In this article we will discuss about how to measure intelligence.
In 1904, when psychology was just emerging as an independent field, members of the Paris school board approached Alfred Binet with an interesting request- Could he develop an objective method for identifying the children who, in the language of that era, were described as being mentally retarded, so that they could be given special education? Binet was already at work on related topics, so he agreed, enlisting the aid of his colleague, Theodore Simon.
In designing this test Binet and Simon were guided by the belief that the items used should be ones children could answer without special training or study. They felt that this was important because the test should measure the ability to handle intellectual tasks—not specific knowledge acquired in school.
To attain this goal, Binet and Simon decided to use items of two basic types- ones so new or unusual that none of the children would have prior exposure to them, and ones so familiar that almost all youngsters would have encountered them in the past. Children were asked to perform the following tasks- Follow simple commands or imitate simple gestures; name objects shown in pictures; repeat a sentence of fifteen words; tell how two common objects are different; complete sentences begun by the examiner.
The first version of Binet and Simon’s test was published in 1905 and contained thirty items. Much to the two authors’ pleasure, it was quite effective: With it, schools could readily identify children in need of special help. Encouraged by this success, Binet and Simon broadened the scope of their test to measure variations in intelligence among all children. The revised version, published in 1908, grouped items by age, with six items at each level from three to thirteen years. Items were placed at a particular age level if about 75 percent of children of that age could pass them correctly.
Binet’s tests were soon revised and adapted for use in many countries. In the United States, Lewis Terman, a psychologist at Stanford University, developed the Stanford-Binet test a test that was soon put to use in many different settings. Over the years the Stanford-Binet has been revised several times. One of the features of the Stanford-Binet that contributed to its popularity was the fact that it yielded a single score assumed to reflect an individual’s level of intelligence the now famous (some would say infamous) IQ.
1. IQ: Its Meaning Then and Now:
Originally, the letters IQ stood for intelligence quotient, and a “quotient” is precisely what the scores represented. To obtain an IQ score, an examiner divided a student’s mental age by his or her chronological age, then multiplied this number by 100. For this computation, mental age was based on the number of items a person passed correctly on the test: Test takers received two months’ credit of “mental age” for each item passed.
If an individual’s mental and chronological ages were equal, an IQ of 100 was obtained; this was considered to be an average score. IQs above 100 indicated that a person’s intellectual age was greater than her or his chronological age—in other words, that the individual was more intelligent than typical students of the same age. In contrast, numbers below 100 indicated that the individual was less intelligent than her or his peers.
Perhaps you can already see one obvious problem with this type of IQ score: At some point, mental growth levels off or stops, while chronological age continues to grow. As a result, IQ scores begin to decline after the early teen years! Partly because of this problem, IQ scores now have a different definition.
They simply reflect an individual’s performance relative to that of persons of the same age who have taken the same test. Thus, an IQ above 100 indicates that the person has scored higher than the average person in her or his age group, while a score below 100 indicates that the person has scored lower than average.
2. The Wechsler Scales:
To overcome this and other problems, David Wechsler devised a set of tests for both children and adults that include nonverbal, or performance, items as well as verbal ones, and that yield separate scores for these two components of intelligence. However, he developed these tests at a time when the multifaceted nature of intelligence was not yet well understood, and it is not clear that Wechsler’s various subtests actually do measure different aspects of intelligence.
Despite such problems, the Wechsler tests are currently among the most frequently used individual tests of intelligence. An overview of the subtests that make up one of the Wechsler scales, the Weschsler Adult Intelligence Scale—Revised (WAIS—3 for short) is presented in Table 11.1.
Wechsler believed that differences between scores on the various subtests could be used to diagnose serious psychological disorders. However, research on this possibility has yielded mixed results at best.
A Wechsler test for children, the Wechsler Intelligence Scale for Children (WTSC), has also been developed; it too is in widespread use. Patterns of scores on the subtests of the WTSC are sometimes used to identify children suffering from various learning disabilities.
Some findings indicate that children who score high on certain subtests, such as Picture Completion and Object Assembly, but lower on others, such as Arithmetic, Information, and Vocabulary, are more likely to suffer from learning disabilities than children with other patterns of scores. Once again, however, not all findings point to such conclusions, so the value of the WISC (now in its third revision, WISC—3) for this kind of diagnosis remains somewhat uncertain.
One more point- Tests such as the WAIS are designed to measure the ability to acquire new information or skills, an ability implied by our definition of intelligence. Psychologists describe such tests as aptitude tests, and scores on these tests are often used to predict future performance. In contrast, achievement tests are designed to measure what you have already learned; for example, the tests you take in this class are designed to be achievement tests.
The distinction between these two types of tests is not precise, however. For instance, many aptitude tests (including college entrance examinations) include measures of vocabulary, which, to a degree, reflect past learning. But in general it is useful to think of aptitude tests as ones designed to predict future performance and achievement tests as ones that reflect current performance and the results of past learning.
3. Individual Tests of Intelligence: Measuring the Extremes:
Individual tests of intelligence such as the Stanford-Binet or WISC are costly. They must be administered one-on-one by a psychologist or other trained professional. Why, then, do these tests continue in widespread use? The answer is that these tests have several practical uses and provide benefits that help to offset these costs. The most important of these uses involves identification of children at the extremes with respect to intelligence those who are mentally challenged (i.e., who suffer from some degree of mental retardation), and those who are intellectually gifted.
The term mental retardation refers to intellectual functioning that is considerably below average combined with varying degrees of difficulty in meeting the demands of everyday life. As shown in Figure 11.4, persons with mental retardation are typically described according to four broad categories of retardation: mild, moderate, severe, and profound.
Individuals’ level of retardation is determined by at least two factors: their test scores and their success in carrying out activities of daily living expected of persons their age. As you can guess, persons whose retardation is in the “mild” category can usually learn to function quite well.
What causes mental retardation? In some cases it can be traced to genetic abnormalities such as Down syndrome, which is caused by the presence of an extra chromosome; persons with Down syndrome usually have IQs below 50. Mental retardation can also result from environmental factors, such as inadequate nutrition or use of drugs or alcohol by expectant mothers, infections, toxic agents, and traumas resulting from a lack of oxygen during birth. Most cases of mental retardation, however, cannot readily be traced to specific causes.
Intelligence tests have also been used to identify the intellectually gifted—persons whose intelligence is far above average. A study on exceptionally bright persons first begun by Lewis Terman in 1925. This study has followed the lives of about 1,500 children with IQs of 130 or above to determine the relationship between high intelligence and occupational success and social adjustment.
As a group, these gifted persons have experienced high levels of success. They have earned more degrees, attained higher occupational status and salaries, experienced better personal and social adjustment, and were healthier, at each age, than the average adult.
4. Group Tests of Intelligence:
Both the Stanford—Binet and the Wechsler scales are individual tests of intelligence. They are designed for use with one person at a time. Obviously, it would be much more efficient if group tests could be administered to large numbers of people at once. The need for such tests was driven home at the start of World War I, when the armed forces in the United States suddenly faced the task of screening several million recruits.
In response to this challenge, psychologists such as Arthur Otis developed two tests; Army Alpha for persons who could read and Army Beta for persons would could not read or who did not speak English. These early group tests proved highly useful. For example, the tests were used to select candidates for officer training school, and they did accurately predict success in such training.
Psychologists interested in studying intelligence have moved beyond tests such as the Stanford-Binet and Wechsler scales in an attempt to try to identify the basic cognitive mechanisms and processes that underlie intelligence—and that enable people to score high on intelligence tests. This work has led to two major developments. First, several tests have been constructed that are based on the findings of cognitive psychology and on our growing understanding of many aspects of cognition.
Among these the most noteworthy are the Kaufman Assessment Battery for Children and the Kaufman Adult Intelligence Test, and the Woodcock-Johnson Test of Cognitive Abilities. The Woodcock-Johnson Test, for instance, attempts to measure important aspects of both fluid and crystallized intelligence.
Second, a growing body of research has focused on the finding that the speed with which individuals perform simple perceptual and cognitive tasks (processing speed) is often correlated with scores on intelligence tests. For example, significant correlations (on the order of -.30 to -.40) have often been found between various measures of reaction time (one measure of processing speed) and scores on intelligence tests.
Another and even more promising cognitive measure of intelligence is a measure known as inspection time. It reflects the minimum amount of time a particular stimulus must be exposed for individuals to make a judgment about it that meets some pre-established criterion of accuracy.
To measure inspection time, psychologists often use procedures in which individuals are shown simple drawings and are asked to indicate whether the longer side occurs on the left or right. Participants are not told to respond as quickly as possible. Rather, they are instructed to take their time and to be accurate. Inspection time is measured by the time they take to make such decisions at a pre-specified level of accuracy—for example, 85 percent.
What does inspection time measure? Presumably, the amount of time individuals require for the intake of new visual information. Supporters of this measure argue that this task perceiving new information is basic to all higher-level mental operations in human thought.
Further, they note that this measure is closely related to current theories of perception and decision making, theories emphasizing that new visual information is perceived in discrete samples and then combined into judgments such as “I see it” or “I don’t see it.” Growing evidence indicates that inspection time is indeed closely related to intelligence, as measured by standard tests. In fact, inspection time and scores on such tests correlate -.50 or more.
Can we trace individual differences in intelligence to differences in neural functioning? The answer suggested by a growing body of evidence is yes. Such research suggests, first, that nerve conduction velocity—the speed with which nerve impulses are conducted in the visual system—correlates significantly with measures of intelligence (e.g., the Raven Progressive Matrices test).
Other, and related, research has examined metabolic activity in the brain during cognitive tasks. Presumably, if intelligence is related to efficient brain functioning, then the more intelligent people are, the less energy their brains should expend while working on various tasks. This prediction, too, has been confirmed; the brains of persons scoring highest on written measures of intellectual ability to expend less energy when these individuals perform complex cognitive tasks.
Finally, some findings suggest that there is a link between brain structure and intelligence. Specifically, scores on standard measures of intelligence such as the WAIS are related to the size of certain portions of the brain, including the left and right temporal lobes and the left and right hippocampus. Moreover, this is true even when corrections are made for individuals’ overall physical size.
In sum, it appears that the improved methods now available for studying the brain and nervous system are beginning to establish the kind of links between intelligence and physical structures that psychologists have long suspected to exist. Such research is very recent, so it is still too soon to reach firm conclusions.