The sample consisted in all substantiated cases of physically abused black and white children who were 11 years old in Ramsey County (Minn.) in 1992. Severity of abuse was measured from descriptions of the child's first incident of substantiated abuse. Cases were coded as currently poor if they were receiving Aid to Families with Dependent Children or General Assistance. The persistence of poverty was calculated as the percentage of the child's life before the abuse that involved public assistance. Results revealed that these relationships are more complicated than previous research has indicated. Persistent poverty had no significant effect on the severity of the abuse, and the observed relationship between current poverty and severity was eliminated once race was controlled. Moreover, although poverty persistence was significantly related to recurrent abuse, almost all of this effect appeared to be explained by factors associated with the stress of economic deprivation. Footnotes, tables, appended table, and 74 references (Author abstract modified)