In a two-stage community study of 3069 Chinese schoolboys in Hong Kong, those entering stage 2 were classified by scores on Rutter's teacher and parent questionnaires into: (1) a mixed hyperactive/conduct-disturbed (HA-CD) group; (2) a pure hyperactive (HA) group; (3) a pure conduct-disturbed (CD) group; and (4) a normal control group. The four groups of children were compared on a series of psychosocial, cognitive and neurodevelopmental measures. This was followed by a regression analysis to examine the specificity of the differential patterns of associations between HA and CD. There was a mix of negative and positive findings defying a simple, definitive conclusion. However, the positive findings that did emerge supported a growing body of recent literature which favoured a separation of HA from CD and their cross-cultural validity: the former was associated with neurodevelopmental impairments, the latter with family disharmony. The mixed condition, HA-CD, was a hybrid of its two constituent conditions, displaying the attributes of both.