The Measurement of Intelligence

Dec 21, 12 The Measurement of Intelligence

To address the issues surrounding today sulphurous intelligence, or simply to help them understand what is the famous “IQ” (intelligence quotient), a small detour through the intricacies of measuring intelligence can be useful.

The Measurement of Universal Intelligence Test

The Measurement of Universal Intelligence Test

The first intelligence test (the Binet-Simon) was published in 1905 by Alfred Binet, following a request from the National Education. His goal was to predict academic success. At every age corresponded to a series of questions and tasks usually successful with children of this age. For example, 3 years children are usually able to repeat two digits, and 7 years they are able to remember 5. Thus, the ability to repeat 5 digits is characteristic of 7 years. Such considerations can define, for a given child, “mental age” is the typical age profile corresponding to its responses.

The Measurement of Intelligence Test

The Measurement of Intelligence Test

Around 1912, the German psychologist William Stern had the idea of dividing the mental age by the chronological age for which took the name of Intelligence Quotient, and that usually expresses pourcents1. For a given age, the distribution of IQ forms a bell-shaped curve (Gaussian), a mean of 100 and standard deviation of about 15, which means that the “normal” or usual lie between 85 (100 – 15) and 115 (100 +15).

Wechsler scales

Water has flowed under the bridges, the intelligence theories have bubbled and many other scales were invented. The most widely used at present in the world is due to David Wechsler. The first version of his intelligence test for adults, WAIS (Wechsler Adult Intelligence Scale) was published in 1939. This was followed ten years later by a version for children 6 to 16 years, the WISC (Wechsler Intelligence Scale for Children) and children under 6 years, the WPPSI (Wechsler Preschool and Primary Scale of Intelligence).

Subtests of the WISC

The Measurement of Intelligence

The Measurement of Intelligence

The various events of the WISC IV (latest version) are:


Vocabulary (define words, such as “light”)

Understanding (explain social rules, such as “why we dress in the morning”)

Cubes (reproduced with two-colored cubes construction presented in an image)

Identification of concepts (select a set of images forming a concept – for example, must be grouped trees – the common concept is not provided)

Matrices (the beginning of a series of figures is given, and we must find the next)

Memory numbers (again, at the place or to a series of numbers)

Sequences of numbers and letters (numbers and letters repeat the heard, but in alphabetical order)

Code (code a series of numbers)

Symbols (symbols identify a series of lines)

Measurement of Intelligence

Measurement of Intelligence

Wechsler tests provide an overall standard, defined so that the average value is 100 and the standard deviation of 15, to resemble the old IQ. This result is still called an IQ although no longer the result of a division. Contrary to earlier measures Binet IQ and Stern, the latest version of the WISC can distinguish different types of intelligence, corresponding to four major factors2, but also gives a total IQ, which is an approximation of the g factor of Spearman.

The test consists of various tasks (or items) grouped into tests or subtests. An event usually has thirty items. In one of the tests (similarities), the child must find that two objects or properties in common. An item of this test is for example “how blue and red look like? “. The expected response is “what are the colors.” The similarity test has 23 questions of this type.

Each event provides a note.

From these values, we calculate the scores corresponding to four main factors:

Verbal comprehension factor is calculated from the similarity tests, vocabulary, comprehension. It can be interpreted as an index of general understanding of words, concepts, or rules.

Factor includes perceptual reasoning tests cubes, concept identification, and matrices. It can be interpreted as a score of reasoning in situations where visual perception is important.

Working memory factor is calculated from tests memory for numbers and sequences of numbers and letters. Psychologists call working memory resource hypothetical mental allows both to retain short term aggregate information (short term memory) and processing this data. In the test sequences of numbers and letters must both remember the letters and numbers you hear and do processing on the data: it is therefore working memory (and not just short-term memory) that it is.

The processing speed factor finally gathers the last two races and code symbols, where is the number of cases in a short time to calculate the bill.

One can imagine three levels of analysis results in such a test. At the coarsest level, we do not focus only IQ total average of the four scores. We will have a useful idea, but approximate capacity of the child all the tasks considered. At the second level, as advocated by Wechsler, we can build a more accurate mental skills by separating what is comprehension, memory, speed, and visualization. Finally, a more detailed level would be to analyze the events one by one.