| 1 |
Identifying preschool measures most predictive of language outcomes at 11 years in the Early Language in Victoria Study ...
|
|
|
|
Abstract:
Language provides a foundation for quality of life and development, but language difficulties, affecting 10% of children, are generally only identified late in childhood. This study aims to identify a parsimonious set of preschool indicators that predict language outcomes, using data from the Early Language in Victoria Study (n=851). Between 8 months and 3 years, parents completed surveys relating to their child and at 11 years the children completed a standardised language assessment. This study uses random forests and SuperLearner machine learning methods to identify which parent-reported questions best predict children’s language skills at 11 years. We hypothesise that gestures, vocabulary, morphosyntax, family history of Language Disorder, socioeconomic status and parent-child interactions predict language outcomes. A short, accurate set of predictors of language outcomes would allow targeted recruitment of participants at risk of persisting language difficulties into intervention studies and could be ...
|
|
Keyword:
Communication Sciences and Disorders; First and Second Language Acquisition; FOS Languages and literature; language development; language disorders; Linguistics; longitudinal studies; machine learning; Medicine and Health Sciences; predictive value of tests; random forests; Social and Behavioral Sciences; SuperLearner
|
|
URL: https://osf.io/nqht7/
https://dx.doi.org/10.17605/osf.io/nqht7
|
|
BASE
|
|
Hide details
|
|
|
|
|
