Abstract
This article contains a review of issues and documentation concerning the possibility of "measuring " the intelligence of students. The construct validity of "intelligence, " the role of cultural context as a modifier of the meaning of test results, and the lack of meaningful predictive validity of IQ tests are discussed. A discussion of the utility of the intelligence construct and IQ measures, and their relationship to the design of beneficial pedagogy follows. Based on reviews of empirical evidence, it is concluded that the measurement of intelligence as practical, at present, makes no contribution to the design of instruction that is beneficial to students, as far as academic achievement is concerned. The main effects of the present popular intelligence measures are found to be negative.