
> Third, the temporary effects of interventions to increase
> intelligence are in large part because the interventions
> themselves are temporary and usually extremely
> short-lived. If you have a child living in extreme
> poverty, in a challenging and possibly dangerous
> environment, and with parents who are not in a position to
> provide the best possible education for
> their children, it is not surprising that short
> interventions--the kinds most easily funded by
> grants--are difficult to maintain. Consider an oft-made
> analogy to exercise. You can exercise to improve your
> muscles. But if you stop exercising, your muscles revert
> to what they were before. The same is true of your
> intelligence, and research by Carmi Schooler and others
> shows precisely that.
> Fourth, it is fallacious to believe that brain development
> is etched in stone. Research by William Greenough, Marian
> Diamond, and others has shown that learning changes the
> brain--permanently. Experience matters for brain
> development. Charles Murray had the good fortune to be
> exposed to experiences that children in many parts of the
> United States and elsewhere never will have. Indeed,
> children growing up in war zones often need to devote all
> their resources just to staying alive. They cannot have
> the kind of schooling that optimizes their scores on the
> tests of which Mr. Murray is so fond.
> Fifth, our own peer-reviewed, published research has shown
> that broader measures of abilities--based on the
> "multiple intelligences" that Murray disdains--can
> substantially improve prediction of academic success at
> the college level at the same time that they reduce
> ethnic-group differences. These assessments do not replace
> traditional measures--they supplement them. They are not
> "refutations" of the existence of the analytical skills
> measured by tests of general ability, but rather,
> demonstrations that such measures are relatively narrow
> and incomplete in their measurements of abilities. These
> conventional tests measure important skills, but not the
> only skills that matter for academic and other forms of
> success. Indeed, teaching to a broader range of abilities,
> our research shows, also can significantly improve school
> achievement over teaching that is more narrowly focused.
> In sum, Murray's column gives a false and misleading view
> of the state of research on intelligence. I blieve
> responsible scientists will not take it seriously.
> Unfortunately, many laypeople will not be in a position
> to realize that the statements are seriously misleading
> and paint a picture of research on intelligence that does
> not correspond to reality.
> Robert J. Sternberg
> (Robert J. Sternberg is Dean of the School of Arts and
> Sciences and Professor of Psychology at Tufts University.
> Previously, he was IBM Professor of Psychology and
> Education and Professor of Management at Yale University
> and President of the American Psychological Association.)
> On Jan 16, 2007, at 5:51 PM, J. Mark Jackson wrote:
> > This article ran in today's WSJ. The link below takes
> > you directly to the full article without registration.
> > Scary, very scary!
> > http://www.aei.org/publications/pubID.25452
> ,filter.all/pub_detail.asp >
> > Mark
> > _______________________________________________
> > xmca mailing list
> > email@example.com
> > http://dss.ucsd.edu/mailman/listinfo/xmca
> David Preiss, Ph.D.
> Profesor Auxiliar / Assistant Professor
> Pontificia Universidad Catolica de Chile
> Escuela de Psicología
> Av Vicuña Mackenna 4860
> Macul, Santiago
> Fono: 3544605
> Fax: 3544844
> e-mail: firstname.lastname@example.org
> web personal: http://web.mac.com/ddpreiss/
> web institucional: http://www.uc.cl/psicologia
> xmca mailing list
> xmca mailing list
> [Attachment: winmail.dat]
xmca mailing list
This archive was generated by hypermail 2b29 : Thu Feb 01 2007 - 10:11:33 PST