The Evidence on Teaching

Genetic Regulation of Physical Fitness in Children: A Twin Study of 15 Tests from Eurofit and Fitnessgram Test Batteries

Silventoinen K, Maia J, Sillanpää E, Sund R, Gouveia ÉR, Antunes A, Marques G, Thomis M, Kaprio J, Freitas D · 2024

grade Ctwin-adoptionindependentunreplicated
Sample
198 Portuguese twin pairs (40% monozygotic), ages 6-18
Population
Portuguese child/adolescent twins, ages 6-18; 15 field tests of fitness and motor coordination
Design
Single Portuguese cohort, modest sample (198 pairs), wide age band pooled 6-18. Classical ACE/AE twin modeling of directly measured field tests rather than self-report.
Key findings
Across 15 Eurofit/Fitnessgram tests, genes explained 52% (standing long jump) to 79% (sit and reach) of variance; 12-min run 75%, handgrip 71%, 10x5 m shuttle 65%, flamingo balance 60%. AE model fit best: no detectable shared-environment effect, so family, school, and coaching do not explain youth fitness or coordination differences.
Genetic confound
Classical twin design, equal-environments assumption. Same-age co-twins typically share school and coaching, which makes the absent C estimate notable.

Effects

OutcomeMetricValueMeasureTimingVsHorizonClass
Physical fitness and motor performance across 15 testsh2 range52% (standing long jump) to 79% (sit and reach)standardizedages 6-18, single assessmentnonenot-applicablehealth
Cardiorespiratory endurance (12-min run/walk)h275%standardizedages 6-18nonenot-applicablehealth
Handgrip strength; 10x5 m shuttle run (agility); flamingo balanceh271%; 65%; 60%standardizedages 6-18nonenot-applicablehealth
Shared environmental variance (C) in youth fitnessC variancenot significant; AE model without sex-specific genetic factors fit beststandardizedages 6-18nonenot-applicablehealth

Cited by