We used data from 11 states (5,449 respondents) to examine the association between self-reported consistent use of occupant restraints for children under 11 years of age and factors related to this practice. Multivariable statistical methods that model the effect of several variables simultaneously on child restraint use were used. Adults' own self-reported safety belt use, age of youngest child in the household (child restraint use decreased with increasing age), and educational attainment were significant predictors of child restraint use. Respondent age, race/ethnicity, sex, marital status, household income, and employment status were not significantly associated with child restraint use. We also found that adult and child occupant restraint use was higher in states with an adult safety belt law than in states without such a law.