Binet and Simon went about constructing such an age scale by observing the kinds of Neo-Scholastic things most children knew or could do at different ages, and then making up a lot of short questions and simple tasks that would incorporate these ordinary kinds off general knowledge and skills. They did next want to assess a child’s knowledge of what was specifically taught in school. Their idea was to obtain a much broader assessment of the child’s abilities, which they could then compare with his performance in school. The many questions and tasks that Binet made up were tried out on representative groups of children of different ages. Binet recorded the percentage of children with each one-year age interval from age 3 to age 15 who could ” pass” each item. He noted those items that showed the most clear-aft age differences in percent passing and assigned them to the specific age differences in percent passing and assigned them to the specific age levels for which they were the most discriminating.
The items thus form an age scale in terms of which a child’s total score on the test can be meaningfully expressed. A child who passes as many items as the average number passed by all five-year olds is said to have a mental age of 5, regardless of what the chronological age happens to be. A child who has a mental age of 5 and is, in fact, 5 years old is said to have average ability for his age. But a five year old with a mental age of 3 is considered retarded. Such a child could fail to keep his age-mates in the very first year of school. Binet and Simon thought that “nearly all the phenomena with which psychology concerns itself are phenomena of intelligence: sensation, perception and intellectual manifestations as much as reasoning.
” To them, the French word ‘intelligence’ meant something other than what we mean by intelligence today. But they selected a fundamental part of intelligence and that was the part they were trying to measure with their tests. They wrote, “It seems to us that in intelligence there is a fundamental faculty, the alteration or the lack of which is of the utmost importance for practical life. This faculty is judgement, otherwise called good sense, practical sense, initiative, the faculty of adopting oneself to circumstances. To judge well, to understand well, to reason well, these are the essential activities of intelligence. The rest of the intellectual capacities seem of little importance in comparison with judgement.
” “What does the word ‘jealous’ mean?” (Stanford Binnet), measures intelligence, we must know whether people of the age for which the test is designed can define the world ‘jealous’ in the way the designer of the test regards as correct, without their parents’ vocabulary or their schooling or other factors apart from “real” intelligence playing a major role. This means that although the psychologists have no precise common definition of intelligence, they have a common ides of the kind of thing intelligence is. When they select problems to measure intelligence they are acting on this common approximation to what intelligence is. In practice this means that the designers of intelligence tests collect problems similar to those which occur in existing tests.
Such problems are said to have face validity. But sometimes someone designs a problem of a type which has never occurred before. For instance, a psychologist devised an entire intelligence test on the problem “Draw a picture of a person.” and many problems in modern tests were once quite novel types. Where this kind of problem is under study, we consider that a high correlation between the results of the new problems and results of well known intelligence tests guarantees that they are problems requiring intelligence.
In order to be fully satisfactory the correlation must be 0.80. The correlation between Stanford-binet and WAIS, is 0.85- 0.
95 in large heterogeneous groups. However, most intelligence tests have a mush lower correlation with each other-somewhere between 0.50 and 0.85. It is not clearly understood that an intelligent test is not intended to measure an aptitude for passing intelligence tests. In that case no criteria of validity would have been needed. The question “What validity has this test?” would be answered quite simply by studying the problems in the test.
The test is intended to measure attribute intelligence. Opinions differ as to how important it is to have an exact definition of intelligence. In America psychologists have decided that they can do useful work by measuring intelligence without defining it. In this respect, they are doing the same as physicists did when they were studying heat. Long before the physicists had agreed on a reasonable definition of heat; they had already invented thermometers with which they could measure changes in temperature and with these instruments they could measure changes in a position to discover many important physical laws.
In America, intelligence researchers have a common idea of what the test results they want to produce should be like. And in broad outline the following is a definition of intelligence: (1) The test results from a large group people, representative of a whole population, must be divided up in the same way as body measurements. The group must not contain proportionately more of any particular group than the population as a whole. This means if IQ units are used (other kinds of markings are generally used now but the principle is the same) that about 68% of the group must have I.
Q ratings between 85 and 115. About 13% must have scores between 70 and 85, the same number between 115 and 130 marks and 2-3 % will have an IQ score of less than 70 or more than 130. (2) There must be no difference between the average test results of men and women of the same age. (3) The IQ must not show any increase with chronological age. But it must decrease slightly after the age of 20 to between 50 and 60, and thereafter more sharply. (4) When compared, the average test results must reveal clear differences between different social groups. The lower social and economic groups must have lower marks than the higher ones.
(5) The test results must show a clear and close relationship with school results and results in university studies and professional work.