We aimed to identify the factors influencing child height-for-age z-scores (HAZ) as a measure of child nutritional status in Rwanda, and to examine the role of child feeding and health practices. We conducted a cross-sectional study involving 379 children (aged 6–23 months) and their mothers in northwest Rwanda. Data were collected using a pre-tested, structured questionnaire. An infant and young child feeding practices index (ICFI) and health practices index (HPI) were developed and categorised into tertiles, and linear regression analyses were performed to assess their association with child HAZ. Overall, mothers of non-stunted children exhibited better feeding and health practices than those of stunted children. ICFI was positively associated with child HAZ. We found an adjusted mean HAZ difference of 0·14 between children whose mothers were in high ICFI tertile compared with those in low tertile. Neither HPI nor any of its components were significantly associated with child HAZ. Other factors that were positively associated with child HAZ were infant birth weight (P < 0·001) and maternal height (P < 0·001). Child age, sex (male) (P < 0·05) and altitude (P < 0·05) were negatively associated with child HAZ. Diarrhoea (P < 0·05) and respiratory infections (P < 0·05) were negatively associated with HAZ in younger children aged 6–11 months. Policies to reduce stunting in this population must focus on both pre- and postnatal factors. Appropriate child feeding practices, particularly breast-feeding promotion and improvement in children’s dietary diversity combined with measures to control infections should be given priority.