The Evidence on Teaching

The effect of grammar teaching on writing development

Andrews R, Torgerson C, Beverton S, Freeman A, Locke T, Low G, Robinson A, Zhu D · 2006

grade Creviewindependentnot-applicable
Sample
Reports two EPPI-Centre systematic reviews; the sentence-combining in-depth review synthesised 18 studies, the formal-grammar (syntax) review a comparably small set
Population
English-medium schooling, ages 5-16
Design
Journal summary (British Educational Research Journal 32(1), 39-55) of two systematic reviews commissioned by the DfES and conducted to EPPI-Centre protocol - one on formal grammar (syntax) teaching, one on sentence combining. Narrative synthesis; no pooled effect size is reported in either. The authors are explicit that the constraint is the primary literature, not the synthesis method: the field's studies are mostly old, mostly American, and mostly under-reported.
Key findings
Verbatim from the abstract: "The results show that there is little evidence to indicate that the teaching of formal grammar is effective; and that teaching sentence-combining has a more positive effect. In both cases, however, despite over a hundred years of research and debate on the topic, there is insufficient quality of research to prove the case with either approach." The authors call for a review of policy and practice in England on the teaching of sentence structure - a call the 2014 National Curriculum then moved in the opposite direction from.
Genetic confound
Not directly addressed. The underlying corpus is experimental and quasi-experimental rather than correlational, so the direction of effect is not a heredity artefact; but the review makes no genetically informative claims and none should be read into it.

Effects

OutcomeMetricValueMeasureTimingVsHorizonClass
Writing quality/accuracy, formal grammar (syntax) teachingnarrative synthesis verdictlittle evidence of effectiveness; no pooled ES computedresearcher-designedpost-treatmentunclearend-of-treatmentdomain-skill
Writing quality/accuracy, sentence combiningnarrative synthesis verdictmore positive effect than formal grammar; evidence quality judged insufficient to prove the caseresearcher-designedpost-treatment and (rarely) delayedunclearuncleardomain-skill

Cited by