From Baby Babble to Childhood Chatter: Predicting Infant and Toddler Communication Outcomes Using Longitudinal Modeling

Abstract

The importance of early communication development for later literacy, school achievement, and social interaction cannot be understated, according to a large body of literature. There is a consensus in the field of childhood communication that intervention could be applied earlier to prevent developmental delays from becoming disabilities. The Early Communication Indicator (ECI) measures Gestures, Vocalizations, Single Word Utterances and Multiple Word Utterances in infants and toddlers. Important information to be gained from the ECI is how key skills may predict themselves of other skills at later measurements on the ECI, which would give information to when would be the best window of intervention for children. Kansas Early Head Start programs administered the ECI quarterly to 4445 non-disabled children as part of an accountability program. Multiple imputation procedures were done on the data to recover key information. Longitudinal structural equation modeling lends itself well to this type of developmental data. Univariate panel models were applied to each of the key skills, a multivariate panel model was applied to all of the key skills integrated into a complete model, and a growth curve was used to model the growth in Total Communication (composite ECI score). Key skills predicted significantly to themselves at subsequent time points in the univariate panel model, and to themselves and others in the multivariate panel model. Total communication growth between the ages of 6 and 15 months was a significant predictor of status at 42 months. Limitations and future directions are discussed.