Modelling test performance on some components of executive function in early childhood

In this talk we assess change in performance of young children on repeated measures of some executive functions, based on a study in which 115 children were administered a battery of tests at three time intervals, six months apart. Through our statistical approach i) we address developmental questions by modelling children's performance on single trials ii) we emphasize the importance of incorporating design aspects such as test ordering into the analysis and show that failure to do so may cast doubt on the validity of research reported in the literature in which large numbers of tests are administered to children within a battery. Finally, we present an approach to test hypotheses of separability between two of the psychological constructs under scrutiny.