Abstract
The validation procedure recommended by the College Entrance Examination Board (CEEB) for their foreign language placement tests determines whether individuals (1) with high and low grades in the same course, and (2) from courses at different levels, score differentially on the CEEB tests. If both these differences are statistically significant, high predictive validity is inferred. The alternative proposed in this paper is to ask the student directly upon completion of the language course to which he had been assigned, whether he had been correctly placed by the CEEB test. High predictive validity is inferred if a large proportion of the students believe that they were correctly placed.
CEEB placement was evaluated by the alternative validity procedure and was found to be significantly more valid than was a second placement method in which a regression equation was used to predict level of placement from background variables. It is argued that the alternative validation procedure complements CEEB validation because in not relying on grades, it provides different information, serves different purposes, and facilitates comparisons between different placement schemes.
Get full access to this article
View all access options for this article.