The Effectiveness of Parent-Implemented Language Interventions: A Meta-Analysis
Roberts MY, Kaiser AP · 2011
grade Cmeta-analysisindependentmixed
Sample
18 studies
Population
Children aged 18-60 months with primary and secondary language impairments (below this archive's 4-18 range; recorded for the delivery-agent question)
Design
Random-effects meta-analysis across seven language outcomes, analysed separately by comparison group and by whether the measure was parent-reported or directly observed — the parent-report versus observation split is the measure-type discipline this archive requires and it is rarely done in this literature.
Key findings
Parent-implemented language interventions had significant positive effects on receptive and expressive language for children with and without intellectual disabilities, with effect sizes (g) ranging from -0.15 to 0.82 depending on the outcome and the comparison group. The range is the point: the same intervention looks strong or null depending on what it is compared against and who reports the outcome.
Genetic confound
Low — controlled comparisons within a trained-parent design.
Replication notes
Superseded in part by Heidlage et al. 2020, which restricted to RCTs and found the receptive-language effect no longer significant.
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| Child language outcomes, parent-implemented intervention | Hedges' g range across 7 outcomes and comparison groups | -0.15 to 0.82 | mixed | end of intervention | unclear | end-of-treatment | domain-skill |
Cited by
- The parent as instructor — what survives when the person teaching is the person who raised the childmixedconf: mediumgc: medium