Philadelphia, stimulated by the work of Rice, carried on a series of tests in spelling by the composition method, extending from June, 1896, to June, 1898. In 1903 he published the results of this investigation in a volume entitled Spelling in the Elementary School. Though Cornman's data were carefully gathered and methodically tabulated, the results which he obtained were little more satisfactory than those of Rice, in that he counted all words in the composition and determined the ratio of the spelled words and misspelled words in terms of per cent. He not only counted the recurring words which were spelled correctly but the recurring misspelled words as well. This accounts for his percentages being lower than those reported by Rice. The work of Rice and Cornman stimulated many young men to begin work in educational research. Edward L. Thorndike, who has since become the wizard of the objective standard, wrote in the Forum in 1905 as follows: "The study of education is beginning to be quantitative, we are becoming properly disgusted with the one-sided booking which only takes account of dollars spent and neglects the debit side, the income in knowledge, habits, power, zeal, and ideals. This ambition toward an exact objective measurement of the results of educational endeavor is a symptom of healthy scientific fervor and also of common-sense wisdom. No one possessed of science or sense will deny the value of successful quantitative study of school work." Ayres's Spelling Scale.1 - All of the words which compose⚫ the Ayres Spelling Scale were selected on the basis of their frequency in the letters from some 7000 children in eighty1 Distributed by the Russell Sage Foundation, New York City. four widely separated cities, in newspaper articles, and in selections from standard literature. The manuscripts examined aggregated 368,000 words from 2500 different writers. The one thousand words most frequently used were selected for the Ayres Spelling Scale. These words were grouped into fifty lists of twenty words each. The difficulty of these groups was tested by giving them to the children of a number of cities. After the first test the words were regrouped into fifty lists of twenty words each and again presented to the children of the grades. On the basis of these two tests the original one thousand words were arranged into twenty-six groups. The systematic arrangement of these twenty-six groups, with the difficulty each offers expressed in per cent, constitutes the well-known and popular Ayres Spelling Scale. The Buckingham Spelling Scale.1 The author of this scale chose fifty words from five spelling books, each of which was found in at least two of the spelling books. These words were arranged into two lists of twenty-five each by means of a statistical method. In the final arrangement the lists are consistently progressive in difficulty from the third grade to the eighth. The relative difficulty of each word is likewise determined. The comparatively few words employed in this test limits its practical value. Starch's Spelling Scale.2 - Starch has a spelling scale which is composed of six lists each of which contains one hundred words. These words were chosen from Webster's New International Dictionary. The plan consisted of 1 Prepared by Prof. R. B. Buckingham, University of Illinois. choosing the first defined word on the even-numbered pages of the dictionary and of the arrangement of these in the order of the number of letters composing them. Jones's Spelling Demons.1 - Jones has listed the one hundred words which he found most frequently misspelled by children. He calls this group the "One Hundred Spelling Demons of the English Language." Teachers in the upper grades should drill the pupils upon these words until they are mastered. Method of Teaching Word Lists. - Good sense as well as good method prompts one to determine the relative difficulty of the words listed for the term or year by giving to the class without previous preparation all of the words listed. These words should then be arranged in the order of their relative difficulty. The relative difficulty of the words in any list is indicated by the number of persons who misspell each. WRITING STANDARDS Thorndike's Handwriting Scale. The first satisfactory result, from a practical point of view, of all the agitation for quantitative standards of measurement occurred in 1910. In that year Thorndike's Scale for Judging Handwriting appeared in the Teachers College Record. Referring to this scale, Ayres says, "The credit of developing the first measuring scale for handwriting belongs to Professor Edward L. Thorndike of Teachers College, Columbia University. The publication, in March, 1910, of his Handwriting Scale constituted a most important contribution not only to experimental pedagogy, but to the entire movement for the scientific study of education." 1 Child's Own Spelling Book, Jones. In reference to the need of such a scale, Thorndike said: "At present we can do no better than estimate a handwriting as very bad, good, very good, or extremely good, knowing only vaguely what we mean thereby, running a risk of shifting our standards with time and only by chance meaning the same by a word as some other student of the facts means by it. We are in the condition in which the students of temperature were before the discovery of the thermometer, or any other scale for measuring temperature beyond the very hot, hot, warm, lukewarm, and the like, of subjective opinion." The Thorndike scale is composed of the handwriting of children in grades five to eight inclusive. The writing from these grades was grouped into eleven groups on the basis of quality. The quality of the groups is represented by figures 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, and 17 respectively. Quality 7 represents the poorest samples taken from grade five, and quality 17 represents the best samples taken from grade eight. The steps of difference between the qualities were equal in the sense of being called equal by from twenty-three to fifty-five competent judges. This means that 14 is as much better than 13 as 13 is than 12; that 13 is as much better than 12 as 12 is better than II, and so on; that quality 14 is two times as far above zero merit in handwriting as quality 7. The scale includes quality 18, which was taken from a copy book, and qualities 4, 5, and 6. Samples 5 and 6 were taken from the fourth grade and sample 4 was manufactured for the purpose of extending the scale below the merit of fourth-grade children. In the November issue of the Teachers College Record for 1914, Thorndike presents a more fully developed hand writing scale. This edition is in many respects a marked improvement over the 1910 edition. The wide variation of form presented in this standard makes it exceedingly valuable in evaluating the legibility of various types of handwriting. Teachers who habitually think of quality in terms of grades can, for all practical purposes, easily transfer the qualities of the Thorndike scale into grades, by multiplying the numbers of the scale by 5.8. Those who have measured the merit of handwriting with this or the Ayres scale will not be content to judge the merit of writing in terms of personal experience. Ayres's Handwriting Standard. - In November, 1911, Leonard P. Ayres of the Russell Sage Foundation began a preliminary experiment to determine the relative legibility of different samples of handwriting. He early concluded that quality of legibility is measurable, and proceeded to perfect a writing scale with that in mind. His first printed scale appeared in February, 1912. In discussing the merits of this scale he says: "The method by which the present scale has been produced, and the criterion on which it rests as a basis, differ radically from those adopted by Professor Thorndike. The difference in the basis is that in the present case legibility has been adopted as a criterion for rating the different samples in place of 'general merit' used as the basis of the Thorndike scale. The change substitutes function for appearance as a criterion for judging handwriting.” Ayres gathered 1578 samples of writing from forty school systems. The samples were read by ten readers, each of whom by means of a stop watch recorded the exact number of seconds required to read each sample. The samples |