The Evidence on Teaching

Fundamental movement skill interventions in youth: a systematic review and meta-analysis

Morgan PJ, Barnett LM, Cliff DP, Okely AD, Scott HA, Cohen KE, Lubans DR · 2013

grade Cmeta-analysisindependentreplicated
Population
School-age youth; 19 interventions, mostly primary school (field)
Design
22 articles / 19 interventions: only 6 RCTs, 13 quasi-experimental, 3 pre-post. Almost all primary-school settings, delivered by PE specialists or highly trained classroom teachers. Outcomes are process-based FMS batteries (TGMD etc.) at immediate post-test; many studies scored poorly on risk-of-bias items.
Key findings
Teaching fundamental movement skills demonstrably works ON THE TAUGHT SKILLS — large immediate effects, largest for locomotor, smaller for object control (the more coordination-loaded, more heritable class). But this is acquisition-phase evidence on the trained tasks, mostly from non-randomized designs with poor bias scores. Whether gains survive after the program ends, and whether they cascade to physical activity, is exactly what Barnett 2022 rates indeterminate. Skill moves; capacity and downstream behaviour do not follow automatically.
Genetic confound
None controlled; comparison-group designs but no genetically informative arm and no test of aptitude x treatment interaction (who responds to FMS instruction).
Replication notes
Replicated in direction by later FMS meta-analyses (childcare/kindergarten 2017; school-based 2025), all flagging low-quality evidence and missing follow-up.

Effects

OutcomeMetricValueMeasureTimingVsHorizonClass
Overall gross motor proficiencySMD1.42 (95% CI 0.68-2.16), p<.0002standardizedacquisition/immediate post-intervention; field (school)unclearend-of-treatmentdomain-skill
Locomotor skill competencySMD1.42 (95% CI 0.56-2.27), p=.001standardizedacquisition/immediate post; fieldunclearend-of-treatmentdomain-skill
Object control skill competencySMD0.63 (95% CI 0.28-0.98), p=.0004standardizedacquisition/immediate post; fieldunclearend-of-treatmentdomain-skill
Retention at delayed follow-upFollow-up effect sizeNot established — evidence base is dominated by immediate post-test; later FMS meta-analyses (2017-2025) repeat that long-term follow-up is largely absentstandardizedretention; unmeasuredunclearuncleardomain-skill

Cited by