Predictors of Variation in Maternal Talk to Children: A Longitudinal Study of Low-Income Families

Abstract

Objective. The goals of this study were (1) to examine the extent of variation in amount and lexical diversity of maternal talk to young children within a sample of low-income families, (2) to determine the patterns of change over time in maternal talk during the early childhood years, and (3) to consider specific predictors of variation in maternal communicative input. Design. Low-income mother - child dyads (n = 108) were videotaped at child ages 14, 24, and 36 months during semistructured play in the home. Videotapes were transcribed and analyzed using the conventions of the Child Language Data Exchange System to determine amount (word tokens) and lexical diversity (word types) of maternal talk to children. Background information collected from mothers at study entry served as predictors of variation in communicative input. Individual growth modeling methods were used to analyze data. Results. Mothers varied greatly in the number of tokens and types produced during interaction. Mothers were found to increase in their total amount of talk and diversity of vocabulary as children aged. In addition to child age, maternal education, language and literacy skills, depression, and age helped explain variation in level of maternal talk, but none of the predictors explained variation in growth. Different combinations of predictors explained variation in tokens versus types. Conclusions. Findings highlight the importance of studying specific predictors of parental talk to children, as even in low-income samples large variation in communicative input is evident.