Scoring. Simply + or −. No attention is paid to faults of pronunciation.
Remarks. There is unanimous agreement that this test belongs in the 3-year group. Although the child has not had as much opportunity to learn the family name as his first name, he is almost certain to have heard it more or less, and if his intelligence is normal the interest in self will ordinarily cause it to be remembered.
The critic of the intelligence scale need not be unduly exercised over the fact that there may be an occasional child of 3 years who has never heard his family name. We have all read of such children, but they are so extremely rare that the chances of a given 3-year-old being unjustly penalized for this reason are practically negligible. In the second place, contingencies of this nature are throughout the scale consistently allowed for in the percentage of passes required for locating a test. Since (in the year groups below XIV) the individual tests are located at the age level where they are passed by 60 to 70 per cent of unselected children of that age, it follows that the child of average ability is expected to fail on about one third of the tests of his age group. The plan of the scale is such as to warrant this amount of leeway. But even granting the possibility that one subject out of a hundred or so may be unjustly penalized for lack of opportunity to acquire the knowledge which the test calls for, the injustice done does not greatly alter the result. A single test affects mental age only to the extent of two months, and the chances of two such injustices occurring with the same child are very slight. Herein lies the advantage of a multiplicity of tests. No test considered by itself is very dependable, but two dozen tests, properly arranged, are almost infinitely reliable.
III, 6. Repeating six to seven syllables
Procedure. Begin by saying: “Can you say ‘mamma’? Now, say ‘nice kitty.’” Then ask the child to say, “I have a little dog.” Speak the sentence distinctly and with expression, but in a natural voice and not too slowly. If there is no response, the first sentence may be repeated two or three times. Then give the other two sentences: “The dog runs after the cat,” and, “In summer the sun is hot.” A great deal of tact is sometimes necessary to enlist the child’s coöperation in this test. If he cannot be persuaded to try, the alternative test of three digits may be substituted.
Scoring. The test is passed if at least one sentence is repeated without error after a single reading. “Without error” is to be taken literally; there must be no omission, insertion, or transposition of words. Ignore indistinctness of articulation and defects of pronunciation as long as they do not mutilate the sentence beyond easy recognition.
Remarks. The test does not presuppose that the child should have the ability to make and use sentences like these for purposes of communication, or even that he should know the meaning of all the words they contain. Its purpose is to bring out the ability of the child to repeat a six-syllable series of more or less familiar language sounds. As every one knows, the normal child of 2 or 3 years is constantly imitating the speech of those around him and finds this a great source of delight. Long practice in the semi-mechanical repetition of language sounds is necessary for the learning of speech coördinations and is therefore an indispensable preliminary to the purposeful use of language. High-grade idiots and the lowest grade of imbeciles never acquire much facility in the repetition of language heard. The test gets at one of the simplest forms of mental integration.
Binet says that children of 3 years never repeat sentences of ten syllables. This is not strictly true, for six out of nineteen 3-year-olds succeeded in doing so. All the data agree, however, that the average child of 3 years repeats only six to seven syllables correctly.
III. Alternative test: repeating three digits
Procedure. Use the following digits: 6–4–1, 3–5–2, 8–3–7. Begin with two digits, as follows: “Listen; say 4–2.” “Now, say 6–4–1.” “Now, say 3–5–2,” etc. Pronounce the digits in a distinct voice and with perfectly uniform emphasis at a rate just a little faster than one per second. Two per second, as recommended by Binet, is too rapid.