Factor analysis of variability measures in eight independent samples of children and adults

Variability in strategy use has been measured and conceptualized in several ways. The present study sought to determine whether a sample of variability measures could be reduced to a smaller number of factors that the individual measures had in common. Factor analysis of five commonly used variability measures was performed separately on data from eight independent samples of children and adults. The results revealed that the variability measures could be reduced to one of two general types of variability: strategy diversity, which refers to the sheer number of strategies used, and strategy change, which refers to the number of trial-by-trial changes in strategy use. Support for this two-factor model of variability was found for all samples of children and adults. Moreover, each type of variability had distinctive relations with cognitive performance that varied with age. In general, strategy diversity was positively related to recall for children but not adults, whereas strategy change was negatively related to recall for adults but not children. Based on this evidence, three recommendations for researchers studying strategy variability are described.