A ^- rv ~ —* 2 >!] Q — O *'l Q ^^^^= 5> 1 iYFACII COO 7 =^ ■■■■■■■■■■inHi 5814 P8D4 descriptive Southern Branch of the University of California Los Angeles Form L I HARVARD BULLETINS IN EDUCATION NUMBER V SOUTHERN BRANCH, DIVERSITY OF CALIFORNIA, LIBRARY, 4-0S ANGELES, CAUF, This Bulletin, the second issued under the series-title " Harvard Bulletins in Education," continues the series begun as " The Harvard-Newton Bulletins." It is therefore numbered V. The volumes heretofore issued are: I. The School System as an Educational Lab- oratory. II. Scales for the Measurement of English Com- position. III. Bridging the Gap: The Transfer Class. IV. A Selected Critical Bibliography of Voca- tional Guidance. HARVARD BULLETINS IN EDUCATION A DESCRIPTIVE BIBLIOGRAPHY OF MEASUREMENT IN ELEMENTARY SUBJECTS BY HENRY W. HOLMES, Professor of Education, Harvard University G. S. BEAN, Locke School, Arlington, Mass. C. H. BEEK, West Concord Grammar School, Concord, Mass. H. D. BIXBY, State Normal School, Lowell, Mass. O. K. COLLINS, Superintendent of Schools, Hingham, Mass. ROY DAVIS, Mechanic Arts High School, Boston, Mass. S: B. PAUL, Mason School, Newton, Mass. R. E. PINKHAM, Danversport Grammar School, Danvers, Mass. H. T. PRARIO, Hancock School, Lexington, Mass. PUBLISHED BY HARVARD UNIVERSITY CAMBRIDGE, MASS. • COPYRIGHT, igi 7 HARVARD UNIVERSITY OO IO.\ to FOREWORD This Bibliography is neither complete nor authoritative. It probably contains, however, most of the important titles in its field, and the descriptive comments, since they were written by responsible and experienced persons engaged in educational work, may be taken as fairly representative reactions on the books and articles listed. As editor, I have unfortunately been obliged to revise the lists during an absence from Cambridge which has taken me far from my own sources of information, so that I have been unable to supply omissions or correct mistakes with any certainty. For such errors as appear the responsibility must therefore be mine; the merits of the bibliography are due to the solid labor of my coworkers. It is our expectation to continue this bibliography yearly for an indefinite period; and we hope in future issues to correct and amplify the lists. It will be of great service in this work if any one who discovers an omission, misstatement, or inadequate description will inform the editor, i The movement for educational measurement is now well advanced. The theory of tests and scales is fairly well worked out, and a considerable number of valuable tests and scales have been produced; what is most needed now is the application and criticism of these products in the class room. It is time for teachers and principals to take a more constructive part in the movement. We hope this bibliography will help them to do so. The Bibliography is limited to the field of elementary educa- tion and to measurements in the subjects actually taught in school. It contains no references to psychological tests, and VI FOREWORD refers to general surveys only when these include reports of the use of tests and scales in subjects. Mr. Roy Davis and Mr. H. D. Bixby have contributed to the publication of the Bibliography more largely than the other members of the Seminary by which it was prepared, and have generously relieved me of the labor of proof reading and of certain final revisions; the thanks of the whole Seminary are due them for their careful and continuous effort. Henry W. Holmes. CONTENTS I. The Theory of Tests and Scales. Discussions of the Movement for Educational Measurement 3 II. Reports on the Use of Tests in Various Subjects; Lists of Tests and Scales 9 III. Reading 12 IV. Penmanship 19 V. Spelling 24 VI. Composition 28 VII. Grammar 33 VIII. Arithmetic 34 IX. Geography 41 X. Drawing 41 XI. Addresses 42 Index 45 A DESCRIPTIVE BIBLIOGRAPHY OF MEASUREMENT IN ELEMENTARY SUBJECTS BIBLIOGRAPHY OF MEASUREMENT IN ELEMENTARY SUBJECTS THE THEORY OF TESTS AND SCALES. DISCUSSIONS OF THE MOVEMENT FOR EDUCATIONAL MEASUREMENT Ayres, Leonard P. Measuring educational processes through educational results. School Review, Vol. xx, No. 5, pp. 300- 309, May 1912. Discusses the change in attitude of educators toward scientific study of education in the past fifteen years. The author says this has been forced by exactions of this scientific age. Education has had to fall in with the movement for efficiency. Advance has been made from the examination of method to the study of the product. This study has led to reforms in administration to the benefit of the schools. Evidence has taken the place of " authority." Surveys have resulted from the demand for evidence and standards for measurement have been pro- duced. This measurement of results is scientific method applied to education. The measurement of educational processes and products. Conference on Educational Measurements, The Extension Division of Indiana University, Bloomington, Ind., pp. 127-133, April 16 and 17, 1915. Discusses the method of formulating and standardizing a series of tests, with special reference to the Ayres Spelling Scales. This is fol- lowed by questions from those present at the conference. An article of value to students engaged in constructing tests and measures. It shows one method of standardizing tests. Bagley, W. C. The need of standards for measuring progress and results. Proceedings, National Education Association, 191 2, pp. 634-639. Shows that standards produce economy in five ways: (1) by pro- viding the teacher with a definite goal toward which to direct his 3 4 BIBLIOGRAPHY OF MEASUREMENT efforts, (2) by furnishing the pupil with a powerful incentive, (3) by- giving supervisors a means of detecting weaknesses, (4) by providing a means of testing new methods, (5) by furnishing a means of deter- mining with a nicer degree of accuracy the growth in efficiency of a school system. Discusses the imperative need of reducing teaching experience to a trustworthy basis, and so formulating the experience that it may be readily assimilated and applied by beginners, and asserts that the construction of standards offers a most effective means of meeting this need. Shows the two types of standards already estab- lished: (1) those represented by the exact definitions resulting from the work of the Carnegie Foundation and the North Central Associa- tion of Colleges and Secondary Schools; (2) those represented by the work of Binet, Courtis, Thorndike, and Ayres. Points out lines of future progress, and shows the dangers lurking in the present interest in the construction of standards, and the necessity for settling down to patient, painstaking, systematic investigation. Ballou, Frank W. Improving instruction through educational measurement. National Education Association Bulletin, Vol. iv, No. 6, pp. 196-203, May 1916. States the aim of the department of educational investigation and measurement in Boston, enumerates the steps in the process of im- provement of instruction, and describes the present status of educa- tional measurement in that city. It contains a very clear and com- prehensive account of the work done with the Courtis Tests, and in spelling, geography, penmanship, and addition of fractions. Especial attention is given to the use of the Courtis Tests in measuring improve- ment in arithmetical results. The tables and observations concerning the results are both interesting and enlightening. The significance of educational measurement. Journal of Education, Vol. lxxx, pp. 61-78, July 16, 1914. An excellent exposition of the need of standards of achievement in education and of scales for the measurement of that achievement. Emphasizes the difference between the two, and outlines the results of their use. The meaning of the movement for educational measurement is simply and clearly set forth. - Work of the department of educational investigation and measurement, Boston, Mass. The 15th Yearbook of the National Society for the Study of Education, pp. 61-68, 1916. A. Measurement of educational results. B. Supervision of a revision of the elementary course of study. C. Organization of a plan for the promotion of teachers on merit. IN ELEMENTARY SUBJECTS 5 Birch, T. Bruce. Standard tests and scales of measurement. Psychological Clinic, Vol. x, pp. 49-57, April 15, 1916. An article outlining the need of measuring the achievements of pupils. Explains the functions of a successful test and outlines a plan whereby tests may be more scientifically prepared. A valuable article for one unacquainted with the aims and functions of tests. Bliss, Don C. School measurements and school administration. Educational Administration and Supervision, Vol. 1, No. 2, pp. 77-88, Feb. 1915. A strong argument in favor of the use of standard tests. The article shows, by charts, the results of using test's in penmanship, spelling, and arithmetic to stimulate individual improvement and to give the teacher an exact idea of the condition of the class and of the individual. It does not deal, except in the case of penmanship, with the use of so-called " standard tests," but rather with the use of tests that are "standard" for the city or for the school. Of interest to superinten- dents, principals, and teachers who would know definitely about the progress of pupils in subjects in which the progress can be definitely measured. — The application of standard measurements to school adminis- tration. The 15 th Yearbook of the National Society for the Study of Education, pp. 69-78, 1916. Standard tests were applied in fundamental subjects at Montdair, N. J. in 1 91 2. The report accompanied by charts and tables lays emphasis on subnormal pupils. Brown, George A. Help from the use of standard scales of attain- ment in school subjects. School and Home Education, No. 33, p. 300, April 1913-14. States briefly the real purpose of tests, showing, for example, just what the Courtis Tests are supposed to measure. Colvin, Stephen. Marks and the marking system as an incentive to study. Education, Vol. xxxn, pp. 560-572, May 1912. The writer advocates a carefully graded scale of objective measure- ment as a scientific and vital means of measuring the progress, both relative and absolute, of pupils. He believes that such standards should give a body of exact data for the use of the educational expert, should furnish a definite standard for the relative place and assignment of pupils, and should also prove a vigorous stimulus for the pupils themselves. 6 BIBLIOGRAPHY OF MEASUREMENT Courtis, S. A. Educational diagnosis. Educational Adminis- tration and Supervision, No. i, pp. 89-116, Feb. 1915. A plea for increasing the efficiency of school work by diagnosing individual cases by the aid of the Courtis Standard Tests. Graphs, diagrams, etc., with their explanations, are given, showing the general results of these tests. From these the conclusion is drawn that most schools are inefficient. Other graphs and diagrams are given showing typical and interesting cases brought to light by the use of these tests and for which methods of treatment are suggested. The claim is made that by such an analysis of conditions and by the intelligent remedying of defects, made possible by the use of standard tests, a great gain in efficiency will be made. The article is important for those who are seeking school efficiency through the use of standard tests. Supervisory control by means of objective standards. Con- ference on Educational Measurements, The Extension Division of Indiana University, Bloomington, Ind., pp. 37- 68, April 16 and 17, 1915. An illustrated lecture based chiefly on Courtis Tests in arithmetic presented to show the value of standard tests as an aid in supervision. Urges setting standards of attainment for different grades in different subjects and setting children to other tasks when the standard is reached. Shows that in the mechanical elements of education, the size of class is immaterial. Pictures a new school plant resulting from the use of standards. An article of much significance for school administration. Littell, H. V. Testing efficiency in school. Proceedings 28th Annual Meeting Associated Principals and Council of Ele- mentary Principals and Teachers, pp. 67-75, I 9 I 2 - A discussion of the lack of efficiency in the present teaching process. The results of the application of the standard tests and scales show that there is a need of wider use of tests and standards. Proof of the values of scientific measuring of ability. Maxwell, William H. Efficiency of schools and school systems. Proceedings, National Education Association, pp. 395-402, i9 J 5- A vigorous protest, couched in definite statements, against certain methods of testing the efficiency of a school system. Deprecates the use of the Courtis Tests, and attempts to prove that statistical measure- ments do not apply in education. Gives a list of 15 " conditions " or propositions by which the efficiency of a school system may be deter- mined. Gives in brief a common point of view. IN ELEMENTARY SUBJECTS 7 Monroe, Walter S. Standard tests and their uses. Teaching, Vol. i, pp. 9-14, Nov. 15, 1914. Discusses the weaknesses in the present method of grading pupils' work in accordance with the personal opinion of the individual teachers. Presents the value of educational standards and objective measure- ments, and contains a short description of the Courtis Tests in arith- metic, and Ayres' and Thorndike's penmanship scales. The facts are briefly and simply told. A good article for young teachers to read. Scott, Fred N. Efficiency for efficiency's sake. School Review, Vol. xxhi, pp. 34-42, Jan. 1915. Rehearses the story of the rapid growth of interest in educational efficiency and points out the danger of overlooking the real end and nature of education in the zeal for measurement, and the danger of pursuing efficiency solely for efficiency's sake. Shows the real work of a true teacher, and emphasizes the superior value of the things in teach- ing that can not be measured. A thoroughly human article by one who appreciates the significance of the efficiency movement in educa- tion, but who sees the need of differentiating between the measurable and unmeasurable, and senses the relative value of the two. Smith, James H. Individual variations in arithmetic. Elemen- tary School Journal, Vol. xvh, No. 3, pp. 195-200, Nov. 1916. A report of a study made to determine the cause of individual varia- tions within any class and to remedy this by methods of instruction. Test material used was taken from that of Cleveland Survey. Test work was done orally by individual pupils and written report prepared by investigators. Nine main sources of error were found and frequency of each shown by tabulation. Probable causes of certain of these are discussed. Three general methods of instruction for improvement were followed and final results of work as shown by second test shown in comparative graphs. A good account of the use of standard tests as a means of diagnosis. Stockton, James L. Exact measurements in education. Row, Peterson & Co., Chicago and New York, 1915. The author discusses in a scholarly way four theses: (1) Measure- ment in education should have for its goal the computation of work and rate of work (power) in the sense in which these terms are used in mechanics. (2) Scales of force, space, and time exist or can be made for school subjects, and the standard units of these scales should be combined into standard units of work and power. (Units are worked out for penmanship and illustrated by experiments with the Thorndike scale.) (3) Many units in many school subjects should be supple- 8 BIBLIOGRAPHY OF MEASUREMENT merited by a single unit for the computation of mental work and mental power in all school subjects. (A theory of a scale of intelligence is presented.) (4) To consider either force, space, or time alone or to combine them in an arbitrary manner gives unreliable results. (Illustrates this from penmanship rating and from use of Binet-Simon test.) Points the way clearly to future work in scale making and in using scales. Valuable to scientific investigators. Strayer, G. D. By what standards or tests shall the efficiency of a school or system of schools be measured ? Proceedings, National Education Association, pp. 560-566, 191 2. Calls attention to the possible fines of work by the National Council Committee on Standards and Tests which may result in the production of a significent report later on. These deal with (1) producing an annotated bibliography of all the studies on educational standards or test; (2) bringing together the experiences of all engaged in such in- vestigations and organizing it in such form as to aid in the development of a system or scheme of measurement which will take into account all the types of educational activity, i. e., business organization, educa- tional administration, and measurement of achievement of pupils, classes and schools. Points out that demand for measurement is not new, that the present day is dissatisfied with the rough approximations of the past and that progress has been made, that investigations to-day are partial and may be injurious unless the relation to the whole of the activity investigated be kept clearly in mind. Interesting as a definite appreciation of what is to be done, and as indicating that efforts are being made to do the things necessary. The use of tests and scales of measurement in the administra- tion of schools. Proceedings, National Education Associa- tion, pp. 579-5 82 > J 9i5- Discusses the four uses of tests and scales: (1) To compare achievements of children in different school systems. (2) To compare achievements of children in the same school system. (3) To study progress of pupils in a subject from grade to grade. (4) To indicate progress of any group for a certain period. Illustrates the use of Courtis Tests, spelling tests, Hillegas Composi- tion Scale, and Thorndike Handwriting Scale. Gives a few hints as to conclusions to be drawn from the use of tests and scales, and the value of their use to teachers and pupils. A brief presentation that will tend to encourage the use of standard tests and scales by educational executives. - Report of the committee of the National Educational Associa- tion on tests and standards of efficiency in schools and school systems. Proceedings, National Education Association, 191 5. IN ELEMENTARY SUBJECTS 9 The report is accompanied by discussions of leading educators on the subject. It urges the development of scientifically derived scales of measurement for subjects and school systems. Thorndike, Edward L. The measurement of educational prod- ucts. The School Review, Vol. xx, No. 5, pp. 289-300, May 1912. Discusses the need of some means of objective evaluations of the work of the schools, and the difficulty of securing such an evaluation. States the requirements of an ideal scale. Shows how a large number of subjective judgments may become the basis of an objective scale as well as to determine the very necessary zero point. Dr. Thorndike recognizes the probable main objections to standard scales or tests but offers refutation for these arguments. An article of value to all stu- dents in the field as an aid to understanding the validity of tests and scales which have been scientifically handled. II REPORTS ON THE USE OF TESTS IN VARIOUS SUBJECTS; LISTS OF TESTS AND SCALES Babbitt, Franklin. Report of the school survey of district number one in the city and county of Denver, Colo. The School Survey Committee, Denver, Colo., pp. 50-74, 1916. Contains an account of measuring the quality of compositions in grades 4-8 inclusive. A scale was derived from the compositions written in the test, and then the compositions were rated by the use of this scale. Grades were tested in nineteen schools and the results tabulated. The scale used is printed in full in the report. Contains, also, an account of a test in penmanship, given in all the schools in the district, to obtain measures in speed and form. Results in form were measured by the Ayres scale. The speed results are given in terms of words written per minute. Tables and charts show the records of the different grades in each school, and the findings are briefly discussed. A good example of the practical use of standard scales and tests. Childs, H. G. A half-year's progress in the achievement of one school system. The 15 th Yearbook of the National Society for the Study of Education, pp. 79-90, 1916. Progress as measured by the Thorndike visual vocabulary Test and by the Courtis Tests, series B, given in Bloomington, Indiana. Charts and tables show the progress in achievement in this school system. 10 BIBLIOGRAPHY OF MEASUREMENT Cubberley, E. P. Use of standard tests at Salt Lake City, Utah. The 15th Yearbook of the National Society for the Study of Education, pp. 107-110, 1916. A measure of instruction in writing, spelling, composition, arithme- tic, and reading in the elementary schools. Used Ayres, Hillegas, Thorndike, and Courtis tests and scales. Salt Lake City averaged above the standard. Gray, William S. Descriptive list of standard tests. Elementary School Journal, Vol. xvn, pp. 24-34, Sept. 1916. An article telling of standard tests. — their nature, purpose, price, and the places at which they are obtainable. Very valuable as a refer- ence list. Harlan, Charles H. A comparison of the writing, spelling, and arithmetic abilities of comity and city children. Educational Administration and Supervision, Vol. 11, No. 9, pp. 560-573, Nov. 1916. A study undertaken to compare rural schools and city schools in writing, spelling, and arithmetic. Material was taken from 21 schools. The handwriting of the rural schools, on the average, was found of approximately the same quality and was produced at about the same speed as that of city schools. In spelling the country averages were uniformly lower than the city averages. In arithmetic the rural schools ranked approximately two years below the city schools. There were 13 more pupils per hundred in the rural schools who were average than in the city schools. Valuable for rural superintendents. Johnston, Joseph H. A brief tabular history of the movement toward standardization by means of scales and tests of educa- tional achievement in the elementary school subjects. Educa- tional Administration and Supervision, Vol. 11, No. 8, pp. 483-492, Oct. 1916. A brief record of those investigations that have resulted in the for- mation of definite tests or scales in the elementary school subjects, and other more noteworthy attempts at standardization from the use of these tests and scales. Tabulates data for thirty-two investigations, extending from 1895 to 191 5, in arithmetic, drawing, handwriting, English composition, language, reading, oral reading, silent reading, spelling, and vocabulary. The data include names of investigators, subject investigated, date of study, general method employed, and results. A bibliography indicates where the details of each investiga- tion may be found. IN ELEMENTARY SUBJECTS II Judd, Charles Hubbard. Measuring the work of the public schools. The Survey Committee of the Cleveland Founda- tion, 290 pp., Cleveland, Ohio, 1916. Tells of poor conditions found in Cleveland schools by the Survey Committee. Points these out definitely by tests in all subjects and comparison with tests of many other cities. In summaries of results of tests in each subject are given suggestions for improving conditions. Tests given are shown in appendix. Manahan, J. L. A bibliography of educational surveys and tests. University of Virginia Record, Extension Series, Vol. 11, No. 3, pp. 50-92, Nov. 1916. An annotated bibliography of educational surveys and tests giving a brief summary of each reference presented. These are of four kinds, (a) school surveys, state, city, and county; (b) standard tests in school subjects; (c) studies in the application of these tests; (d) tests of mental intelligence and studies in their application (not annotated in this issue). Sufficient data are presented to give a fair idea of the scope and value of each reference, together with all information needed to obtain it. The list is by no means complete, but, so far as it goes, it is valuable for reference. Springer, Isidore. Teachers' year book of educational investiga- tions. Division of Reference and Research, Department of Education, New York City, Pub. No. 14, 1914, 12 pp. 8vo. (Bulletin no. 1.) Contains the following scales and comments upon them: — Courtis Arithmetic and English Tests, Harvard Newton Scale for Composition, Buckingham Spelling Test, Thorndike Reading Scale and Penman- ship Scale, Kendall Geography Tests, Thorndike Drawing Scale, and also Tests of educational administration. Starch, Daniel. Educational measurements. Macmillan Co., 197 pp., New York, 1916. A collection of the greater number of the tests for which standard scales have been determined. Gives the tests in full and more or less complete directions for applying them. Four chapters which accom- pany the collected tests give excellent reasons for the use of such tests. This volume is very useful to those who desire to learn the nature of the various tests and as a reference book for those who contemplate using them. Strayer, George D. Report of a survey of the school system of Butte, Mont. School Board of Butte, Mont., pp. 69-96, 1914. 12 BIBLIOGRAPHY OF MEASUREMENT A chapter in the report of the Survey Commission of Butte dealing with the achievement of pupils in the elementary schools of that city. The achievement in spelling, composition, handwriting, and the fun- damental processes in arithmetic was tested by means of standardized tests and scales. In spelling, tests of a known standard derived by Ayres were used; in composition, the scoring was done with the aid of the Hillegas scale. The Thorndike scale was used in scoring the pen- manship papers, and the Courtis Tests were used in arithmetic. The chapter contains the exact tests given, states the method of giving and of scoring, shows the results by means of tables and graphs, and com- ments instructively on these results. Ill READING Brown, H. A. The measurement of efficiency of instruction in reading. Elementary School Teacher, Vol. xiv, No. 10, pp. 477-490, June 1914. Discusses the need and effectiveness of tests of reading ability. Describes procedure in testing reading in seven New Hampshire school systems. Gives the tests and the method of scoring results and dis- cusses reasons for the use of given standards. The author uses results of tests given as a basis for analysis of efficiency of teaching of reading, thus showing one of the virtues of standard tests. This is a painstaking piece of work, but the method of scoring is open to criticism because of its dependence on subjective judgments. The measurement of ability to read. N. H. Dept. of Public Instruction in cooperation with the General Educational Board. Bureau of Research, Bulletin, No. 1. Discusses the importance and need of reading tests. Gives tests and describes method used in testing reading in many New Hampshire communities. Presents data in tabular and graphic form, together with a key to the scheme for scoring papers. Discusses the use of these data as aids in diagnosis, for comparison of results of various teaching methods, and to help in solution of other problems of administration in this field. Courtis, S. A. Standard tests in English. Elementary School Teacher, Vol. xiv, pp. 374-392, April 1914. The writer differentiates between standard tests on the one hand and ordinary examinations and the "pure science" tests of psychol- ogists on the other, and states definitely and clearly the four functions of standard tests. He describes briefly the six English tests he has IN ELEMENTARY SUBJECTS 1 3 devised for use in measuring ability in reading, and to illustrate the types of results obtained gives a score sheet showing scores made by an adult, and also graphs showing the results from measurements in one school. The discussion accompanying the graphs is both interesting and instructive. School executives and all teachers of English will find the article of distinct value. The study being made by the author, as here outlined, ought to produce ultimately a valuable set of standards in English. Courtis, S. A. Standards in rates of reading. National Society for the Study of Education, Fourteenth Year Book, Pt. I, pp. 44-58, Feb. 1915. A discussion of the teaching of reading for the mastery of technique. Gives reasons for use of a standard scale. States that training in technique of reading is useless beyond a certain point or mental age. Provides a series of sample tests with tabulations and graphs of scores. Discusses scores as a basis for diagnosis of weaknesses. Shows the value of the use of standard scores in supervision. A thoroughly scientific article by an authority in this field. Gray, William S. A selected bibliography upon practical tests of reading ability. National Society for the Study of Educa- tion, Fourteenth Year Book, Pt. I, pp. 59, 60, Feb. 191 5. A bibliography of the best publications on this subject appearing prior to February, 191 5. Methods of testing reading. Elementary School Journal, Vol. xvi, No. 5, pp. 231-246, Jan. 1916. A detailed description of certain rough general tests which may be administered under ordinary school conditions. There is also a dis- cussion of some standard tests organized to secure more exact results. Gives directions for preliminary tests and a scoring scheme. Presents a set of standard tests with methods of scoring and directions for work- ing up results. Provision is made in the method of administration for valuable comparisons of any grade with those contiguous to it, as a means of checking up progress. A clear presentation of a carefully constructed series of tests. Should be compared critically with similar work by Starch, by Thorndike, and by Kelley. There are several commendable features about these tests, and the article as a whole is worthy of a careful reading by all interested in this field of educational activity. Methods of testing reading, II. Elementary School Journal, Vol. xvi, No. 6, pp. 281-198, Feb. 1916. Continuation of an article by the writer in the preceding issue of this periodical. Presents a new series of tests of oral and silent reading, 14 BIBLIOGRAPHY OF MEASUREMENT with directions for their administration carefully discussed. A scoring scheme is presented and discussed. Speed and mechanical difficulty of reading and speed and quality of reproduction are scored. Sample scores are tabulated. Concludes with an interesting discussion of the value of tests to administrative officers, and gives suggestion for grading the element of expression. One of the best of the articles on standard reading tests, containing some commendable original fea- tures. The article should be read by progressive administrative officers; and the class-room teacher could profit by study of it and the subsequent use of the tests. Gray, William S. A study of the emphasis on various phases of reading instruction in two cities. Elementary School Journal, Vol. xvn, No. 3, pp. 178-186, Nov. 1916. A critical comparison of the results of the tests of reading in St. Louis and Cleveland to show how different phases of instruction are empha- sized. Data secured during educational surveys of these cities are analysed in the light of known practices in these cities. No test material is presented. Graphic representations are presented and dis- cussed to show differences in results which come from different prac- tices in instruction. Points of strength and weakness are called to attention and means of improvement suggested. A cooperative study in reading in eleven cities of northern Illinois. Elementary School Journal, Vol. xvn, No. 4, pp. 250-265, Dec. 1916. Reports a cooperative study by a branch of the Illinois Teachers Association by means of the author's standard tests. Both oral and silent reading were tested. Tabulations of data are extensively shown and thoroughly discussed and compared. Test materials used are not introduced. A highly interesting article showing the great value of standard tests for purposes of comparison and diagnosis. Haggerty, M. E. Scales for reading vocabulary of primary children. Elementary School Journal, Vol. xvn, No. 2, pp. 106-115, Oct. 1916. Describes the derivation of scales by use of the tests suggested by Robinson G. Jones in the Fourteenth Year Book of the Society for the Study of Education. The tests are lists of 118 phonic and 236 sight words and are used to determine the relative difficulty of the words. Tests, tabulated data, and discussion of problems suggested by the results follow, with a critical discussion of the method of grading fol- lowed by the author of the tests. An interesting account of an experi- ment with tests. IN ELEMENTARY SUBJECTS 1 5 Jones, Robinson G. Standard vocabulary. The National Society for the Study of Education, Fourteenth Year Book, Chap. 4, pp. 37-43, Parti, Feb. 1915. A test of vocabulary based on words occurring in ten standard primers. Phonic and sight words are given values from which scores are made. A tabulation of scores of two schools are used as a basis for comparison of quality of instruction. Limitations and validity of the tests as measures of pupils and teachers are discussed. A method of determining a coefficient of efficiency of the teacher is introduced. A valuable article, especially for primary teachers. Judd, C. H. Reading. The Fifteenth Yearbook of the National Society for the Study of Education, pp. 111-119, 1916. A report of the Cleveland survey, methods used, and results obtained. Accompanied by tables. Of interest to all investigators of oral and silent reading. Reading tests. Elementary School Teacher, Vol. xiv, No. 8, PP- 365-373. A P ril I9H- An article written by Professor Judd as representative of the Com- mittee of Standards and Tests of the National Council of Education. Its object is to secure wider cooperation of school officials in the testing of reading. The author tries to allay the teachers' fear of such tests by showing that they are not essentially new. Methods and objects in testing are discussed, and suggestions for conducting tests in connection with regular class work are given. To those interested in securing the cooperation of teachers in measuring the result of their work this article will provide excellent material. Report of the committee on tests and standards of efficiency in schools and school systems. National Educational Associa- tion, Journal of Proceedings, Sect. B, Reading Tests, pp. 561-565, Oakland, Cal., 1915. A report of the progress of standard tests in reading with names of the leading men engaged in the preparation of tests. Discusses types of investigations and gives suggestions for testing by teachers in daily class work. Attention is called to overlapping of grades and to inter- class variations. Discusses the distinction between power of recogni- tion and retention, and between oral and silent reading; and states that ignorance or disregard of these distinctions is the source of much loss of time and effort. A very excellent report of the progress, methods, tendencies, and problems of the measurement by objective standards of the product of instruction in reading. 1 6 BIBLIOGRAPHY OF MEASUREMENT Judd, C. H. Measuring the work of the public schools. Cleveland Educational Survey, pp. 290, Cleveland, Ohio, 1916. A report of a careful investigation of reading in the Cleveland schools by means of a series of carefully graded tests not previously used elsewhere. Children were tested in oral and silent reading to determine rates of reading and extensive tabulations of data are given. Understanding of content was also tested and relation between rate and quality of reading determined. Results of all tests were analysed to determine the influence of various social factors on the ability to read. Kelley, F. J. The Kansas reading tests. Journal of Educational Psychology, Vol. vii, No. 2, pp. 63-80, Feb. 1916. Describes the formulation of the " Kansas silent reading tests." Discusses the " four criteria of all tests " and shows how these tests meet these criteria. Tests are produced in full with directions for administering and scoring. Tabulations of data from these tests in 19 cities are shown, and discussed with reference to the class and individ- ual variations made evident by them. Attention is called to valuable comparative studies based on data secured by means of these tests. Gives median scores for grades 3 to 12. This article is of value to all students in this field, and especially to administrative officers and teachers. Test material is carefully selected and graded and of unusual quantity. A wider section of the grades is provided for than in most tests of this nature. Simplicity of administration and scoring is espe- cially commendable. Mead, Cyrus D. Silent vs. oral reading with 160 sixth grade children. Journal of Educational Psychology, Vol. vi, No. 6, pp. 345-348, June 1915. Describes a series of tests given to determine the relative value of the two methods of reading as a means of securing the content of the matter read. No test material is given except as references to standard texts used. Directions are given for administering and scoring the tests, and results in five classes tested are tabulated. The conclusion is drawn that too much time is given to oral reading. An interesting description of an experiment. Oberholzer, E. E. Testing the efficiency of reading in the grades. Elementary School Journal, Vol. xv, No. 6, pp. 313-322, Feb. 1915. A report of tests given to about 1800 pupils in the elementary schools of Tulsa, Oklahoma, under the direction of the writer, superinten- dent of the Tulsa schools. Both oral and silent reading were tested for speed; and the latter for grasp of content. The tests were essen- IN ELEMENTARY SUBJECTS 1 7 tially speed tests, as no tabulation of results is given for test of grasp of content. The method of scoring is simple and easy for the average teacher to understand. The results and the conclusions drawn from them are valuable as indices of the quality of our teaching. An extended list of test material is appended. Otis, Arthur S. Considerations concerning the making of a scale for the measurement of reading ability. The Pedagogical Seminary, Vol. xxm, No. 4, pp. 528-549, Dec. 1916. The. author formulates a definition of reading ability, analyses the reading act to determine what factors constitute reading ability per se, and then presents an outline of a scale suitable for teaching. He also suggests the character of investigation necessary to perfect such a scale. Pitner, Rudolf and Gilliland, A. R. Oral and silent reading. Journal of Educational Psychology, Vol. vn, No. 4, pp. 201- 212, April 1916. A report of an experimental determination of the relative economy of oral and silent reading as means of securing content of matter read. Covers elementary, high school, and college grades. Rate of reading and amount reproduced were the points considered. Graded test material was used, five sets of cards of increasing difficulty being employed in appropriate grades. The test material used is not given. Results of the experiment are tabulated and discussed. The conclu- sions reached are: (a) oral and silent reading are equally efficient in the third and fourth grades; (b) up to college grades oral reading is slower than silent reading and efficiency about the same; (c) adults show the superior efficiency of silent reading. The general conclusion is that much time is wasted on oral reading. Richards, Alva M. and Davidson, P. E. Correlations of single measures of some representative reading tests. School and Society, Vol. rv, No. 88, pp. 375-377, Sept. 2, 1916. Discusses the difference between the foremost standard reading tests. Calls attention to differences of opinion of their authors as to what constitutes a measure of reading ability. Computes correlations between certain tests of Starch, Kelley, Trabue, Thorndike, and a narrative completion test. Shows tests as used. Correlations are tabulated after being computed by Spearman Footrule Formula. Conclusion is that " completion and vocabulary tests are somewhat more significant than either of so-called reading tests." A valuable study to those students in this field who plan to use or formulate reading tests. 1 8 BIBLIOGRAPHY OF MEASUREMENT Starch, Daniel. The measurement of efficiency in reading. Journal of Educational Psychology, Vol. vi, No. i, pp. i- 24, Jan. 1915. Analyses reading under headings (a) comprehension; (b) speed; (c) pronunciation. The common method of measuring speed is fol- lowed and simple tests are given. Special tests for comprehension are given and a system of scoring pronunciation is suggested. A critical discussion of the validity of the tests is given. Standards of efficiency are indicated by standard scores. Individual variations within grades and the overlapping of grades is illustrated by tabulated scores. The article is interesting and instructive, but the tests resemble those produced by Thorndike. The method of scoring comprehension seems arbitrary and questionable, but this procedure is ably defended in the context. Thorndike, Edward L. The measurement of ability in reading. Teachers College Record, Vol. xv, No. 4, pp. 207-277, Sept. 1914. Presents Scale A, a measure of visual vocabulary, consisting of lists of words to be classified by the pupils. Results are tabulated and a key provided to assist in the interpretation of the data. The scale is absolutely objective and requires no exercise of the teacher's judgment. Full directions are given for administering and scoring the tests, and its advantages and defects are discussed. This is an extremely valua- ble discussion, which should be read by every student in this field. In connection with it an article by the same author appearing in Novem- ber, 191 5, should be studied. An improved scale for measuring ability in reading. Teachers College Record, Vol. xvi, No. 5, pp. 31-53, Nov. 1915. A continuation and expansion of an article by this author which appeared in this periodical in Sept., 1914. Presents a revised form of his earlier scale, Alpha I, with an excellent key for scoring and estimat- ing ability in silent reading. Grasp of content is the element tested, as this seems the most essential form of reading ability. The tests are carefully graded and coefficients of relative difficulty are scientifically derived. The tests are designed to be placed in the hands of any efficient teacher for administration, and would reveal the cause of failure of many pupils to do assigned work requiring silent reading. The tests may be open to criticism, in that the pupils are permitted free access to the test passages, while answering questions on their content. Uhl, W. L. The use of the results of reading tests as a basis for planning remedial work. Elementary School Journal, Vol. xvii, No. 4, pp. 266-275, Dec. 1916. IN ELEMENTARY SUBJECTS 1 9 Describes an experiment carried on with certain children in the Oshkosh, Wis., schools showing a use of standard tests as a basis for remedial action. A rough test was made by means of the Kansas Tests (Kelley's) and these were followed by Gray's individual oral tests and by the silent reading tests of the same author. With data thus secured as a basis remedies were applied to secure improvement and at the end of a six-week period the participants were again tested. Results are tabulated and discussed. Common errors are called to attention and individual differences shown by tabulations. Six conclusions are drawn each of which has decided significance. Zeidler, Richard. Tests in silent reading in the rural schools of Santa Clara County, California. The Elementary School Journal, Vol. xvn, No. i, pp. 55-62, Sept. 1916. An account of the use of the Starch Reading Tests for the purpose of comparison of rural and city schools. Contiguous grades were also tested for the purpose of comparison. Method of procedure is described and tabulated data are given, with a discussion and comparison of results in different schools. Six very suggestive conclusions are drawn. IV PENMANSHIP Ashbaugh, Ernest J. Handwriting of the Iowa state children. Extension Division Bulletin, State University of Iowa, Iowa City, No. 15, pp. 1-23, March 1, 1916. The method by which the penmanship ability of the Iowa school children was measured is given in this bulletin. The Ayres measuring scale was used. There appeared very little difference between the writing ability of rural school children and of city children. The writ- ing of Iowa children compares favorably with the writing found else- where in the United States. Excellent graphs and tables are given to show the comparisons drawn by the author. Ayres, Leonard P. A scale for measuring the quality of hand- writing of school children. New York: Russell Sage Founda- tion, Bulletin, No. 113, pp. 16, 1912. The report of the investigation from which the Ayres scale in hand- writing was developed. Each step in the formation of the scale is dis- cussed and illustrated by charts and tables. The scale was formed by measuring 1578 samples of children's writing. An average reading time was computed for each sample after it was read by ten different persons. Specimens were chosen to represent eight degrees of legi- 20 BIBLIOGRAPHY OF MEASUREMENT bility. There is a sample of slant, medium, and vertical writing for each degree of legibility. The limits of the Ayres scale are 20 and 90. Each step is designated by a division of ten. One of the most popular hand- writing scales. Breed, Frederick S. and Culp, Vernon. An application and critique of the Ayres handwriting scale. School and Society, Vol. 11, pp. 639-647, Oct. 1015. Data from fifteen Michigan schools to show that the Ayres scale cannot be regarded as a satisfactory scale for the measurement of the legibility of handwriting. The high correlation between Ayres scale values and Freeman form values confirm the idea that the Ayres scale measures form rather than legibility. Of interest to any one contem- plating the use of the Ayres scale. The relation of legibility and form in handwriting. School and Society, Vol. iv, No. 101, pp. 870-872, Dec. 2, 1916. An attempt to evaluate certain factors of form in handwriting. One hundred samples of handwriting were used for the legibility-form cal- culations. Legibility is measured by the speed-of-reading method. Form is measured by the Freeman scales. A table is given to show what factors need emphasis in the scoring of handwriting when the aim is legibility. This article should be read in connection with a previous study published in School and Society by the same authors. Valuable for teachers as a guide in ranking penmanship. Freeman, Frank N. An analytical scale for judging handwriting. Elementary School Journal, Vol. xv, No. 8, pp. 432-441, April 1915. A scale is described for the purpose of eliminating difficulties found in the Ayres and Thorndike Scales, namely, lack of uniformity in results obtained and ambiguity as to the characteristics in writing which are to be used as a basis of judgment. Uniformity in slant and alignment, quality of line, letter formation, and spacing were con- sidered necessary characteristics for ranking. Charts were formed containing three grades of excellence for each characteristic in question. Specimens chosen were measured, so far as possible, objectively. Handwriting tests for use in school surveys. The Elementary School Journal, Vol. xvi, No. 6, pp. 299-302, Feb. 1916. The material is presented for the benefit of those who wish to make surveys in their own school systems. Two kinds of writing are studied. First, a composition test is given to measure the appearance or form of the writing. Second, a formal writing test is given to measure the IN ELEMENTARY SUBJECTS 21 quality and speed of the writing. The test does not involve the use of a handwriting scale, judgment of form and quality of writing being left to the opinion of the teacher. The lists are simple and easily applicable to any school system. Gray, Truman C. The training of judgment in the use of the Ayres scale for handwriting. Journal of Educational Psychol- ogy, Vol. vi, No. 2, pp. 85-98, Feb. 1915. A discussion of an experiment undertaken to discover causes for the considerable variation in grading handwriting by the Ayres scale. Three students in education without experience in the use of scales were given twenty-five samples of writing a week for twenty weeks to grade according to the Ayres scale. Upon the twentieth week five hundred additional samples, mainly of the elementary grades, were judged. Results showed that training of judgment in the use of the Ayres scale was essential in procuring uniform results. An article of practical importance in testing. A score card for the measurement of handwriting. University of Texas, Austin, Texas, Bulletin No. 37, pp. 1-50, 1915. An exposition of the method by which a score card for the measure- ment of handwriting was evolved. The steps in the method were (1) the determination of the elements of handwriting to be used on a score card, and (2) the evaluation of these elements. This score card is meant for the use of teachers and others in measuring handwriting, and in recording the detailed study of the writing of individuals. Johnson, George L. Measuring the quality of handwriting. The Elementary School Journal, Vol. xvi, No. 6, pp. 302-313, Feb. 1916. A series of short essays, by different authors, on the making of a handwriting scale, together with a discussion of the purpose and use of such a scale. Several tables are given to show each step in the forma- tion of the scale which is presented in the article. The scale is formed by judging all samples according to standards previously derived from an analysis of the qualities sought in penmanship as it is taught in St. Louis schools. The judgment of the samples was thus rendered ex- ceptionally careful and well informed. This scale will be found useful when ratings of detailed characteristics are desired. King, Irving and Johnson, Harry. The writing abilities of the elementary and grammar school pupils of a city school system measured by the Ayres scale. Journal of Educational Psy- chology, Vol. in, No. 9, pp. 514-520, November, 1912. A study to test the practicability of the Ayres scale and to determine the age, grade, and school variations in legibility of handwriting. 22 BIBLIOGRAPHY OF MEASUREMENT Grade 5 B was the only marked deviation from the upward trend of improvement. The most striking improvement was in 2 B over 1 A, ages 6 to 7 years. The averages of the deviations of a single judge and of eight judges were small, indicating that the subjective factor in judg- ment of handwriting may be largely reduced. Manuel, H. T. The use of an objective scale in grading hand- writing. (Results of a study of the Ayres scale.) Elementary- School Journal, Vol. xv, No. 5, pp. 269-278, Jan. 1915. This article reports some of the main results of a study of the Ayres scale and their apparent implications as to variability in marking with the scale. Unpracticed markers vary widely in their use of the scale when their training is only independent practice; practiced markers vary considerably. Markers should be technically trained. All scales should have the same foundation, — an objective definition of the grades of handwriting to be recognized. Pitner, Rudolf. A comparison of the Ayres and Thorndike handwriting scales. Journal of Educational Psychology, Vol. v, No. 4, pp. 525-536. Nov - IQ i4- An experiment to find which scale will show the least amount of deviation in the ratings of the same samples by a number of individ- uals. The Ayres scale considers legibility alone. The Thorndike considers legibility, beauty, character, ease, etc. The Thorndike scale showed a greater uniformity in results — due, probably, to the fact, that it took into account all the factors that go to influence our judg- ment of handwriting and did not depend on legibility alone. Sackett, Leroy W. Comparable measures of handwriting. School and Society, Vol. iv, No. 95, pp. 640-645, Oct. 21, 1916. A study of the measures of handwriting. Thirty-six university sophomores were asked to provide two samples of writing. Two addi- tional samples were taken exactly comparable with the first two except that the pupils knew that their work was to be judged for penmanship. The 144 samples were graded by practiced judges. Six different scales were used. In order to make results comparable the judgments were all reduced to a percentage basis. Tables are given to show the grades given each group by the judges, also to show how the attempt to improve writing affects speed. Six reference tables are printed to denote how the readings by each scale may be interpreted in terms of the other scales. Valuable to students who wish to know what the scale units which they have used may mean in terms of other scale units. IN ELEMENTARY SUBJECTS 23 Starch, Daniel. The measurement of handwriting. Journal of Educational Psychology, Vol. iv, No. 8, pp. 445-464, Oct. 1913- The Thorndike scale, the pioneer contribution, has a high degree of accuracy. The Ayres scale has several shortcomings. It does not con- tain enough steps; constructed to measure legibility, it actually judges form or appearance. Legibility is measured on the time required to read a specimen; producibility, by measuring the speed and ease of writing; form or beauty is of little practical consequence. Madison, Wis., children showed little improvement in speed and form in the first three years but speed increased rapidly after that. Improvement in legibility showed the reverse tendency, being at its height in the fourth year. Standards of attainment to be reached by each grade were obtained and are now used in Madison. The measurement of efficiency in writing. Journal of Educa- tional Psychology, Vol. vi, No. 2, pp. 106-114, Feb. 1915. A presentation of the results obtained from a writing test made with 4074 pupils in nineteen schools. The Thorndike scale was used because it offered a wider range of rankings. Tables and charts are given to show how simply the efficiency of pupils can be recorded by graphs. Results showed that because of the wide range of individual abilities found in the writing, classes differ from each other only by small amounts. Informing and suggestive. Thorndike, Edward L. Handwriting. Teachers College Record, Vol. xi, No. 2, pp. 1-93, March, 1910. Describes the means by which a scale for handwriting may be made, presents such a scale for the handwriting of children in grades 5-8, and explains how the scale is to be used. The scale is the result of some 20,000 ratings by more than 40 competent judges of the handwriting of more than 1000 pupils. It is intended to measure quality rather than quality and speed combined. There is also given a scale for the measurement of adult women's handwriting. As a whole the report is very carefully presented. It gives the reader opportunity to form his own opinion of the results, and is of great value to any one investigat- ing penmanship. (See also Teachers' Estimates of the Quality of Speci- mens of Handwriting, Thorndike, E. L. Teachers' College Record, Vol. XV, No. 5, pp. 1-10, Nov. 1914.) Means of measuring school achievements in handwriting. Educational Administration and Supervision, Vol. 1, No. 5, pp. 300-305, May 191 5. A presentation of a means for undertaking the measurement of quantity and quality of handwriting in grades 4-8 inclusive. Instruc- 24 BIBLIOGRAPHY OF MEASUREMENT tions are given for sampling and scoring the writing. Detailed instruc- tions for giving the tests are included. Important for those who con- template the testing of handwriting. Tobey, Lucy. Value of the Ayres' scale in writing to the grade teachers. School and Home Education, No. 33, pp. 300-301, April 1913-14. Shows how a standard scale works in the everyday class room. Gives the results of two formal applications of the Ayres scale and shows how the daily use of the scale " created enthusiasm for writing." Witham, Ernest C. A method of measuring handwriting. American School Board Journal, Vol. 48, No. 5, pp. 18-19, 72-73, May 1914. A presentation of a method for measuring the speed and legibility of handwriting. Suggestions are given for regulating form and beauty of writing. Scores and tables are included to illustrate the method. Directions are also given for testing the pupils. A simple method for measuring handwriting, easily applicable to any school system. All the elements of handwriting measured. Education Ad- ministration and Supervision, Vol. 1, No. 5, pp. 313-324, May 1915. A presentation of the results found by measuring all the elements of handwriting in two seventh grades. Grade specimens, scales, and tables are reproduced in this article. All elements are measured. The results with respect to form and beauty are reduced to the Weiss index and comparisons drawn between the Witham and other hand- writing scales. The author's scale being nearer the average is considered the more reliable. V SPELLING Ayres, Leonard P. Measurement of ability in spelling. New York: Russell Sage Foundation, pp. 1-56, 191 5. A report of an investigation in spelling from which was produced the Ayres Spelling Scale. The different steps in the formation of the scale are explained. Two features of the scale are: first, the selection of the thousand commonest words in the English language, and second, the determination of the difficulty of these words as spelled by children of different grades. On the basis of data received from 84 cities the words were so arranged in twenty-six columns that each column presents ap- IN ELEMENTARY SUBJECTS 2$ proximately equal steps in difficulty. Numbers at the top of the scale indicate what percentage of the words in each column is spelled cor- rectly by the children of each grade. An interesting effort at scale construction and of fundamental interest to teachers of spelling. Buckingham, B. R. Spelling ability: its measurement and dis- tribution. Contributions to Education, No. 59, Teachers College, Columbia University, 1913. Outlines the need of a uniform method of testing spelling ability and gives an account and explanation of experiments carried on in certain elementary schools of New York City to form a scale whereby words might be ranked according to the tested difficulty of spelling them. Jones, W. Franklin. Concrete investigation of the material of English spelling. University of South Dakota, 1914. A study to discover the words used by pupils of the different grades in their free written language. The material consists of the approxi- mate writing vocabulary of 1050 grade students, approximately 150 from each grade above the second. The theme writing began in the third school month and continued until the vocabulary was exhausted. The number of themes required to exhaust the vocabularies varied from 56 to 105. The total number of themes was slightly more than 7500. The total number of words was approximately 15,000,000. The total number of different words was 4532. These 4532 words are tabulated as used grade by grade. A list of 100 words most commonly misspelled — "The 100 Spelling Demons" — is compiled. Eleven valuable conclusions are drawn from the study, and an excellent bib- liography is added. One of the most valuable contributions to the study of spelling yet printed. Lewis, E. E. Spelling abilities of Iowa school children. Elemen- tary School Journal, Vol. xvi, No. 10, pp. 556-564, June 1916. This test describes an attempt to measure the spelling abilities of 8,624 school children in ten Iowa cities. The Buckingham scales were used. Tables are given to show the percentages of the Iowa children as compared with those in other cities in the United States. The writer believes that present standards are unfair, unreliable, and of small value. Otis, Arthur S. The reliability of spelling scales involving a " derivation formula "for correlation. School and Society, Vol. rv, No. 96, pp. 676-698, Oct. 28, 1916. Continued, pp. 716- 722, Nov. 4; pp. 75 ~75°> Nov - "J PP- 793-796, Nov. 18. 26 BIBLIOGRAPHY OF MEASUREMENT A discussion of great merit concerning the reliability of scales. Notes good and bad points of some of the scales now used and makes rec- ommendations as to the using of same. An extremely technical dis- cussion, of value to investigators, superintendents, and principals. Sackett, L. W. Measuring a school system by the Buckingham spelling scale. School and Society, Vol. n, No. 50, pp. 860- 864, Dec. 11, 1915. An adaptation of the Buckingham method for testing spelling in the grades. The article does not justify or criticize the methods by which the Buckingham scale was originated. It gives valuable data showing practical use of the scale. Measuring a school system by the Buckingham spelling scale. School and Society, Vol. 11, No. 51, pp. 894-898, Dec. 18, 1915- An application of the Buckingham scale for testing spelling in the grades. " A division was made on the basis of sex," and tests in two cities show that girls have " superior ability [in spelling] in all grades, but the superiority is not equally marked throughout the grades." Sears, J. B. Spelling efficiency in Oakland schools. School and Society, Vol. 11, No. 41, pp. 53*-537, 5^9-574, Oct. 9, 16, 1915- The aim of this study was to stimulate a professional interest in spelling by presenting problems that were interpreted and explained so as to provide teachers and principals with a full knowledge of the problems of the subject in their respective schools. The Ayres spelling test was used, illustrated by charts and scales. The evidence proved that in spelling the children are poorly classified and graded and that the standard of achievement varies greatly among classes, among grades, and among schools. Starch, Daniel. The measurement of efficiency in spelling and the overlapping of grades in combined measurements of reading, writing, and spelling. Journal of Educational Psychology, Vol. vi, No. 3, pp. 167-186, March 1915. A criticism of several methods of testing spelling, including the Ayres and Buckingham tests. The author presents his own fist of 600 words, selected from Webster's New International Dictionary after discarding all technical, scientific, and obsolete words. These were divided into six lists. Each list contained approximately the same number of words of every degree of difficulty. These lists were given to 2500 pupils in twelve different schools and showed that the range of IN ELEMENTARY SUBJECTS 2J ability in spelling closely conformed to that found in reading and writing. It shows that grading does not conform to rank in these sub- jects as measured by tests. Studebaker, J. W. Results of an investigation of pupils' ability to spell. Newson & Co., Boston, pp. 57, 1916. An account of an examination of the ability in spelling of the pupils in the public schools of Des Moines, given in 191 5. Tests were given on the words studied during the year, also on the hundred " demons," and on words from the Ayres list and from the Buckingham list. The results are tabulated, also shown by graphs. The lists used are printed, and the method of giving the tests explained. Interesting facts are noted and commented on. The pamphlet contains the Ayres list of 1000 commonest words. Suzzalo, H. and Pearson, H. C. Comparative experimental teaching of spelling. Teachers' College Record, Vol. xiii, No. 1, pp. 1-66, Jan. 1912. This article is divided into two parts. Part one contains a discussion of the worth of our present methods of teaching spelling; the relation of pedagogy and psychology to teaching processes and the value of comparative psychology to our teaching methods. Many problems requiring experimental investigation are offered. Part two gives an account of a comparative experiment in the teaching of homonyms. Shows that better results are obtained by teaching homonyms together rather than separately. Also, four experiments were conducted to find out which was the more efficient way of teaching spelling, the class- study method or the independent study method. More effective results were obtained by the use of the class-study method. Thorndike, Edward L. Means of measuring school achievements in spelling. Educational Administration and Supervision, Vol. 1, No. 5, pp. 306-312, May 1915. A presentation of a means for testing achievements in spelling by the use of classified groups of words. These words were made the means of elaborate study by B. R. Buckingham. Imaginary records are given to illustrate how comparisons of school systems in respect to achievement may be obtained. Words are arranged in series approxi- mately equal in difficulty. Careful instructions for a comparative sampling are included. Tedyman, W. F. A descriptive and critical study of Bucking- ham's investigation of spelling efficiency. Educational Administration and Supervision, Vol. 11, No. 5, pp. 290-304, May 1916. 28 BIBLIOGRAPHY OF MEASUREMENT The article is divided into two parts, first, a description of the method employed in forming the Buckingham scale, second, a criticism of the method and procedure employed in testing efficiency in spelling. The author contends that the scale is a statistical and not a pedagogical instrument, that individual differences are sacrificed for mass results, that the objection to standardization is the use of the instrument rather than the instrument itself. Especially interesting to efficiency experts. Wallin, J. E. Wallace. Spelling efficiency in relation to age, grade, and sex, and the question of transfer. Warwick and York, Baltimore, pp. 91, 191 1. " An experimental and critical study of the function of method in the teaching of spelling." Approximately 1000 children in public schools of Cleveland, from the fourth to the eighth grade, were tested on lists of from forty to fifty words. These words were embodied in compositions prepared by teachers and dictated to the pupils. Eleven conclusions are drawn from the tests. VI COMPOSITION Ballou, Frank W. Scales for the measurement of English com- positions. The Harvard-Newton Bulletins, No. 2, pp. 1-93, Harvard University Press, Cambridge, Mass., Sept. 1914. States the purpose of the study to be the preparation of a scale for the measurement of English composition, particularly in the eighth grade. Gives a detailed account of the preparation of the scales, which are founded on compositions written in the eighth grade in the schools of Newton, Massachusetts. There are four scales, corresponding to the four kinds of composition: description, narration, exposition, and argument. Each scale is made up of six compositions, arranged in steps of 10 % in order of merit from 95 % to 45 %. Each composition in each scale is accompanied by a statement of its merits and defects. A care- fully developed piece of work. The scales are likely to be of value in the last years of the elementary school and the first year in high school. Their method of preparation suggests how similar scales may be pre- pared specially adapted to local needs. Brownell, Baker. A test of the Ballou scale of English composi- tion. School and Society, Vol. iv, No. 103, pp. 938-942, Dec. 16, 1916. Reports a test of the " Ballou-Harvard-Newton scale for English composition," which was carried out by the Bureau of Educational IN ELEMENTARY SUBJECTS 29 Measurements and Standards at the Kansas Normal School in 1915- 16. A careful and critical examination of the claim of the Ballou scale to be considered as a final standard. Earhart, G. and Small, J. English in the elementary school. The Elementary School Journal, No. 16, pp. 32-48, Sept. An experimental investigation of methods for improving English usage and the mechanics of English, conducted for six years in the public elementary schools of Boise, Idaho. A preliminary classification was made of common errors in English speech as follows: verbs, double negatives, pronunciation, pronouns, adverbs, colloquialisms, etc. Pupils were taught to attend especially to their own errors according to this classification. Three tests were given each year on the mechanics of writing, the same test being given in all grades from 3-8. The printed tests and tables of results give a practical way by which any school may test its own work and compare it with that of Boise. Green, J. A. Minimum essentials in elementary school subjects. Journal of Experimental Pedagogy, Vol. in, No. 6, pp. 380- 383, Dec. 1916. An exposition, from an Englishman's point of view, of the work of Dr. E. M. Rice in attempting to measure the efficiency of teaching. Dr. Rice's tests are based upon the reproduction by pupils of a short- story which had been read to them. Hillegas, Milo B. A scale for the measurement of quality in English composition by young people. Columbia University, Teachers College Record, Vol. xiii, No. 4, 191 2. Describes the preparation of and the method of using a scale for the measurement of English composition by young people. The introduc- tion points out the great need of standards for measuring the results in school work. The final scale is composed of ten sample compositions with about equal degrees of merit between successive compositions. These compositions were chosen from about 7000 samples varying from the very poorest to the very best work. Some 450 readers, many of whom were of national reputation as teachers, professors, or writers took part in the selection of material. The steps by which the scale was built up are explained in detail and accompanied by graphs and reference tables. The quality or fitness of the judges is carefully weighed and the probable value of the scale discussed. A well known and valuable contribution, that furnishes the best data by which the reader may himself judge the probable value of the results. The article lacks a summary and is not always in language likely to be clear to the average teacher. 30 BIBLIOGRAPHY OF MEASUREMENT Hosic, James F. The essentials of composition and grammar. School and Society, Vol. I, No. 17, pp. 581-587, April 1915. Gives a brief account of " attempts to measure the effectiveness of school work in English " composition. Compares the Hillegas-Thorn- dike scale with the Harvard-Newton scale, pointing out why the latter is the more effective. Gives a brief account of Franklin S. Hoyt's tests on the value of grammar, and of the tests of T. H. Briggs, which sup- plement Hoyt's tests, and also gives a brief account of W. W. Charter's study of errors in grammar. Hudelson, Earle. Standards and measurements in English composition. Second annual conference on educational measurements, Extension Division of Indiana University, Vol. xiii, No. n, April 191 5. Report of a test in English composition made in the seventh, eighth, and ninth (first year in high school) grades in Bloomington, Indiana. The Harvard- Newton scales were the basis of the test, but topics more suitable to the locality were substituted for those of the Harvard- Newton scales. A valuable adaptation of a standard scale to local conditions. Jenkins, Frances. A test of the ability of children to use language forms. The Journal of Educational Psychology, Vol. vi, No. 6 , PP- 335-344, June 191 5 Report of tests made in Decatur, Illinois, in 191 1, 191 2, and 1914. These tests were given in all grades, 5-8. They consisted of the same or similar dictation exercises given with great care and graded according to (1) number of words misspelled, (2) number of words incorrectly used, (3) sentences correctly written. A clear account of a simple and practical test. Johnson, Franklin W. The Hillegas-Thorndike scale for the measurement of quality in English composition. School Review, Vol. xxi, No. 1, pp. 39-49, Jan. 19 13. The writer presents the tabulated results from the rating of eight selection in English composition by forty-seven high school teachers of English, and sixteen graduate students in the University of Chicago, who used as a basis for the rating the Hillegas-Thorndike scale. Tables are given which show so wide variation in the rating that the writer is led to doubt the practicability of this scale for measuring the merit of so highly complex a product as the composition of young people in their teens. The article is a valuable contribution for the consideration of those who plan to use such scales. IN ELEMENTARY SUBJECTS 3 1 Kayfetz, Isidore. A critical study of the Hillegas composition scale. Pedagogical Seminary, Vol. xxi, No. 4, pp. 559-577, Dec. 1914. A critical review of the scale by the writer to determine its scientific validity and its applicability for administrative and pedagogical pur- poses. He discusses the scale as to aim, means, method, procedure, and results, and concludes that the methods used in deriving the scale are purely statistical, and not experimental-pedagogical; that it is not objective in that it is derived by a statistical study of subjective data; that the qualities of the compositions were determined by literary rather than by pedagogical standards; that the study lacked a pre- liminary series; that the selections do not represent real children's expression; and that the scale regards the work of pupils from the standpoint of the result, without regard to the influences that may prompt the work. Interesting, but misleading. A critical study of the Harvard-Newton composition scales. The Pedagogical Seminary, Vol. xxm, No. 4, pp. 325-347, Sept. 1916. Gives the scale in its entirety, and its history fully under the head- ings: aim, means, methods, and mode of proceedure. After mention- ing the criticisms of the scale by others, the author attacks the scale as to aim, means, methods, proceedure, results, and conclusions. In his final summary he gives nine reasons for declaring the scale invalid from a scientific point of view, and rehearses a long list of dangers that menace the proper development of a true science of education. The article gives an excellent idea of the scale and its development. The criticisms are generally unsound. Noyes, Ernest C. and Miller, Edith. Progress in standardiz- ing the measurement of composition. The English Journal, PP- 53 2 ~53 6 > Nov. 1912. States the need of scales in educational work. Gives an appreciative account of the value and use of the Hillegas scale for measuring the merit of English writing. A statement likely to prove valuable to any one not yet convinced of the value of scales in general. Rice, J. M. Educational research: the results of a test in language. The Forum, Vol. xxxv, pp. 269-293, Oct. 1903. Statement of results of a test involving 8300 pupils in 22 schools distributed through nine cities. The test was given in grades 4-8. The requirement was reproduction of a fable, the same story being used for all grades. Papers were divided into five groups, group one being the poorest and group five the best. Results are given by grades, schools, and cities. A clear statement of a practical test. 32 BIBLIOGRAPHY OF MEASUREMENT Sears, Isabel and Diebel, Amelia. A study of the common mistakes in pupils' oral English. The Elementary School Journal, Vol. xvii, No. i, Sept. 1916. An investigation made in the Cincinnati elementary schools to deter- mine to what extent formal grammar may be of assistance in overcom- ing common errors of speech. For four days all teachers from grade 3-8 inclusive, recorded all errors noticed in the oral language of pupils. These errors were classified as wrong use of verbs, pronouns, mispro- nunciations, etc. The investigation tends to show that (1) most grammatical errors are not dependent for their correction upon a knowledge of grammar; (2) except as a preparation for the study of a foreign language the following subjects are of little use to a child in the grades: formal parsing, classification of adverbs, and adjectives, the moods, conjugation. Stark, William E. Measurement of eighth grade composition. School and Society, Vol. 11, No. 22, pp. 208-216, Aug. 7, A careful study of the writing ability of eighth grade pupils in the public schools of Hackensack, N. J. The aim was (1) to estimate the writing ability of the pupils in 191 1 and (2) to compare the composi- tions written in 191 1 with similar ones written in 1914. The examiners tried both the Harvard-Newton scale and the Hillegas scale, but the final ranking was based on a special scale prepared from the composi- tions written in 191 1. Results are given in terms of the Hillegas scale also. The compositions written in 1914 show, on the whole, little improvement over those of 191 1. A practical and careful attempt to use standard scales. Stoddard, W. E. A comparison of the Hillegas and Haroard- Newton scales in English composition. The Pedagogical Seminary, Vol. xxiii, No. 4, pp. 498-501, Dec. 1916. Five hundred thirty-nine compositions written by pupils in grades 5-9 inclusive in Orono and Old Town, Maine, were graded by the Hille- gas and by the Harvard-Newton Scales. The findings are shown by table and graph. The author thinks that the close correlation of the results tends to show the reliability of measurement scales generally, that both scales efficiently measure the mechanics of composition, and that the Harvard-Newton scale more accurately measures the literary value. IN ELEMENTARY SUBJECTS 33 VII GRAMMAR Charters, W. W. English Grammar. Bulletin of the Univer- sity of Missouri, Columbia, Mo., Vol. xvi, No. 2, pp. 1-45, Jan. 1915. An investigation to determine what errors connected with grammati- cal rules were made by the children of Kansas City and to determine on this basis what rules should be taught to children in the elementary grades. A valuable report. Maxwell, William H. English grammar. Proceedings, National Education Association, pp. 565-573, 191 5. A discussion of standards and quantitative tests for English gram- mar. States that the aims of grammatical study are (1) orderly and logical thinking, (2) power to interpret language, (3) habit of correct expression, (4) knowledge of grammatical facts, (5) organization of a basis for study of other languages. Suggests with much detail kinds of questions to test the pupils' knowledge. The tests proposed are of the nature of regulation examination questions. Article has value for its suggestiveness. Rapeer, Louis W. The problem of formal grammar in elementary education. Journal of Educational Psychology, Vol. iv, pp. 125-137, March 1915. Report of tests made to determine the utility of grammar in the elementary school. Follows the method of Franklyn G. Hoyt (see Teachers College Record, Nov. 1906). Gives a clear statement of the aims and method of Hoyt and is intended to be a verification of his tests. Rapeer's deductions agree with those of Hoyt, tending to show that there is little correlation between knowledge of grammar, ability in composition, and ability in interpretation of literature. A kind of article of particular value in educational measurement, being a report on an attempt to test the accuracy of a previous investigation. Appen- dixes are of practical value, containing questions for tests and giving, in tabular form, results of tebts. Starch, Daniel. The measurement of achievement in English grammar. The Journal of Educational Psychology, Vol. vi, No. 10, pp. 615-626, Dec. 1915. Gives scales for measurement of (1) grammatical correctness, (2) knowledge of formal grammar. (3) punctuation. Describes with some 34 BIBLIOGRAPHY OF MEASUREMENT detail methods of preparation of the scales. Explains how they are to be used and points out their defects. The scales for testing grammati- cal correctness seem fairly comprehensive; those for testing gram- matical knowledge are inadequate. Thorndike, Edward L. Note on the significance and use of the Hillegas scale for measuring the quality of English composi- tion. English Journal, Vol. n, Nov. 1913, pp. 551-561. The writer indicates that the purpose of the scale is to measure roughly the difference in paragraph writing of high school students and that rated as especially good performances of recognized masters of English prose. The second purpose is to measure the amount of error to be expected in grading specimens of English writing by a scale. He points out the fact that the errors in the use of the scale at fiist will be large, but will diminish with practice, and that the errors made by teachers using the scale will be smaller than those made by teachers grading for general merit. The Hillegas scale is included in the article. VIII ARITHMETIC Brown, J. C. An investigation on the value of drill work in the fundamental operations of arithmetic. Journal of Educational Psychology, Vol. 11, No. 2, pp. 81-88; Vol. 111, No. 9, pp. 485-491, Feb. 191 2. An experiment given under same conditions as Stone tests in five minute drill-periods upon the fundamentals to find the value of short drill exercises. These were found to be beneficial in the 6th, 7th, and 8th grades. Proved by comparison of sections subjected to the drill with sections not subjected to the drill. Benefit most marked in 6th grade and least marked in 8th. Buckingham, B. R. Notes on the derivation of scales in school subjects: with special application to arithmetic. The 15th Yearbook of the National Society for the Study of Educa- tion, pp. 23-40, 1916. A preliminary report on an arithmetic investigation in New York City given to develop standardized material and to derive scales. Accompanied by tables and figures. Suggestive to those intending to make scales. IN ELEMENTARY SUBJECTS 35 Cole, Lawrence W. Adding upward and downward. Journal of Educational Psychology, Vol. in, No. 2, pp. 83-94, Feb. 1912. A series of tests to determine the difference in accuracy and time in adding the same problems upwaid and downward. They weie added more rapidly and less accurately when added upward than when added downward. Counting lines to left was also slower and more accurate than counting to the right. The factor of habit seemed to produce a saving of time at the expense of accuracy. Courtis, S. A. Measurement of growth and efficiency in arith- metic. Elementary School Teacher, Vol. x, No. 2, pp. 58- 74, Oct. 1909; Vol. x, No. 2, pp. 177-199, Dec. 1909; Vol. xi, No. 4, pp. 171-185, Dec. 1910; Vol. xi, No. 7, pp. 360- 370, March 1911; Vol. xn, No. 3, pp. 127-137, Nov. 1911. Series of experiments to establish a standard from which to measure the success or failure of a reorganization of the mathematics courses in the writer's charge, and to trace the development of ability in arith- metic from the primary grades through the high school. Illustrated by many tables, charts, plots, and samples tests. An argument for a real science of teaching. The Courtis Tests in arithmetic enjoy a well- deserved reputation as the most successful scientific measures in any subject. Report on the Courtis tests in arithmetic. Interim Report, Pt. 11, Sec. D, New York, School Survey, New York City, 1911-12. A complete report on the Courtis Tests as given in New York. Explains reasons for their being and relates how they were formed. Contains many tables and graphs of results of Tests. Discusses the results and tells values to be derived from Tests. This report is also included in the final edition of the New York City Survey. Mr. Courtis issues from 82 Eliot St., Detroit, a manual of his tests, and accounts of results from the use of them. The reliability of single measurements with standard tests. Elementary School Teacher, Vol. xiii, 326-352, March 1913- A criticism of the article published in the Elementary School Teacher in Oct. 191 2, giving the results of a study made with the Courtis Tests No. 1 and drawing conclusions therefrom. The writer takes issue with certain methods of the study and with certain of the conclusions. He declares that the test was put to a use for which it was not intended, states the real purpose of the test and the good effects from its proper 36 BIBLIOGRAPHY OF MEASUREMENT use. He shows by tables and graphs the results of testing 55,200 pupils in many sections of the country, and compares these results with the results of the test before mentioned. He concludes that first scores, when the effects of other factors are at a minimum, measure more reliably than any others the actual difference in readiness of associa- tion, that the " hypothetical first scores," considered by those who gave the test as the true measures of the initial ability of the individ- uals, are not such in reality, and that when properly used, test No. 1 does measure a definite fundamental ability in arithmetic. Of interest to school executives and all upper grade teachers in arithmetic. Courtis, S. A. The reliability of single measurements with stan- dard tests. Elementary School Teacher, Vol. xm, pp. 486- 504, June 1913. The conclusion of a criticism, begun in the Elementary School Teacher of March 1913, of a recent article on the reliability of single measurements. The writer brings forth evidence from tests to show that the results from single measurements will ordinarily not vary more than five points from the true score, that Courtis Test, No. 1, when rightly used, does measure one of the factors concerned in column addition, and that the results, rightly interpreted, have a diagnostic value. He argues in favor of measuring the efficiency of the whole school to reveal individual variations and needs and to disclose gross inefficiency, to the end that schools may be organized so that they may deal with large masses yet give to each individual the special attention he may need. Fully illustrated with graphs and tables. An article of fundamental importance in the theory of educational measurement and in the development of the technique of individual instruction without abandonment of class work. Graham, J. W. A measure of progress in the mechanical operations of arithmetic. Elementary School Teacher, Vol. xiv, No. 7, pp. 348-349, March 1914. The use of the pendulum suspended at the front of the class room is suggested as a means of accelerating oral work in the mechanical operations of arithmetic. The device is simple and may be easily tried in any school room. Gray, P. C. Norms of performance in the fundamental processes of Arithmetic. Journal of Experimental Pedagogy, Vol. 11 No. 5, pp. 310-317, June 1916. This is a discussion of the mode of procedure and of tests in the fundamental processes of arithmetic given in the City of Leeds by the writer. The tests are the same as those given by Dr. Ballard in the elementary schools of London and reported in the March 191 5 and Dec. 1914 numbers of the Journal of Experimental Pedagogy. IN ELEMENTARY SUBJECTS 37 Haggerty, M. E. Arithmetic: a cooperative study in educational measurements. Indiana University Bulletin, Vol. xn, No. 18, PP- 3 8 5-5 o8 > March 1915. A report showing by tables and graphs the results in the Courtis Standard Tests obtained in twenty cities of Indiana. The important facts brought out by the tests are noted and an attempt is made to point out the causes for some of them. Hahn, H. H. and Thorndike, E. L. Some results of practice in addition under school conditions. Journal of Educational Psychology, Vol. v, No. 2, pp. 65-83, Feb. 1914. This is a report of a series of tests in the addition of columns of ten one-place numbers given to fourth, fifth, sixth, and seventh grade pupils for the purpose of determining the results of practice upon (1) the amount of improvement, (2) the rate of improvemnent, (3) im- provement in rate, (4) effect of period-length upon the rate of improve- ment, (5) individual differences in rate of improvement, (6) individual differences in initial efficiency, and (7) the relation of rate of improve- ment to initial ability. The results are carefully tabulated and show (1) that, properly administered, speed-drills will improve accuracy as well as speed, (2) an enormous overlapping of the different grades in respect to ability in adding and, (3) that equal practice of groups of different initial ability increases the initial differences. Jessup, W. A. Current practices and standards in arithmetic. N. E. A. Dept. of Superintendence, igi 5. Journal of Educa- tion, Vol. lxxxi, p. 272, March 11, 1915. A short article summarizing briefly the conclusions to be reached from the results of tests made by Rice, Stone, Van Houten, and Courtis. Also briefly presents suggestions of the investigators. Lane, Henry A. Standard tests as an aid to supervision. The Elementary School Journal, Vol. xv, pp. 378-386, March Courtis Tests, series B, were given to about 500 pupils in Houghton, Mich., in Oct. 1914. From these tests, standard of achievements were formulated for each grade in each of the fundamental operations. The tests were repeated in June, and tended to prove that reasonable class standards could be obtained in a very short time. They also showed the need of attacking the problem of securing individual attainment of reasonable standards. 38 BIBLIOGRAPHY OF MEASUREMENT Mead, Cyrus D. and Sears, Isabel. Additive subtraction and multiplicative division tested. Journal of Educational Psy- chology, Vol. vii, No. 5, pp. 261-270, May 1916. A discussion of tests in additive subtraction and multiplicative division given to a second and third grade respectively. A " take away " group in subtraction and an " into " group in division were also conducted in separate classes of the second and third grades and the results compared with the new methods. The classes taught by the old methods were slightly in the lead. Shows that the new methods must be administered in a different manner. Of interest to all who contemplate using new fundamental methods. Monroe, Walter S. A report of the use of the Courtis standard research tests in arithmetic in twenty-four cities. Bulletin, Vol. in, Kansas State Normal School, Emporia, pp. 1-94, Jan. 8, 1916. An article describing how these tests were given in the 24 cities and giving tables and graphs of the results. Conclusions to be made from a study of the tables are also given. Munson, Kezia. The use of tests in arithmetic. School and Home Education, No. xxxiii, pp. 301-302, April 1913-14. Explains the writer's difficulties in finding suitable standard tests in arithmetic. Shows why the Courtis Tests in arithmetic met the author's requirements. Suggests, somewhat vaguely, how a teacher may make arithmetic tests of her own and use the Courtis system of grading. Otis, Arthur S. and Davidson, Percy, E. The reliability of standard scores in adding ability. Elementary School Teacher, Vol. xiii, No. 2, pp. 91-105, Oct. 1912. An account of an investigation, conducted with 270 eighth grade pupils in San Jose, Cal., to learn (1) the approximate reliability of a single score as a measure of individual ability to make addition com- binations in writing, (2) the number of scores required of an individual in order to obtain a reasonable degree of reliability, and (3) the size of the group that may be measured with reasonable accuracy by means of one score from each member of the group. Twenty-five trials were given with Courtis Test No. 1 and four rearrangements of it under carefully controlled conditions. The results are shown in tables and graphs, and ably discussed. The conclusions are that a single test is unreliable; that twenty-five trials are probably necessary correctly to measure the ability of eighth grade pupils to write addition combina- tions; that groups of 25 may be tested with an error of 1.7 combina- IN ELEMENTARY SUBJECTS 39 tions, groups of 50 with 1.2 and groups of 100 with .7; that the useful- ness of the Courtis Test No. 1 as a test in addition may be questioned; and that it is hardly possible to explain the improvement noted as being due to increased readiness of mental association. Of interest to all arithmetic teachers and school executives. For reply, see above under Courtis, S. A. Phelps, C. L. A study of errors in tests of adding ability. Ele- mentary School Teachers, Vol. xrv, pp. 29-39, Sept. 1913. The data was collected from giving twenty-five arrangements of Courtis Test No. 1 to 270 eighth grade pupils in San Jose, Cal. The purpose was to determine the relative difficulty of the various com- binations, and what kinds of errors are made in writing the answers to the combinations. Mistakes were found to increase in number as the combinations increased in size, and certain type errors were discov- ered. Speed and accuracy are shown to have a negative correlation. A valuable contribution. Of interest to all teachers of arithmetic. Philips, Frank M. A comparison of the work done in successive minutes or a ten minute practice period in the fundamentals of arithmetic. Journal of Educational Psychology, Vol. vn, No. 5, pp. 271-277, May 1916. An investigation to determine how far the results of a one-minute test in the fundamentals of arithmetic are an accurate measure of a pupil's ability, and to what degree the results of a more extended test may be expected to vary from the results of the first minute. The Courtis Tests were used. A one-minute test is shown to be reliable. Rice, J. M. Rice's reasoning tests, a test in arithmetic. The Forum, Vol. xxxiv, pp. 281-297 an d 437-452, 1902. A series of tests given by the writer to 6000 children in grades 4-8 in seven cities, together with results obtained. He shows that the mechanical side of arithmetic is closely related to the thought side, that the best thinking is accompanied by the fewest mechanical errors. Starch, Daniel. A scale for measuring ability in arithmetic. Journal of Educational Psychology, Vol. vn, No. 4, pp. 213- 222, April 1916. A scale designed to measure ability in arithmetical reasoning such as is involved in the solution of concrete problems ; arranged in the order of steps of increasing difficulty from one to fifteen. Shows individual differences, overlapping grades and sex differences in solving problems. Of value to those who are using scales in aiithmetic. 40 BIBLIOGRAPHY OF MEASUREMENT Taylor, E. H. A comparison of the arithmetical abilities of rural and city school children. Journal of Educational Psychology, Vol. v, No. 8, pp. 461-466, Oct. 1914. This is a comparison of the scores in Courtis' Tests made by the pupils in twenty-eight rural schools in Illinois and the scores of city school pupils as given by Courtis. The scores of the rural children were found to be from one to two grades lower than those of the city children, showing that accuracy is too little emphasized in rural schools. Thorndike, Edward L. Measurements of ability to solve arith- metical problems. Pedagogical Seminary, Vol. xxi, No. 4, pp. 495-503, D ec 1914. This article consists of the tabulated results obtained from giving a test composed of four arithmetical problems to forty-five hundred sixth, seventh, and eighth grade pupils to determine their ability to solve in a given time problems which increased in difficulty. The results are tabulated to show sex differences in problem solving within a grade, and also compare the achievements of the three grades, as well as the work of young and old pupils in the same grade. It is a valuable study in individual and sex differences and retardation, and may be used to advantage in any school system. Woody, Clifford. Measurements of some achievements in arithmetic. School and Society, Vol. iv, No. 86, pp. 299-303, Aug. 19, 1916. This article describes the process of developing scales for measuring the achievements of pupils in arithmetic. It is valuable in its reference to these particular scales and interesting in its incidental comment upon school measurements in general. Zeidler, Richard. Tests of efficiency in the rural and village schools of Santa Clara, Cal. Elementary School Journal, Vol. xvi, No. 10, pp. 542-555, June 1916. A study to measure results of arithmetic teaching in rural and village schools and to ascertain the status of such schools in the teaching of the fundamentals in arithmetic as compared with some other western schools. One-room rural schools obtain better results in accuracy than do two-room rural schools, three-room rural schools obtain better results in accuracy than the other three; village schools fall below the rural schools in accuracy, in number of examples attempted, and in number correctly worked. All the Santa Clara schools fell below the median scores of western cities. Illustrated by eight tables, also com- pared with Courtis Tests. The findings were contrary to expectations and are a strong argument for using scales and tests for measuring product and for supervision. IN ELEMENTARY SUBJECTS 4 1 IX GEOGRAPHY Witham, Ernest C. A minimum standard for measuring geog- raphy. American School Board Journal, Vol. l, No. i, PP- i3-!4, Jan. 1915. A test for measuring sixth grade geography, or the geography of the United States, together with results obtained from its use. It consists almost wholly of the location of geographical features and the indica- tion of the various industrial regions on outline maps. It is of value within narrow limits, but not a comprehensive geographical test. X DRAWING Childs, H. G. Measurement of the drawing ability of 21JJ children in the Indiana city school systems by a supplemented Thorndike scale. Journal of Educational Psychology, Vol. vi, No. 7, pp. 391-408, Sept. 1915. Measurement to determine (1) growth in ability from grade to grade; (2) standards or norms of ability for each grade; (3) limita- tions of the scale in actual use; (4) what administrative problems would be brought to light which might be desirable for supervising authorities to know. Thorndike scale was supplemented to have greater degree of uniformity by adding samples where units were not of equal value. The tables of distribution and curves show that the average child develcps more ability in drawing before entering school at the age of six or seven years than he does in the entire eight years of his elementary school course. Thorndike, Edward L. The measurement of achievement in drawing. Teacher's College Record, Vol. xiv, No. 5, pp. 1- 38, Nov. 1913. A scale by which achievement and improvement in drawing can be measured. Defines by samples different degrees of merit in drawings and represents them by numbers. By the use of this scale a teacher can tell a pupil how well he has drawn or how well he should have drawn. A supervisor can tell teachers similarly what he expects of children in a given grade. Comparisons of achievement can also be made from time to time. 42 BIBLIOGRAPHY OF MEASUREMENT XI ADDRESSES The following information is taken from A Descriptive List of Standard Tests, by William S. Gray, Elementary School Journal, Vol. xvn, Sept. 1 91 6. Thorndike's Drawing Scale. Bureau of Publications, Teachers' College, Columbia Uni- versity, New York City. Ayres' Handwriting Scale. Russell Sage Foundation, Division of Educational Research, New York City. Freeman's Handwriting Scale. Houghton, Mifflin Company, Boston, Mass. Gray's Score Card for the Measurement of Handwriting. C. T. Gray, University of Texas, Austin, Texas. Thorndike's Handwriting Scale. Bureau of Publications, Teachers' College, Columbia Uni- versity, New York City. Starch's English- Grammar Tests. Daniel Starch, University of Wisconsin, Madison, Wis- consin. Thorndike's Scale Alpha for Measuring the Understanding of Sentences. Bureau of Publications, Teachers' College, Columbia Uni- versity, New York City. Harvard-Newton Composition Scale. Harvard University Press, Cambridge, Mass. Hillegas' Composition Scale. Bureau of Publications, Teachers' College, Columbia Uni- versity, New York City. Ayres' Spelling Scale. Russell Sage Foundation, Division of Educational Research, New York City. Buckingham's Spelling Ability. Bureau of Publications, Teacher's College, Columbia Uni- versity, New York City. IN ELEMENTARY SUBJECTS 43 Courtis Standard Research Tests. S. A. Courtis, 82 Eliot Street, Detroit, Michigan. Jones' Concrete Investigation of the Material of English Spelling. University of South Dakota, Vermilion, South Dakota. Starch's Spelling Tests. The Macmillan Company, New York City. Cleveland Arithmetic Tests. Dr. Charles H. Judd, School of Education, University of Chicago, Chicago, Illinois. Starch's Arithmetic Tests. Daniel Starch, University of Wisconsin, Madison, Wisconsin. Stone's Arithmetic Tests and Stone's Arithmetical Abilities, and Some Factors Determining Them. Bureau of Publications, Teacher's College, Columbia Uni- versity, New York City. Brown's Silent Reading Tests. H. A. Brown, Bureau of Research, 25 Capitol Street, Con- cord, N. H. Gray's Reading Tests. William S. Gray, School of Education, University of Chicago, Chicago, Illinois. The Kansas Silent Reading Tests. Bureau of Educational Measurements and Standards, Kansas State Normal School, Emporia, Kansas. Jones' Scale for Teaching and Testing Elementary Reading. Rockford Printing Company, Rockford, Illinois. Starch's Silent Reading Tests. Daniel Starch, University of Wisconsin, Madison, Wiscon- sin. Thorndike's Visual Reading Tests. Bureau of Publications, Teachers' College, Columbia Uni- versity, New York City. INDEX Ayres, Leonard P., 3, 19, 24. Ashbaugh, E. J., 19. Babbitt, F., 9. Bagley, W. C, 3. Ballou, F. W. ; 4, 28. Birch, T. B., 5. Bliss, D. C, 5. Breed, F. S., 20. Brown, G. A., 5. Brown, H. A., 12. Brown, J. C, 34. Brownell, B., 28. Buckingham, B. R., 25, 34. Charters, W. W., $^. Childs, H. G., 9, 41. Cole, L. W., 35. Colvin, S., 5. Courtis, S. A., 6, 12, 13, 35, 36. Cubberley, E. P., 10. Culp, V., 20. Davidson, P. E., 17, 38. Diebel, A., 32. Earhard, G., 29. Freeman, F. N., 20. Gilliland, A. R., 17. Graham, J. W., 36. Gray, P. C, 36. Gray, T. C, 21. Gray, W. S., 10, 13, 14. Green, J. A., 29. Haggerty, M. E., 14, 37. Hahn, H. H., 37. Harlan, C. H., 10. Hillegas, M. B., 29. Hosic, J. F., 30. Hudelson, E., 30. Jenkins, F., 30. Jessup, W. A., 37. Johnston, J. H., 10. Johnson, F. W., 30. Johnson, G. L., 21. Johnson, H., 21. Jones, R. G., 15. Jones, W. F., 25. Judd, C. H., 11, 15, 16. Kayfetz, I., 31. Kelley, F. J., 16. King, I., 21. Lane, H. A.. 37. Lewis. E. E., 25. Littell, H. V., 6. Manahan, J. L., 11. Manuel, H. T., 22. Maxwell, W. H., 6, 33. Mead, C. D., 16, 38. Miller, E., 31. Monroe, W. S., 7, 38. Munson, K., 38. Noyes, E. C, 31. Oberholzer, E. E., 16. Otis, A. S., 17, 25, 38. Pearson, H. C. ; 27. Phelps, C. L., 39. Philips, F. M., 39. Pitner, R., 17, 22. 46 INDEX Rapeer, L. W., 33. Rice, J. M., 31, 39. Richards, A. M., 17. Sackett, L. W., 22, 26. Scott, F. N., 7. Sears, I., 32, 38. Sears, J. B., 26. Small, J., 29. Smith, J. H., 7. Springer, I., 11. Starch, D., n, 18, 23, 26, 33, 39. Stark, W. E., 32. Stockton, J. L., 7. Stoddard, W. E., 32. Strayer, G. D., 8, 11. Studebaker, J. W., 27. Suzzalo, H., 27. Taylor, E. H., 40. Thorndike, E. L., 9, 18, 23, 27, 34, 37, 40, 41- Tidyman, W. F., 27. Tobey, L., 24. Uhl, W. L., 18. Wallin, J. E. W., 28. Witham, E. C, 24, 41. Woody, C, 40. Zeidler, R., 19, 40. UCLA-Young Research Library Z5814.P8 D4 yr L 009 514 789 8 UC SOUTHERN REGIONAL LIBRARY FACILITY AA 001 242 330 7 ■: -