Abstract
Studies evaluating the trainability of intelligence have been dependent, in the main, upon a relatively small number of contemporary measurement procedures largely derived from a common base. A brief review of these tests is provided, and it is concluded that the procedures possess a number of characteristics that make the evaluation of change in intelligence a tenuous proposition. The most serious concern is seen as revolving about the evidence of test validity. The case is advanced that at least 4 conceptually distinct types of validity should be required of these tests and that in most instances this is now assumed rather than demonstrated. As a result, many of the conclusions drawn from efforts to evaluate change in intelligence may be erroneous; hence, the trainability of intelligence cannot be fully judged from the existing research in which these measures are used.
Get full access to this article
View all access options for this article.
