the letters in their proper places. To give the code
form of a given letter without copy, however, makes a much heavier
demand on attention. Nearly all subjects find it necessary to trace the
code form, in imagination, from the beginning up to each letter whose
code form is sought. Subjects of superior intelligence, however,
sometimes hit upon the device of remembering the position of the
individual key letters e.g. (the first letter of each figure) from
which, as a base, any desired letter form may be quickly sought out.

The test correlates well with mental age, but for some reason not
apparent it is passed by a larger percentage of high-school pupils than
unschooled adults of the same mental level.

The code test was first described by Healy and Fernald in their "Tests
for Practical Mental Classification."[77] The authors gave no data,
however, which would indicate the mental level to which the test
belongs. Dr. Goddard incorporated it in year XV of his revision of the
Binet scale, but also fails to give statistics. The location given
the test in the Stanford revision is based on tests of nearly
500 individuals ranging from a mental level of 12 years to that of
"superior adult." It appears that the test is considerably more
difficult than most had thought it to be.

[77] _Psychological Review Monographs_ (1911), vol. XIII, no. 2, p. 51.

AVERAGE ADULT, ALTERNATIVE TEST 1: REPEATING TWENTY-EIGHT SYLLABLES

The sentences for this test are:--

(a) _Walter likes very much to go on visits to his grandmother,
because she always tells him many funny stories._
(b) _Yesterday I saw a pretty little dog in the street. It had
curly brown hair, short legs, and a long tail._

PROCEDURE. Exactly as in VI, 6. Emphasize that the sentence must be
repeated without a single change of any sort. Get attention before
giving each sentence.

SCORING. Passed _if one sentence is repeated without a single error_. In
VI and X we scored the response as satisfactory if one sentence was
repeated without error, or if two were repeated with not more than one
error each.

REMARKS. The test of repeating sentences is not as satisfactory in the
higher intelligence levels as in the lower. It is too mechanical to tax
very heavily the higher thought processes. It does, however, have a
certain correlation with intelligence. Contrary to what one would have
expected, uneducated adults of "average adult" intelligence surpassed
our