Cognitive-Components Tests Are Not Much More than g: An Extension of Kyllonen's Analyses

Abstract

A battery of 10 traditional paper-and-pencil aptitude tests and a battery of 25 cognitive-components-based tests were administered to 298 men and women to investigate the common sources of variance in those batteries. Earlier confirmatory factor analyses showed each battery to have a hierarchical structure, each with a single higher order factor. The higher order factor in the paper-and-pencil battery had previously been identified as general cognitive ability, or g. The higher order factor from the cognitive-components battery had been identified as working memory. The intercorrelation of the higher order factors from the two batteries was .994, indicating that both measured g. The proportion of common variance because of g was greater in the cognitive-components battery than in the paper-and-pencil battery. The correlations between each factor based on cognitive components and g averaged .946. Despite theoretical foundations and arguments, cognitive components tests appear to measure much the same thing as traditional paper-and-pencil tests.