Chapter 8. What Is An Intelligence Test?

The content of the I Q tests comes out of the experiences children normally have. It is based upon our culture, in the broadest sense of the word. More and more, psychologists are recognizing that children whose experiences and backgrounds differ do not have equal chances for success on I Q tests.

In recent years a number of psychologists have concentrated on developing what were first called "culture-free tests" and are now called "culture-fair tests." The aim of these tests is to minimize the influence of culture. Since it is not possible to construct tests that are "free" of cultural influences, the objective is to make them "fair" to children with different social, economic, and educational backgrounds.

Attempts to create such tests are an indication of the growing dissatisfaction among psychologists and educators with the limitations of all the standard IQ tests. However, these tests have had only a limited development up to now and are not in general use. Therefore it is the content of standard I Q tests that we shall examine.

Intelligence testing is a relatively recent development in the history of human measurement. It appeared and grew as a direct result of the growth of public schools, the greatest part of which has taken place during the twentieth century. The first intelligence test was devised by Alfred Binet, a French psychologist who, interestingly enough, was looking for a means of distinguishing between bright and dull students so that Parisian schools might be more efficiently organized. The Binet-Simon test was created in 1905 and served as the point of departure for the entire intelligence-testing movement. In the half century since, the number of tests that have been developed is almost staggering.

There are two broad categories of intelligence tests, individual tests and group tests. Both claim to measure the same thing, but there are important differences between them. The Binet-Simon test was an individual test, as were all the well-known revisions of this pioneering effort. The Terman Revision of 1916, the Herring Revision of 1922, the Stanford Revision of 1937, and others are nothing more than modifications and adaptations of the French test.

Other individual tests have more or less departed from the Binet-Simon original, being based on somewhat different theories. But the similarities of the overwhelming majority of these tests are much more striking than their differences.

It is important to note that when psychologists speak about intelligence tests they mean individual tests unless they specifically mention group tests. Their theoretical discussions are always based on what is measured by an individual test. However, the most widely used and most practical tests are group tests, which differ from individual tests in several important ways. Most children are tested by group tests and are educationally identified by an IQ score which is an even less accurate index to intelligence than the individual tests scores. And even more than the individual scores, the group IQ score is subject to modification by changes in environment and experience.

Individual tests are given to one child at a time by a trained psychologist or tester. It takes approximately one hour to administer most individual tests. These are made up of a series of subjects, some of which are verbal tests, others "performance" tests.

A performance test is one in which the subject is asked to do or manipulate something, while the verbal test is one which deals with use of language, either oral or written.

Following are samples of some of the subtests in the 1937 Stanford-Binet test, probably the most widely used individual test today.

Year II

1. Place three small blocks into appropriate holes.
2. Point to toys as they are named.
3. Build a four-cube tower after being shown how.
4. Point to parts of a paper doll as they are named.
5. Name common objects shown in different pictures.
6. Use a two-word sentence spontaneously.

Year III

1. Manipulate small toys in response to simple commands.
2. Name common objects shown in different pictures.
3. Point to the longer of two sticks.
4. Name at least three objects shown in a picture.
5. Point to objects identified by use.
6. Tell what should be done in common situations.

Year VI

1. Define five words orally.
2. Make a bead-chain pattern from memory after being shown how.
3. Identify the missing parts of pictured objects.
4. Select a given number of blocks from a pile.
5. Identify one of five pictured objectswhich does not belong with the others.
6. Draw a line through the shortest path in a simple maze.

Year X

1. Define eleven words orally.
2. Explain the foolishness of the pictured activities of a person.
3. Read a 48-word passage and recall a large part of it from memory.
4. Give two reasons justifying an oral statement.
5. Name as many disconnected words as possible in one minute.
6. Repeat digits after hearing them.

An important part of the Stanford-Binet test is a vocabulary subtest, consisting of one hundred words chosen arbitrarily from an 18,000-word standard dictionary and administered and answered orally. This vocabulary test is considered by Terman and his associates, who devised the Stanford-Binet test, to be the most valuable subtest of all, one that in and of itself practically measures what the entire test is supposed to measure.

The Herring Revision places much more emphasis on verbal factors than does the Stanford-Binet. The following is an outline of the subtests, divided into five groups each of which consists of its own subtests plus those of the preceding groups; the examiner may omit any subtest that the subject will obviously pass or f aiL

Group A

1. Name common objects shown in different pictures.
2. Tell what two numbers should come next in a sequence.
3. Read a passage and recall a large part of it from memory.
4. Repeat backwards digit groups of from two to nine numbers.

Group B

5. Show knees, fingers, ear, foot.
6. Repeat sentences of six and seven syllables.
7. Point to the larger figure in three pairs of figures.
8. Aesthetic discrimination using four pairs of faces.
9. Name black, gray, and white.
10. Solve six problem situations.
11. Read a passage; 17 "memories."
12. Define seven abstract words.
13. Read a passage; 12 "memories."

Group C

14. Solve five problem situations.
15. Point out absurdities in eight statements.
16. Make four three-word sentences.
17. Give rhymes for four words.
18. Identify similarities in six groups of three things each.
19. Interpret five proverbs.
20. Read a passage; 13 "memories."
21. Read three disarranged sentences.
22. Solve three arithmetic problems.

Group D

23. Repeat four sentences of from ten to 13 syllables each.
24. Directions test.
25. Directions test.
26. Find the similarities in four groups of three things each.
27. Generalize from four related statements.
28. Explain two verse passages.
29. Sentence completion.
30. Read and solve problems.

Group E

31. Identify five common objects.
32. Compare different forms.
33. Perform three commands.
34. Solve diagram problem.
35. Repeat digits in various series.
36. Repeat three sentences of from 19 to 24 syllables each.
37. Identify proportional relationships.
38. Code writing.

Two well-known and widely used individual tests that are not revisions of the Binet-Simon type are the Kuhlmann Tests of Mental Development and the Wechsler Intelligence Scale for Children. The Kuhl-mann test includes the following subjects:

1. Sitting with support
2. Naming objects shown.
3. Naming objects from memory.
4. Recognizing missing parts in pictures.
5. Repeating numbers.
6. Vocabulary.
7. Finding analogous words.
8. Drawing inversions of upright forms.

The Wechsler test is probably the best example of a test which departs from the Binet type of test. It includes the following subtests:

1. General information.
2. General comprehension.
3. Verbal arithmetic problems.
4. Digit repetition.
5. Word similarities.
6. Picture completion.
7. Picture arrangement.
8. Block design.
9. Object assembly.
10. Digit-symbol pairing.

There are, of course, many other tests which differ somewhat from those just described; but these are among the most widely used and should suffice to show how individual test results may be influenced by experience and learning.

Group tests, more widely and easily applied than individual tests, rely almost entirely on verbal factors to measure intelligence. Some, however, do include subtests which deal with number ability; others include nonverbal items dealing with recognition of forms and their relations to each other. Instead of being examined individually, the test subjects are given printed question booklets, with spaces provided for answers or with separate answer sheets which can be mechanically scored. The questions are usually all of the multiple-choice type; that is, a question is posed and four or five possible answers are given; the subject must choose the best answer from among the possibilities.

In schools, group tests are usually administered by teachers who have been given only the most rudimentary instruction in testing procedures. Almost all group tests have a precise time limit for each sub-test, and each subtest is constructed so that most, if not all, children will be unable to complete all the questions within the given time limits.

Among the most widely used group intelligence tests are: Army Alpha, Terman Group Test of Mental Ability, Pintner General Ability Test, Otis Group Intelligence Scale, Otis Quick-Scoring Mental Ability Test, National Intelligence Test, Henmon-Nelson Test of Mental Ability, Primary Mental Abilities Test, Thorndike's CAVD Scale, and Haggerty Intelligence Examination Delta 1 and Delta 2. These tests all differ somewhat, but their content on the whole is strikingly similar. Here is a random sampling, with examples of each type of question:

1. Verbal analogies: Gun is to shoot as knife is to (fly, glove, meat, hurt, cut).
2. Verbal opposites: Hot is the opposite of (warm, heavy, cool, damp, cold).
3. Vocabulary: Distant means the same as (seldom, far, old, dark, space).
4. Disarranged sentences: are and apples long thin (true, false).
5. Logical reasoning: A room always has (rugs, chairs, floor, people, windows).
6. Common-sense problems: If plants are dying for lack of rain, you should (water them; ask a florist's advice; put fertilizer around them; keep them warm; reseed them).
7. General information: Diamondsare obtained from (mines, reefs, elephants, oysters).
8. Arithmetic problems: How many are ten men and four men? (11, 18, 14, 7, 9).
9. Number sequence: 2 4 6 8: what number should come next? (9, 7, 10, 12, 6).
10. Classification: Choose the word that does not belong with the others (cow, fox, rose, squirrel, tiger).

In addition to these most common types there are frequently subtests based on sentence completion, anagrams, proverb interpretation, picture completion, narrative completion, and the like. On the nonverbal level, there are sometimes geometric figures which must be related and digits and symbols to be paired.

Now the average child's success in any of these tests will depend to a large extent on his training and background, on his direct experience with the materials that make up the content of the test or with materials that are very similar. In the nonverbal performance subtests-that is, those tasks that do not involve verbal responses-even a limited amount of experience with similar objects and materials could make a tremendous difference in any child's ability to place blocks into appropriate holes in a board, to build a four-cube tower, to make a bead chain, to assemble objects from parts, and to make a certain design with blocks.

Of course in those performance tests that involve the use of language-naming toys, the missing parts of pictures, the parts of a doll, the use of certain objects, and so on-the child's success will also depend directly upon whether he has ever seen these objects and been taught their names and functions.

Why so many verbal subtests, and why the overwhelmingly verbal character of all group tests?

Many psychologists feel that the extent to which the verbal skills are developed in a child is usually a fairly accurate measure of the child's ability. Therefore they consider the heavy reliance on verbal factors in so many tests to be justified. Of course other psychologists maintain that measuring verbal factors is to a large extent a measurement of differences in environment and experience, particularly in respect to language.

Let's consider the verbal skills for a moment. There are four of them: listening, speaking, reading, and writing. Each is on a different level of complexity, listening being the most simple and writing the most complex. We all know from experience that people differ widely in their abilities to do these four things. Some people do none of them well, some people do all of them well and some people do only some of them well. For example, there are people who can listen and speak reasonably well, but cannot read or write; they are called illiterate. Others can listen, read, and write well, but cannot speak well. Still others can speak, read, and write well, but are poor listeners. And there are still others who listen, speak, and read well, but have trouble writing.

If intelligence were the only significant factor in determining how well people could utilize and develop verbal skills, then ability in some of these four skills should imply ability in all of them. This is especially true when a person is skilled in the more complex of the four. In any case, these four skills are all closely related; they are all aspects of an individual's ability to use the symbols of language in communication with other people.

If the degree to which the ability to communicate has been developed is higher in one verbal skill than in another, then it seems that this must reflect differences in environmental opportunities to develop these skills. Obviously, the man who can speak well but cannot read or write is, in most instances, simply a man who has never been taught to read or write. For those who have been taught but have not been able to learn, there usually exists a special problem which often has no connection with innate intelligence or ability.

How do we master the skills of language? As infants we listen, learn to distinguish sounds and to identify them with certain meanings. Then we imitate, unintelligibly at first, the sounds we have listened to. Next we learn to identify written or printed symbols with words we have listened to and spoken. And finally we imitate the symbols we have learned to read.
The child whose parents do not talk to him much is generally handicapped by a limited environment; parents who do not encourage their child to speak, who do not listen and who do not stimulate verbal communication, are providing that child with an impoverished language background. In turn, this handicap may well affect the child's ability to learn to read and write. In all likelihood he will enter school at a disadvantage that will not be apparent and that will be mistakenly considered to be an inherent lack of ability. He will probably be the recipient of an education geared to such a lack of ability, and this education in turn will further limit his opportunities to develop what by then probably has been buried under an avalanche of negative experiences and attitudes. This child, the product of a verbally impoverished home, will again and again be measured by verbal means in order to determine what he is capable of learning and doing in life.

Admittedly, this example may be extreme, but it is by no means uncommon. However, it is enough to recognize that there are wide differences in the verbal experiences of children, and to understand that it is completely within the province and the capabilities of parents to affect those differences. Since a child is evaluated to such a large extent on the basis of his verbal skills, and since his development, particularly in school, is to be so seriously affected by that evaluation, parents have a responsibility to see that the child's experience and environment will increase those skills as much as possible.

There is little doubt that a child's vocabulary can be increased in the home, and that this alone would probably help increase his ability to succeed in an IQ test. There is also little doubt that the speed with which a child works the various types of questions and problems found in I Q tests can be increased by training and practice. Since almost all group tests are essentially speed tests, added speed will result in higher scores.

There are exceptions to these generalizations, and we consider them next.