BackHome
Refresh: OffView Personal Schedule
Individual Submission Summary
Share...

Direct link:

A Remote Intervention to Enhance Child-Directed Speech: an RCT For Infants and Toddlers

Wed, April 7, 10:15 to 11:15am EDT (10:15 to 11:15am EDT), Virtual

Abstract

Introduction. Previous studies have demonstrated replicable relationships between the quantity and quality of parental speech to children, and children's language development (Hart & Risley, 1995). The majority of studies use a correlational design and argue that parental speech is a key driver of language development, however, other causal pathways are possible (Dale, Tosto, Hayiou-Thomas, & Plomin, 2015). Randomised Controlled Trials (RCTs) are the gold standard for evaluating causal relationships and can be used to test whether changing parental speech influences child language. Many studies use face-to-face delivery of interventions to groups (Ramírez, Lytle & Kuhl, 2020). Remote digital interventions may also be effective, scalable, and safe (i.e. social distancing).
Method. As part of the Talk Together Study, we are delivering a four-week double-blind RCT to parents of 8 to 36-month-olds in Singapore (Fig. 1). Parents randomly allocated to the intervention group receive twice daily text messages containing evidence-based tips and interactive content about ways of enriching child-directed talk. Parents randomly allocated to the control group receive infrequent, general information by email. Assessments of parental speech and child vocabulary take place at baseline (T1) and after the 4-week intervention period (T2). At both time points, parental speech is recorded in a video-call where parents share an on-screen wordless picture book ‘What a scary storm!’ (DOI: TBC) with their child. Parental speech is evaluated in terms of (1) Total words, (2) Mean length of utterance, (3) Type-token ratio, (4) Turn taking and (5) Partial repetitions. Child language is assessed by a developmental vocabulary checklist.
Analysis Plan. If the intervention delivered by text messages is effective, we predict that the quantity and/or quality of child-directed speech will increase more for the intervention group than control (with child age, parental education and SES controlled). To test this, we will run mixed effects multivariate regression with Group (Intervention/Control), Time (T1/T2) and the interaction as fixed effects, with parental speech measures as dependent variables (Fig. 2).
Furthermore, if the intervention is effective, children in the intervention group will show a larger positive change in total vocabulary (T2 − T1) than the control group, and this effect will be largest for those children whose parents show the most change in their speech between T1 and T2 (with child age, parental education and SES controlled). According to our pre-registered analysis plan, we will conduct dimensional reduction using Principle Components Analysis on the change scores of the five parental speech measures (T2 – T1). Factors arising from the PCA will be analysed as predictors of change in child vocabulary size, along with Group (Intervention/Control) in a mixed effects Multiple Regression, with age as a fixed covariate effect (Fig. 2).
Current progress. To date, 181 families have enrolled in the study, 88 have completed the T1 video-call, and 35 have completed their T2 video-calls. Two video-calls have been transcribed. Data collection is expected to end following the pre-registered stop rule in February 2021. Pre-registration: DOI: TBC_after_blind_Review. Archive of Talk Together intervention materials: DOI: TBC_after_blind_Review.

Authors