Abstract: The current article analyzes 1994-2000 data from the Canadian National Longitudinal Survey of Children and Youth to examine the relevance of family structures to trajectories of parental reports on hyperactivity-inattention among elementary school aged children. We use a latent growth modelling approach to compare children living in intact families, lone-parent families, stepfamilies, and families where parents divorced or separated. The results highlight the apparent advantages to living in intact families and the slightly greater risks experienced by children living in stepfamilies. Children in lone-parent families, while experiencing an initial disadvantage, displayed a similar trajectory on hyperactivity to children in intact families over the 1994-2000 period. With regard to the children of divorce, the current study finds little evidence of a pre-disruption effect, as the children whose parents divorce or separate over 1994-2000 appear initially no worse off than children whose parents stay together. Resume: Cet article analyse les donnees 1994-2000 de l'enquete longitudinale canadienne nationale sur les enfants et les jeunes dans le but d'examiner la pertinence des structures familiales sur les trajectoires de rapports parentaux sur l'hyperactivite et l'inattention chez les enfants du cours primaire. Nous utilisons une approche de modele de croissance non detecte pour comparer les enfants qui vivent dans des families intactes, monoparentales, reconstituees et des familles ou les parents sont divorces our separes. Les resultats ont mis en evidence les avantages apparents de vivre darts une famille intacte et le risque legerement superieur auquel sont exposes les enfants qui vivent darts des familles reconstituees. Les enfants de families monoparentales, bien qu'ils aient un desavantage initial, manifestaient une trajectoire semblable en matiere d'hyeractivite que les enfants des families intactes de la periode 1994-2000. En ce qui concerne les enfants de parents divorces, l'enquete actuelle a trouve tres peu d'indication d'un effet prealable a la rupture etant donne que les enfants dont les parents ont divorce ou se sont separes pendant la periode 1994-2000 ne semblaient pas avoir plus de problemes que les enfants dora les parents demeuraient ensemble. ********** Ample evidence exists to suggest that family contexts can have profound effects on child development, including the likelihood of delinquency and various antisocial behaviours (Harper and McLanahan, 2004; Henry, Tolan, and Gorman-Smith, 2001 ; Wiesner and Capaldi, 2003; Wiesner and Windle, 2004). An issue that has spawned considerable interest in recent years involves the increasing incidence of child hyperactivity in general or attention deficit, hyperactivity disorder (ADHD) in particular. The media have fuelled the public's perception that children and youth have become more difficult over the years, especially as manifested by the perceived increase in children's behavioural difficulties. Indeed, the evidence confirms a rather substantial climb in the percentage of children diagnosed with hyperactivity-inattention disorders in North America--a major concern to parents, educators, and the general public. Nevertheless, despite the prevalence of these disorders, not much research exists on the extent to which family structure and other contextual factors may contribute to the onset and trajectory of such behavioural problems (Schmitz, 2003). The current paper hopes, in part, to remedy this situation. The results from the first cycle of the National Longitudinal Study of Children and Youth (NLSCY) in 1994 indicated that a significant number of elementary aged children in Canada could be classified as exhibiting hyperactive or inattentive behaviour, which was by far the most commonly observed behavioural problem that parents reported. For example, Offord and Lipman (1996) used the NLSCY to estimate that one in ten Canadian children displayed this particular behavioural problem. …