While numerous studies have recently shown that variation in input quantity predicts children’s rate of acquisition across a range of language skills, comparatively little is known about the impact of variation in input quality on (bilingual) children’s language development. This study investigated the relation between specific quality-oriented properties of bilingual children’s input and measures of children’s language development across a number of skills while at the same time taking family constellation into account. Participants were bilingual preschoolers (n = 50) acquiring Dutch alongside another language. Preschoolers’ receptive and productive vocabulary and morphosyntax in Dutch were assessed. Parental questionnaires were used to derive estimates of input quality. Family constellation was first operationalized as presence of a native-speaker parent and subsequently in terms of patterns of parental language use. Results showed that proportion of native input and having a native-speaker parent were never significant predictors of children’s language skills, whereas the degree of non-nativeness in the input, family constellation in terms of parental language use, and language richness were. This study shows that what matters is not how much exposure bilingual children have to native rather than non-native speakers, but how proficient any non-native speakers are.

Many children learn language, in part, from the speech of non-native speakers who vary in their language proficiency. To investigate the influence of speaker proficiency on the quality of child-directed speech, 29 mothers who were native English speakers and 31 mothers who were native speakers of Spanish and who reported speaking English to their children on a regular basis were recorded interacting with their two-year-old children in English. Of the non-native speakers, 21 described their English proficiency as ‘good’, and eight described their English proficiency as ‘limited’. ANCOVAs, controlling for differences in maternal education and child language level, revealed significant effects of group on lexical and grammatical properties of child-directed speech that the literature has identified as positive predictors of child language development. These results suggest that the child-directed speech of native speakers and non-native speakers with good proficiency provide a richer database for language acquisition than the child-directed speech of speakers with limited proficiency.