Com-pletion-Test Language Scales
REVIEWS AND CRITICISM. :Author: Marion Rex Trabue, Ph.D. Teachers College, Columbia University Contributions to Education, No. 77, 1916. Pp. IX + 118.
In this monograph Mr. Trabue reports an attempt to derive one or more scales for the measuring of ability related to language. He does not define “language” in any strict sense, as he believes that on the whole ability to complete his sentences successfully is very closely related to what is usually called “language ability”.
Mr. Trabue finds that his scales of mutilated sentences have an advantage over mutilated paragraphs in that they may be used a second time to reveal improvement. The story of a paragraph is not readily forgotten, and therefore improvement on a second trial need not necessarily denote an improvement in ability. But the sentences deal with a wide variety of subjects, none of which is interesting in itself and consequently they are not readily retained in memory. Using the same scale twice is rarely necessary, however, as the author of the sentences has prepared several other scales (B, C, D, E) which for all practical purposes are equal and interchangeable. Likewise scales I and K are fairly well matched, and L and M differ but slightly in their degree of difficulty. The sentences possess an added advantage over the mutilated paragraph in that they are arranged as steps on a scale with approximately equal intervals between the steps. Thus an individual is not discouraged at having a task set for him which is too hard, nor is he disgusted at being asked to do child’s play. Furthermore it is easier to judge whether a sentence is “right,” “almost right,” or “wrong” than it is to score a paragraph. The comparison of two performances on scales of equal difficulty is a more valid indication of an individual’s ability than his performance on a passage of prose.
The same amount of time is allowed to all pupils. Each pupil is scored two credits for every sentence perfectly completed, and one credit for every sentence almost perfectly completed. This automatically takes account of the time spent on the work, the quality of work done, and the difficulty of the tests one is capable of doing.
From a preliminary list of fifty-six sentences, which he calls the “graded series” Mr. Trabue has derived his Scales A, B, C, D, E, I, K, L, and M. In Scale A the two sentences for each grade are the ones which proved to be just difficult enough to be done by 50 per cent of the children in that grade. The relative distance between grades is calculated on the P. E. or the median deviation from the median of a grade distribution. Scales B and C are chosen entirely from the evaluated sentences of the graded series and consist of ten sentences each, the steps being about 1 P. E. above each other. Scale C is, on the whole, about 4 P. E. more difficult than Scale B. Seven minutes are allowed on each of these scales, and very careful directions are given in order to make sure that the child understands the requirements. Scales D and E are practically of the same difficulty as B and C, but each contains four sentences which do not appear in the graded series. Scales I and K are intended for adults and only five minutes are allowed for their completion. The procedure is the same as for the other scales.
Mr. Trabue’s report contains results from over six thousand children, primarily from New York and New Jersey, although records were obtained from classes in Illinois, Indiana, Michigan, Missouri, and Pennsylvania. The results indicate no great difference in the abilities of children in the same grades in different states. The results from Scale A show that each successive grade makes a higher score than the grade just previous, and that the range of the middle 50 per cent likewise moves steadily upward. This is naturally to be expected, although the relative amounts of progress between grades are unusual and need careful examination. Boys seem to make a somewhat lower median score than the girls in the same grade although the difference is small and the amount of overlapping enormous. There is also a distinct tendency for the older pupils in the grade to make lower scores than the younger pupils. Mr. Trabue suggests that the reason for this is that teachers are possibly inclined to promote dull pupils on age and retard bright pupils because of their youth. To say that a child is twelve years old apparently signifies far less concerning his ability to complete sentences than to say he is in the sixth grade. Mr. Trabue feels that a systematic study should be made, correlating his language scales with other tests of knowledge and ability. Because he believes his sentences are more accurate than mutilated paragraphs, he thinks that correlations with them will be higher than with the prose passages. No such study has as yet been published. Such correlations as have been made “have been worked out incidentally, sometimes as a matter of curiosity and sometimes in connection with other studies.” They are reported by Mr. Trabue in an appendix. Natalie A. Bassett.
Disclaimer
The historical material in this project falls into one of three categories for clearances and permissions:
Material currently under copyright, made available with a Creative Commons license chosen by the publisher.
Material that is in the public domain
Material identified by the Welcome Trust as an Orphan Work, made available with a Creative Commons Attribution-NonCommercial 4.0 International License.
While we are in the process of adding metadata to the articles, please check the article at its original source for specific copyrights.