The Impact of Vocabulary Instruction on Passage-Level Comprehension of School-Age Children: A Meta-Analysis
Elleman, A. M., Lindo, E. J., Morphy, P., & Compton, D. L. · 2009
grade Cmeta-analysisindependentreplicated
Sample
37 studies, PreK-12
Population
PreK-12; general readers + subgroup with reading difficulties; US-dominant.
Design
Separated custom (experimenter-designed) from standardized comprehension outcomes - the decisive contrast.
Key findings
Vocabulary instruction reliably teaches the taught words (custom d=0.50) but transfers weakly to independent standardized comprehension (d=0.10) - a ~5x gap. Effect ~3x larger for reading-disabled readers. Vocabulary is a teachable DOMAIN skill, not a general far-transfer comprehension lever.
Genetic confound
Low for the experimental ES; the tiny standardized transfer (0.10) is consistent with heritable/g-linked general comprehension resisting instruction.
Replication notes
Custom>>standardized pattern echoed by Wright & Cervetti 2017.
DOI / URL
10.1080/19345740802539200
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| Passage comprehension (custom measures) | d | 0.50 | researcher-designed | end-of-treatment | unclear | end-of-treatment | domain-skill |
| Passage comprehension (standardized measures) | d | 0.10 | standardized | end-of-treatment | unclear | end-of-treatment | domain-skill |
| Comprehension (custom) - students WITH reading difficulties | d | 1.23 (vs 0.39 without) | researcher-designed | end-of-treatment | unclear | end-of-treatment | domain-skill |
Cited by
- Background knowledge and vocabulary as drivers of reading comprehensionmoderate supportconf: mediumgc: low