Iq

The start of it all - the first intelligence test

The first intelligence test along modern lines was created by the Frenchman Alfred Binet in 1905. Binet had been set the task by the Parisian school authorities of devising a test that would select those children who were unlikely to learn much from being in ordinary schools, so that they could then be given special education. Binet's test gave different questions to children of different ages and was based on their general knowledge, and their ability to reason and solve problems. His test consisted of some 30 items, ranging from the ability to touch parts of one's face to more abstract concepts. It was sufficiently successful that it correlated well with teachers' estimates of children's ability such that those who scored highly were judged to be 'bright', whereas those who did poorly were judged 'dull' or retarded. A modern version of Binet's test is still much used today - the Stanford-Binet test.

Mental age and intelligence quotient (IQ)

Binet introduced the concept of mental age (MA) which can be defined as an individual's level of mental ability relative to others. If a child with a real or chronological age (CA) of 5 years succeeded at problems usually solved by 7-year-olds her MA would be 7 while her CA is 5 and the child is judged to be bright. Conversely, if a 5-year-old succeeded only at the level of a 3-year-old his MA is below average and he is likely to have learning difficulties at school.

A few years later (1912) William Stern introduced the term intelligence quotient (IQ), and in its original formulation it was simply calculated as a child's MA divided by the child's CA multiplied by 100:

IQ = MA x 100

CA

We can see from this formulation that those children who are exactly average for their age have an IQ of 100; if MA is below CA then the IQ is below 100; if MA is above CA then the child is bright and IQ is above 100.

Intelligence tests

There are four important things to note about IQ tests and IQ scores. [1] The simple formula given above is no longer used, but the purpose of IQ tests is always to compare people's (children's or adults') scores with those from people of the same population and of approximately the same age. [2] The average IQ at a given time is always 100. To ensure this it means that tests are carefully standardized every few years to ensure that the population varies around this mean. This means that test makers provide a conversion chart so that an individual's raw score (i.e., the number of items passed) can be expressed as an IQ score. [3] Children's and adults' raw scores tend to increase from one generation to the next, hence the need for regular standardization of tests - a fuller discussion of this, and the possible reasons for the changes, is given in Chapter 1. [4] The items on IQ tests invariably proceed from the simple to the complex, so that an individual's raw score (and hence their IQ) is derived from the number of items passed before they make mistakes.

There are now several IQ tests that are widely used, and hundreds of tests of specific abilities. Three well-known tests are the following: [1] the Stanford-Binet for the ages of 2 to adulthood; [2] the Wechsler scales (The Wechsler Preschool and Primary Scale of Intelligence - WPPSI - for ages 4 to6 Ð... ; the Wechsler Intelligence Scale for Children - WISC - 6 to 16; The Wechsler Adult Intelligence Scale - WAIS; [3] The Differential Ability Scales - DAS - from infancy to adolescence.

What is intelligence - one ability or several?

To a large extent how intelligence is defined determines how it is measured. We can probably agree that intelligence involves verbal abilities, memory, problem-solving skills, and the ability to adapt and change to meet life's demands. Unfortunately, however, the agreement stops there! The concepts of mental age and IQ suggest that intelligence is a single general ability, and there are those who argue that a general intelligence ability (often referred to simply as 'g') underlies performance on all intelligence tests. Others suggest that intelligence is made up of a number of specific abilities or subskills. Still others have argued that performance on intelligence tests is unrelated to our ability to 'live our lives intelligently'. One commentator has remarked: "Tests have very modest correlations with performance in skills that society deems important" (Deese, 1993, p. 113).

In line with this view Steve Ceci and Jeff Liker (1986) tested 30 "avid racetrack patrons" for their ability to use a sophisticated multiplicative model to handicap races - an important ability if you make a living from horse races! They found no relationship (correlation) between this ability and IQ (the handicappers' IQs ranged from 83 to 130), leading them to the conclusion that "IQ is unrelated to real-world forms of cognitive complexity" (p. 255). In additional research Steve Ceci found little relationship between IQ and income, leading him to the reasonable conclusion that "it's better to be born rich than smart"!

Nevertheless, we have known for many years that occupation is related to IQ: teachers, doctors, accountants, pharmacists, lawyers, and those in similar occupations have a mean IQ above 120; people in semi- or low-skilled occupations such as barber, farmhand, laborer, have a mean IQ below 100. Perhaps we can conclude that intelligence, as measured by intelligence tests, is