COURTIS' TESTS. For several years Stuart Appleton Courtis (q.v.), in the discharge of his duties as head of the department of science and mathematics in the Detroit Home and Day School, made use of comparative tests for purposes of supervision. During this time a gradual evolution of these tests took place; and they finally assumed the shape of an attempt to apply scientific methods to the teaching of the fundamentals in reading, writing, arithmetic and English subjects so as to make possible rapid and reliable measurements of the effects of teaching through testing work practicable under classroom conditions. After considerable com parative investigation along experimental lines, *Series A,' the first of the Courtis' Tests, was published and 18,000 copies were sold at cost or given away as a preliminary to a compara tive investigation to determine standard scores. Failure after failure met these attempts to con trol individual variation; but a test of 5,000 children in Detroit and 33,000 in New York schools for the Hanus Committee on School Inquiry provided the necessary scope and op portunity for the standardization of the tests in 1911. The result of these tests shows that •the supreme factor in education is the variation of the natural abilities of children.* Courtis came to the conclusion, therefore, that •new educational methods that would give each child a chance to develop in his own way and along his own lines would have to be invented.' He experimented along the same lines in the Boston schools in 1912. In October and again in March 20,000 children were tested and several experi mental methods of giving assistance to indi viduals were put into practice in the schools tested. These original tests were applied prin cipally to the teaching of arithmetic; but as it was found that this was not a broad enough basis on which to work, other tests were applied to the teaching of reading and writing.
The Courtis' Tests are divided as follows: (1) Diagnostic. To make evident the actual conditions existing in schools, classes and indi viduals that the weaker points may be noted, cause determined and remedies devised. (2)
Scientific. To discover the natural laws of men tal developments operative in school work. (3) Experimental. To make possible control experi ments that will settle all questions of educational procedure upon a fact basis, through scientific determination of the efficiency of different methods. (4) Supervisory. To secure the in formation needed in setting standards for the guidance of teachers and schools and in deter mining whether or not standards already set are being attained. The Courtis tests have been defined as *a correspondence course in educa tional measurements.° By means of printed tests, folders of instruction, records and graphic sheets superintendents and teachers are enabled to carry on experimental work with a minimum of effort. The tests *aim to secure information for the formation of reasonable aims in terms of objective standards; to measure the efficiency of methods designed to produce desired results; to obtain knowledge of the factors and natural laws of teaching and learning; and to enable comparisons of school with school, or teacher with teacher, to be made upon a scientific, im personal, objective basis.* The Curtis tests differ from those of Binet (q.v.) in that they are more scientific, more carefully worked out by means of co-operation; and that there is but one set of tests for children of all ages. The giving of exactly the same tests under identical conditions to all persons examined is an essen tial feature of the method and distinguishes it from previous methods. The standardized tests cover arithmetic, writing, dictation, reading, spelling, punctuation, reproduction and the com parison of all the work done by the pupil. Con sult Courtis, S. A., 'Manual of Instructions for Giving and Scoring the Courtis' Standard Tests> (Detroit 1914); 'Better Teaching in Arithmetic' (Detroit 1913); 'Teachers' Man ual' (Yonkers 1914).