Abstract
Both this and other recent studies indicate that parents' observations of their children's behaviours are biased. Insights gained here suggest prevention of such biases to be impracticable. These findings have implications for temperament and other investigation having parental observations as its data base. However, the present debate as to whether difficult temperament is a child characteristic or parental social perception is unproductive. Parental observations have advantages and disadvantages which need to be considered in relation to those of alternate methods and to investigate aims. A multi-method Systems approach to measurement, which integrates research and clinical perspectives, is proposed.