nake progress as rapidly as he can.

The results which were measurable, amely, improvement in rate and quality of writing, are shown below by tables and raphs. To make them more easily comarable with each other and with results nother cities, the rate and quality scores were changed to B-scores. That is, each core was reduced to its grade equivalent. or example, a quality score of 50 is the tandard score for the fifth grade the core made by the average fifth grade child

in the United States as a whole-and scores of 50 were therefore re-expressed as 5. Similarly, a quality score of 54, the sixth grade standard, would be re-written as 6; a rate score of 55, the standard for the fourth grade, would be re-written as 4, and so on. In this scheme, 4 means the ability usually attained at the end of the fourth grade. The standards used for this purpose are those printed on the Gettysburg edition of the Ayres Scale.

Table I shows that the pupils beginning e 4A grade in February wrote only as pidly as the average child who has cometed three months' work beyond the end the third grade. (Since there are ten onths in our school year, the tenths may è taken as months.) In June the rate of ese pupils had fallen to that of the avere child six months beyond the end of e second year of instruction.

The quality of their writing in February as below that of the average child finishg the first grade, but in June it had risen that of the average child finishing the urth grade. Their loss in rate during this mester was the amount usually gained seven months but their gain in quality as more than that usually made in three id a half years. The 4A group using this

method of writing in September was found to write at first only as rapidly as the average child finishing the second grade, but in January wrote with the rate characteristic of a child finishing the third grade; the quality of their writing in September was that of the average child two months beyond the end of the first grade, but in January it was that of the average child seven months beyond the end of the fourth grade; their gain was thus 1.1 years in rate and 3.5 years in quality during this semester. The varying fortunes of the children in the higher grades (of the four schools combined) are shown on the next eight lines of Part 1 of the table. The results secured in the rooms using the Palmer system are shown in Part 2 of the first table.

The average gains in all grades together are shown in Table II. The rooms using the Courtis system are here seen to make a gain, when the two semesters are averaged together, of .2 years in rate and 2.8 years in quality, in a half year. The few rooms using the Palmer system made a much larger gain in rate (1.4 years) but a slightly smaller gain in quality (2.5 years). Different ways of combining rate and quality are shown in the columns to the right-first, with both counting equally, then with the quality score counting twice as heavily as the rate, then three times, and so on.

Courtis Palmer Difference

Since the relative importance of rate and quality is unknown, the reader car take his choice as to which of the columns he considers most significant. As rate has never been counted in judging handwriting proficiency in Chicago, the quality score alone will probably be considered important by most persons. Whether rate can safely be neglected, is discussed later. In comparing the Palmer and CourtisShaw systems as to rate of writing, it should be remembered that these children had written for several years in the Palmer manner and that most of them had written only a half year in the other manner.

Av.

No. Feb.- No. Sep.- for Pupils June Pupils Jan. Year 1059 .1 910 .3 .2 130 1.9 190 1.0 1.4 -1.2

With reference to the value of the Palmer system, it is also relevant to raise the question as to whether it has succeeded in Chicago in keeping the handwriting up to the standard attained elsewhere. The schools used in this experiment are schools considerably above the average Chicago school. They are in general excellent schools. Yet in relation to the average results for the country as a whole, as shown in Figures I and II, these four representative schools stand pretty far down. The initial standings, as seen in the February and September entries on the graph, are well below the standard, both in rate and quality. (The school years or grade levels are shown along the left margin of the drawing. Each grade, if up to standard, would have begun a semester just one-half year below the norm, shown on the graph, for that grade.) In quality of writing, these schools, at the beginning of the two semesters were, on the average 1.7 years behind the quality attained in the average city, and in rate 1.0 year.

Another important consideration is the effect of handwriting instruction upon writing in other studies. Two of the

-Quality

Av.

2.8

2.5

.3

1.95

upon

this

principals were consulted point. Both asserted with emphasis that the carry-over is much greater in the Courtis-Shaw writing, and agreed that the reason is the fact that in this type of writing the pupil is left much freer to learn his own method of writing The appearance of the writing developed by different pupils in a given room, all at the same stage as to legibility, is very ditferent-each has worked out a style naural to him. The effect of this is that whe writing is done outside the handwriting class, the child does not have to learn a new type of writing, as is so often true case of children taught by the Palmer method. The type of writing natural this individual is used constantly, and un form quality is maintained in all his wors The development of self-direction, se control, and self-appraisal is also said b these principals to have a striking effec upon the work of the pupils in other studies.

STANDARDS AND NORMS

A standard, as Freeman pointed out. must be based upon three considerations the needs of an adult, the total amour