The Factor of Verbalization—A Further Limitation of the Multiple Choice Test

Autor: T. J. Eskridge
Rok vydání: 1942
Předmět:
Zdroj: The Journal of Educational Research. 35:326-333
ISSN: 1940-0675
0022-0671
Popis: The multiple choice test of the usual type is frequently used to measure understanding. When this type of test is so used, it is tacitly as sumed that, accidental success or failure barred, (1) a subject who selects the right alternative understands, and (2) one who selects an incorrect alternative or (3) one who omits an item altogether does not understand. Test users have no doubt many times questioned the validity of these as sumptions but apparently little effort has been made to study it experi mentally. The data of the present study relate to the validity of the third assumption, namely, the assumption that a subject who intentionally omits an item in the usual type multiple choice test does not understand. The specific purpose of the study is to show how the scores derived from the usual type of multiple choice test for measuring understanding of terms may be affected by the intentional omission of items on the part of children who have correct meanings for the terms omitted. Approximately 400 subjects, pupils in Grades IV, V, VI, and VII in the public schools of Greenwood, South Carolina, were tested on the mean ings of 135 geographic terms, 60 of which were common to the material studied by each of the four grades. The multiple choice test employed was unusual in that the children were given space in which to define terms when they did not recognize correct meanings among the alternatives offered. In order to prevent the subjects from assuming that one of the given alternatives must be correct, for ten of the 60 terms none of the alterna tives was correct. The subjects were warned that items of this kind had been included in the test, and one of the sample items in the pretest was of this type. The subjects were cautioned also not to supply an original definition of a term unless they felt sure that none of the meanings which were given was right. For the purpose of studying the results from children at different levels of geographic attainment each grade was divided according to the 326
Databáze: OpenAIRE