The importance of well-specified research questions in the evaluation of early predictors of later inattention and hyperactivity is examined. In an analysis of a nationally representative sample of 2,717 children aged 4 to 10, latent growth trajectories for television viewing and inattention and hyperactivity are determined and the relationship of the two constructs examined. Analyses reveal a logistic latent growth model as the best description of the trajectory of television viewing across time, whereas a quadratic trend represents the best portrayal of the trajectory for symptoms of inattention and hyperactivity. Results do not support the presence of a meaningful relationship between television viewing and inattention and hyperactivity, which is inconsistent with previous findings from the same data set. The importance of the nature of well-specified research questions and the need to use contemporary longitudinal evaluation techniques to avoid misleading conclusions based on limited analyses and results are discussed. (Contains 2 tables and 1 figure.)