TABLE 13.-Frequency of the ratios between the number of boys and girls retarded one year or more and the total number enrolled in 1908 and 1918.

The reduction in the percentage of over-age pupils since 1908 is pictured in Tables 10, 11, 12, and 13, and in figures 12, 13, 14, and 15. In every instance in the diagrams the mode has shifted to the left for both boys and girls. The general reduction in the percentage of overage pupils is conducive to better teaching. In some of the cities these over-age pupils have been segregated from the younger pupils of their grades and placed in special schools and classes. The subtle suggestion incident to a study of this kind is that cities which have not provided special work for retarded or subnormal pupils should do so if the proportion of over-age pupils justifies such a venture. In view of the norms established in this study, each city superintendent should carefully analyze the status of the pupils in his school system to determine whether special classes should be provided for the over-age children. A study of special schools and classes for the subnormal children is found in another chapter of this Biennial Survey of Education.

FIG. 15.-Distribution of 80 cities according to the percentage of pupils retarded one year or more in 1908

and 1918.

OVER-AGE AND UNDER-AGE PUPILS, BY GRADES.

The degree of retardation or acceleration is not more important than the location of the irregularity in grade distribution. Accordingly, the under-age and over-age pupils in each city have been retabulated to show in what grades retardation and acceleration accumulate. The results of this study for 1918 are summarized by means of arithmetical averages in figure 16.

30

FIG. 16.-Average percentage of pupils over age and under age in 80 cities in 1917-18.

The

Corresponding data for 1908 had not been ascertained. smallest percentage of over-age pupils is found in the first grade, and the highest percentage in the fifth grade, where over 30 per cent of the pupils are retarded. In the seventh and eighth grades the percentages decrease rapidly. This decided drop is undoubtedly due to the decreasing effectiveness of compulsory attendance laws at the ages of 13 and 14. Where extraneous factors do not operate decreases or increases are quite regular. In high school, for instance, the decreasing percentage of over-age pupils is strikingly constant. The curves representing the under-age pupils takes almost a straight course, varying only slightly and always in an opposite direction to the course taken by the curve representing over-age pupils. In

Under

other words, as one curve rises the other falls and vice versa. agedness seems to accumulate with each grade above the fourth, each successive grade contributing its quota to the accelerated groups. The fact that the curves cross at the first and eighth grades indicates that the age-grade groups previously chosen are the most typical of the age-grade distribution. To have chosen the ages 5 and 6 as normal for the first grade would have increased materially the percentage of over-age pupils and decreased almost to zero the percentage of under-age pupils. To have chosen the ages 7 and 8 as normal for this grade would have had the opposite effect of increasing the percentage of under-age pupils and decreasing the percentage of over-age pupils. Similar conditions would have prevailed in the other grades if corresponding disturbances in the grade groups had been made.

TABLE 14.-Distribution of 80 cities according to the percentage of pupils under age in their respective grades, 1918.

TABLE 15.-Distribution of cities of 25,000 population and over according to the percentage of pupils under age in their respective grades, 1918.-GROUP A.

High-school years.

TABLE 16.-Distribution of cities of less than 25,000 population according to the percentage of pupils under age in their respective grades, 1918-GROUP B.

74-75.

Median..

10

13524122

923121

6-7..

8-9.

10-11.

12-13.

14-15.

3

2

31-35.

36-37.

38-39.

40-41.

42-43.

2