A Latin Comprehension Test

Abstract

The development of a new type of Latin test perhaps deserves a special record, particularly as it has seemed to find favor with the teachers and specialists in the fields both of Latin and of measurements. The construction of the test was undertaken by the authors for the American Classical League with the support of the General Education Board.1 This Latin Comprehension Test2 had its origin in the authors' belief that one of the objects of the study of any language, for eign or vernacular, is to comprehend clearly and rapidly the mean ing of the printed page. The number of silent-reading tests now available, the emphasis on silent-reading methods, and the^ development of reading tests in special subjects such as history and general science indicate the general acceptance of this belief for English. In planning the test it was at first thought best to construct four parts, one for each year of the high-school Latin course. After careful consideration it was concluded, however, that a single test, flexible enough for use by all, was more desirable since it readily made possible a comparison between the results of the teaching in the different years. Several types were con sidered: (1) a test consisting of numerous sentences, (2) a test made up of a single long passage, or (3) one composed of a number of short paragraphs. The first was rejected because it failed to test the pupils' ability in reading connected discourse. The second was impossible in a test which was to be given to pupils who had studied Latin for varying periods of time, be cause it would necessitate the preparation of special material in stead of the use of existing tests. The result would be at vari ance with the classroom conditions. The third type, a compro