The purpose of this study was to assess the onset of independent ambulation relative to possible relationships with maternal and infant characteristics. In a cross-sectional study, the health files of 1,553 Turkish children aged 12-23 months were selected by the multistage sampling method in the Nomenclature of Territorial Units for Statistics (NUTS) regions coded as low, medium and high malnutrition levels in Turkey. Children were selected from health centers by systematic sampling technique in each region. Kaplan-Meier analysis and estimated mean values were used for data description; log-rank test and the Cox multivariable regression analysis were applied for data analysis. Maternal education level, occupation, region of residence, gestational iron supplementation, child's gender, child's nutritional status, and presence of anemia in the infant during the survey period demonstrated significant relationships with walking unassisted in the univariate analysis. However, multivariable analysis showed that high maternal education, absence of parental consanguinity and appropriate weight-for-age Z score were positively associated with earlier age of walking. These findings showed the importance of improvement in girls' education, prevention of postnatal growth retardation and improvement in diet quality for children's gross motor development. In addition, counseling programs should be given to decrease the rate of parental consanguinity.