Cluster-randomized controlled trial evaluating the effectiveness of computer-assisted intervention delivered by educators for children with speech sound disorders

Purpose: The aim was to evaluate the effectiveness of computer-assisted input-based intervention for children with speech sound disorders (SSD). Method: The Sound Start Study was a cluster-randomized controlled trial. Seventy-nine early childhood centers were invited to participate, 45 were recruited, and 1,205 parents and educators of 4- and 5-year-old children returned questionnaires. Children whose parents and educators had concerns about speech were assessed (n =275); 132 children who were identified with phonological patternbased errors underwent additional assessment. Children with SSD and no difficulties with receptive language or hearing, typical nonverbal intelligence, and English as their primary language were eligible; 123 were randomized into two groups (intervention n = 65; control n = 58), and 3 withdrew. The intervention group involved Phoneme Factory Sound Sorter software (Wren & Roulstone, 2013) administered by educators over 9 weeks; the control group involved typical classroom practices. Participants were reassessed twice by a speech-language pathologist who was unaware of the initial assessment and intervention conditions. Results: For the primary outcome variable (percentage of consonants correct), the significant mean change from pre- to postintervention for the intervention group (mean change = +6.15, p <.001) was comparable in magnitude to the significant change for the control group (mean change = +5.43, p <.001) with a small between-groups effect size for change (Cohen’s d = 0.08). Similar results occurred for measures of emergent literacy, phonological processing, participation, and well-being. Conclusion: Computer-assisted input-based intervention administered by educators did not result in greater improvement than typical classroom practices.