This study compared the growth of homeless children with National Center for Health Statistics (NCHS) standards and with growth of age-matched domiciled children of similar income level. Homeless children (n = 167) had lower height percentiles when compared with domiciled children (n = 167; P less than .001) and when compared with NCHS standards (P less than .001). The weight-height percentiles of homeless children did not differ from NCHS standards; however, domiciled children had higher weight-heights when compared with the homeless (P less than .001) and with NCHS standards (P less than .001). After controlling via regression analysis for the effects of potentially confounding factors that affect growth, it was found that homeless children from larger families and with single mothers accounted for the lower height percentiles observed. After controlling for confounding factors, domiciled children still had increased weight-height percentiles when compared with the homeless group. Duration of homelessness was not associated with decreased height or weight-height among homeless children. Homeless children in this study exhibited a pattern of stunting without wasting which is characteristic of poor children experiencing moderate, chronic nutritional stress. They exhibited a greater degree of nutritional stress than domiciled children at a similar income level and than that reported in other groups of poor children in the United States. Preexisting social factors in the families of homeless children were important in explaining the observed growth abnormalities. Further exploration of the associations between social characteristics of homeless children and their families and the growth of these children is warranted.