The Reliability and Validity of Four Types of Vocabulary Tests

Numerous methods for testing vocabulary have been suggested dur ing the past few years, but no attempt to determine objectively the best means for securing a measure of vocabulary seems to have been made. Several authors report the reliability and validity of their own tests, but the conditions under which these data are determined vary so much from test to test that they do not justify the drawing of conclusions concerning the best type of test. Dolch1 has made a very satisfactory analysis of the more common types of tests used, and he presents a good discussion of the merits and demerits of each type ; but the major por tion of this discussion is based upon personal opinion, or psychological analysis rather than experimentation. It is the purpose of this paper to report the results of an investigation of the relative reliability and validity of the four types of tests most commonly used for measuring vocabulary. Most of the suggested methods for testing vocabulary and all of the published tests that the writer knows of may be classified under one of the following types: (1) Identification tests, where the words are pre sented orally, visually, or by both methods to the individual subject and the subject asked to respond by orally identifying the word through definition, use, synonym, illustration or otherwise. Terman's Word List, used in connection with the Stanford Revision of the Binet-Simon Tests is the classical example of this type. (2) Multiple-response tests, where along with the test word are presented three, four, or more words or phrases, one of which is the correct definition of the word. The subject identifies the correct definition usually by drawing a line under it. Two of the most commonly used tests, the Thorndike Test of Word Knowledge and the Inglis Test of English Vocabulary, well illustrate this type. (3) Matching tests, where the test words are presented in columns and along with these word? the definitions are presented in a parallel column, but arranged in different order ; the subject being required to arrange the definitions in the same order as the test words. The Detroit Word Recognition Test and the Southington-Plymouth English Vocabulary