Dental caries is caused by a combination of infection and diet. This disease, if left untreated, may lead to pain, and impair the quality of life, nutritional status and development of young children. The objective was to investigate the association between snacking and caries in a population at high risk of dental caries. American preschool children (n = 1,206) were recruited in the offices of paediatricians. Data on sociodemographic characteristics, oral hygiene, breast-feeding, use of bottle and snacking were collected by questionnaire. Plaque presence, the number of teeth and their caries status (deft) were scored. The children sampled were 61% Black, 27% White and 10% Asian. Of the 1- to 2-, 2- to 3- and 3- to 4-year-old children, 93.8, 82.4 and 77.3% were caries free, and their mean caries scores were 0.16, 0.58 and 0.93, respectively. Multivariate partial least squares (PLS) modelling revealed plaque presence, lowest income, descriptors for tooth exposure time (number of teeth and age) and cariogenic challenge (total intake of sugar-containing snacks and chips/crisps, and chips intake with a sugar-containing drink) to be associated with more caries. These differences were also found in univariate analyses; in addition, children who continued breast-feeding after falling asleep had significantly higher deft values than those who did not. PLS modelling revealed that eating chips clustered with eating many sweet snacks, candies, popcorn and ice cream. We conclude that, in addition to the traditional risk indicators for caries – presence of plaque, sugar intake and socioeconomic status -, consumption of chips was associated with caries in young children.