if the pupil were normal. It gives the percentage of normality. The quotient thus derived will indicate whether the pupil has made normal progress for his age or whether he has progressed more rapidly or slowly than the average. Indirectly, at least, the educational quotient throws some light on the pupil's general intelligence. A high educational quotient usually signifies a high intelligence quotient, but not always. Well taught pupils with a low I. Q. may have relatively high educational quotients. When pupils are classified according to their native capacities and educational ages, we may then begin to judge the quality of the teaching more intelligently. A reasonable increment or interest on the mental capital invested will be all that is expected of teachers. Educational norms will serve as goals for teachers and pupils. Parents and supervisors will not expect increments out of proportion to the mental capital invested. On the other hand, teachers will know when pupils are doing an honest day's work, that is, when they are accomplishing a normal amount for their general intelligence and educational level.

26. Classifying Primary Grade Children by Mental Age [LOWELL, Frances, "An Experiment in Classifying Primary. Grade Children by Mental Age," Journal of Applied Psychology, September, 1922, Vol. 6, pp. 289–290.]

From the data obtained from this experiment in classifying primary children on the basis of mental age, the following conclusions have been suggested:

1. In grades where classification is not based upon mental age, one finds between the children of a single class as great a range in mental age as 53% years, and in intelligence quotients as great a difference as 79 points, yet these children are all trying to accomplish the same tasks and assimilate the same amount of subject matter.

2. Both mental ages and intelligence quotients must be considered in arranging classes. Children may have the same mental ages and yet differ so greatly in chronological ages that it would be impracticable for them to try to work together.

3. When mental age and brightness are used as a basis for classification, it is possible to separate the dull and feeble-minded children from the average and superior individuals, so that each may compete with others of his own mental calibre.

4. Slow groups should be formed from the first grade up, and a special curriculum worked out which will fit these children

to earn a living, since most of them never get beyond the sixth grade.

5. The superior children instead of skipping grades, may be given double work, which will bring them up in a semester or two, to the grade for which they are mentally ready, without the loss of any fundamental part of the school work.

6. Where first grade work is standardized upon the basis of the performance of average children six years old chronologically, no child should be allowed to enter the first grade before he has attained a mental age of six years. If he is chronologically six, and has been in the kindergarten two years, or if the school system does not provide the kindergarten training, he should be placed in a pre-primer or "preparatory-to-first-grade" class until he becomes six years mentally.

7. When the teaching and the curriculum are adequate, and when children are not exposed to the work of any grade until they are mentally qualified for it, a mental age classification will. give practically 100 per cent. of promotions in the primary grades.

27. Grouping Children on the Basis of Mental Capacity [COURTIS, Stuart A., "Data on Ability-Grouping from Detroit," TwentyFourth Yearbook, Part 2, pp. 141-147. Bloomington, Public School Publishing Co., 1925.] (Adapted.)

Since the World War, many school systems have attempted to solve the problem of adjustment to individual differences by classifying children on the basis of intelligence test results. Will such grouping accomplish its object? Can children be classified thus into homogeneous groups, and will they remain so during the course of instruction?

Studies made in Detroit and elsewhere prove conclusively that when provision is made for individual progress, children in each group move forward at rates which are very different. Professor Courtis believes that complete individualization of instruction, whenever knowledge or skill outcomes are sought, is the only adequate solution of the problem of adjustment to individual differences and that grouping of pupils on the basis of mental test results is merely an expedient, not a final solution.

Achievement tests furnish definite proof that children do not fall into permanent natural ability groupings that can

be taught as a class of homogeneous ability in all subjects. The consolidated results from twenty-six schools are shown in tabular form:

MEASUREMENT OF CONSTANCY OF ACHIEVEMENT GROUPINGS IN INITIAL TESTS, FEBRUARY, 1924, AND IN FINAL TESTS, JUNE, 1924. NUMBER OF CHILDREN IN EACH ACHIEVEMENT GROUP AND THE NUMBER OF DIFFERENT POSITIONS IN WHICH THEY ARE FOUND.

(Subjects: 162 X Children, 251 Y Children, 140 Z Children. Total 553. Grade 4-B. Based on Scores in Reading, Writing, Arithmetic, Spelling, and Composition.)

In brief, the term's work resulted in smaller number of changes in the number of different positions held by the X's in different subjects, but greatly increased the spread of the Y's and especially the Z's. Taking all groups into consideration, about 80 per cent. of the pupils are classified in from three to four different achievement groups with respect to

the standards in the five subjects. The greater the ability of the pupil, the larger the number of positions he will hold with reference to the standards in the different subjects. Schooling tends to increase rather than decrease the diversity. "However, individualization of instruction, through practice test materials based upon the unit-task, individualprogress idea, provides a solution which is both administratively feasible with large masses of children and educationally efficient." (p. 147.)

28. The Growth of Intelligence

[DOLL, Edgar A., "The Growth of Intelligence," Psychological Monographs, 1921, Vol. 29, No. 2, pp. 127-128.]

Conclusions as to the growth of intelligence are in part as follows:

Intelligence is defined in this study in terms of its essential activities: judgment, comprehension and reasoning. The growth of intelligence is here measured by comparing successive Binet mental ages of the same individuals obtained by repeated examinations over long periods of time. . . .

For practical purposes an empirical straight-line curve of average normal growth in terms of relative units of mental age may be employed as a standard of comparison. . . .

The most recent and extensive evidence suggests that the average adult level of intelligence is between 13 and 14 years instead of being 16 years as has been recently accepted in the field of clinical psychology. . . . The rate of growth of superior children between the ages of 9 and 13 years life age is negatively correlated with mental age and degree of brightness. Individual variability in the rates of intellectual growth of superior children is very marked.

No conclusion is established regarding the age of arrest of intellectual growth among superior children. . .

The contention of several authorities that the intelligence quotient is approximately constant between the ages of 4 and 16 years of life age for all degrees of brightness is not justified either by theoretical considerations or by the experimental evidence now available in the literature of the subject..

The I. Q. instead of being approximately constant is markedly variable in individual cases. The likelihood of an I. Q. re

maining approximately constant can be expressed in terms of probability with a fair degree of accuracy.

Marked irregularities in the rates of growth of intelligence among individual children of feeble-minded, borderline, and superior children are demonstrated by growth curves covering long periods of time. The data are insufficient to indicate the periods at which intellectual growth is retarded or accelerated.

29. Interpretation and Application of Intelligence Quotient [FREEMAN, F. N., "The Interpretation and Application of the Intelligence Quotient," Journal of Educational Psychology, January, 1921, Vol. 12, p. 12.]

There seems to be evidence of considerable weight that the typical intellectual growth follows a uniform rate, within at least a period covered roughly by the elementary school. We are here, of course, assuming that the measure of mental ability is a composite one, which gives opportunity for the exercise of a variety of functions, and for a wide range of scores ranging from considerably below the ability of the youngest children tested to considerably above the ability of the oldest. The evidence seems further to indicate that the individuals above and below the median develop at about the same rate, though in different levels as compared with the median child.

30. The Intelligence of Pupils Who Repeat

[EATON, H. T., "The Intelligence of Pupils Who Repeat," School and Society, February 3, 1923, Vol. 17, pp. 139–140.]

By a previous study the writer determined that the failure of the pupils who left Central High School, Syracuse, N. Y., during the year of 1920-21 was exactly two and one-half times that of the pupils who remained in school during the year. .

In an attempt to ascertain the reason for somc of the failures, the writer conducted the following study during the spring term of 1922. . . . On May 23, 1922, the Terman group test was given to 91 pupils who were repeating various subjects (Latin, French, algebra, geometry, etc.). At the same time, the test was given to nine pupils who were known to be above the average in their school standing. The scores of the group test were translated into mental ages and the intelligence quotients determined. . . . The writer would emphasize the fact that Central High School offers no courses in vocational work, so that most of the pupils take a college preparatory or a general