Another squirm about IQ

The Left hate the whole idea of IQ.  It offends against their nonsensical doctrine that all men are equal.  So when intelligence becomes a topic, they always do their best to denigrate and misrepresent it.  The article below arises from Trump's assertion that he has a higher IQ than Tillerson. It appeared in "LiveScience", which might as well be called "HalfDeadScience".

The whole aim of the article is to show that IQ score is not fixed and does not matter.  But in claiming those things they show how unscientific they are by not looking at the numbers.  Numbers are the inescapable tools of science.  And that matters.  Psychometricians are well aware that the correlations between different measures of ability are not perfect and that some situational factors can influence an IQ score.  But how strong are those influences?  Could the effect of situational factors be entirely trivial, for instance?

To answer that you have to look at the numbers that have emerged from research into IQ.  And they are revealing.  IQ tests are made up of a number of different types of puzzle that are not obviously related to one-another.  And the whole concept of IQ originated from the observation that some people are good at all sorts of puzzles that are not obviously related to one another.

So how strong is that effect?  When scores on the different tests are analysed a very strong first eigenvector arises, which shows that scores on all the different tests are strongly related to one-another.  Correlations between the various puzzles run as high as .70, which is a rare magnitude in psychological research.  So there is a single strong trait in existence that we call IQ and which tells us that a high scorer on an IQ test will be good at solving all sorts of problems.

So IQ is real and important.

What about the various influences described below that can influence an IQ score?  Again the numbers are instructive.  Nutrition, for instance, can have an effect.  A person eating a diet that is seriously deficient in important ways will get a reduced score -- but only by about 5 IQ points.  That is not negligible but it is mostly irrelevant in Western society.  Western diets generally do not harm IQ.  Reduced scores on dietary grounds are generally found in very poorly fed populations in India and Africa.  And IQs in Africa are so disastrously low that no feeding would bring them anywhere near European standards.

Let me look very briefly at some more of the influences trotted out below.  IQ correlates with Birth order.  Yes. It does appear to.  The research is not unanimous but that is probably because the effect is so small:  About 1 IQ point.

The Howard Gardner theory of "multiple intelligences" -- eight of them, would you believe? There is a very clear and simple demolition of the whole Gardner theory here -- which points out that the Gardner theory not only ignores the data but that its criteria for calling something "an intelligence" are so loose that sense of humour, sense of smell, musical ability, athletic ability etc could all be called "intelligences". By adopting similar rules I could say that all cats, dogs and horses are birds -- but that would still not make them so.

I could go on but will finish with one outright misrepresentation below.  An article titled "Self-Discipline Outdoes IQ in Predicting Academic Performance of Adolescents" is described below as showing that "IQ scores also change with the test taker's level of self-discipline and personal motivation and grit".  But the article did not show that.  It showed only that academic performance, not IQ, was influencible by grit etc.  That hard workers do better at school is hardly news but it does not vitiate the fact that High IQ scorers also perform well academically.

So the article below is an exercise in deception, not science


The IQ, or the intelligence quotient, is a measure of a person's mental age divided by their actual age, multiplied by 100. So, a person who is exactly as "mentally old" as one might expect for that individual's chronological age would score a perfectly average 100. People who deviate from that score in either direction are considered to be of above- or below-average intelligence. These scores can change with age and can fluctuate from one testing session to another, according to researchers.

But intelligence is a many-faceted beast. While it is colloquially associated with math and reasoning skills, psychologists assert that there are many kinds of intelligence, with Howard Gardner, a developmental psychologist at Harvard University, classifying seven distinct types, including bodily-kinesthetic, interpersonal, intrapersonal, musical, logical-mathematical, visual-spatial and linguistic.

Given that it's so hard to pin down exactly what intelligence is, the task of measuring it with a standardized test is particularly difficult, experts say. One of the standard IQ tests used today is called the Wechsler Adult Intelligence Scale (used for adults and older teens), which measures verbal and nonverbal cognitive skills, or as the psychologist who developed the test put it, the ability to "adapt and constructively solve problems in the environment."

Trump might not get the clear-cut result he's hoping for, since this test and others like it, including the Stanford-Binet test, don't present some unified quantity of a person's "smartness."

Test results are affected by several confounding variables, such as smoking habits, hours spent playing computer games and various aspects of one's personality, according to past research. IQ scores also change with the test taker's level of self-discipline and personal motivation and grit — all things that can change from testing session to testing session — according to a 2005 study that surveyed the IQ test results of 140 eighth-graders.

"Indeed, IQ tests are influenced by many factors," Cornell University developmental psychologist Stephen Ceci told Live Science. "For example, schooling affects IQ test performance," he added, explaining that for each year that a student falls short of finishing high school, there is a drop of between 1.8 and 4 IQ points compared to peers who did finish high school.

In Vietnam, Ceci explained, people who had a higher risk of being drafted stayed in school longer as a means to defer service compared to those with safer draft numbers. IQ testing revealed that those who stayed in school longer had higher scores — not because they were smarter, but because they had greater exposure to the conditions that would help them answer IQ test questions such as "who wrote Hamlet," Ceci said.

IQ test scores even correlate with birth order among siblings, according to two 2007 studies, as reported by The New York Times.

Therefore, IQ tests measure not just intelligence (however that is defined), but also the environment and context of one's life.

SOURCE

No comments:

Post a Comment

All comments containing Chinese characters will not be published as I do not understand them