From September 1974 to 28th February 1977 80% of the children in New Haven, Connecticut, aged from 1 month to 72 months were screened for blood lead concentrations. This report examines the relationship of several socioeconomic factors to blood lead concentrations. In addition, a set of hypotheses regarding the effect of environmental and social factors on blood level concentrations in racially defined groups was tested. Characteristics associated with increased blood lead concentrations were found to be those that tend to impair the ability of a family to provide the necessary care and supervision for the young child. The risk factors, however, produce different effects on the various race groups. The analyses support the belief that the elimination of childhood lead poisoning as a public health problem will require recognition of social-demographic and family operational factors that underlie the interactions of childhood behaviour and environmental lead potentially available to children.