Frequency of truancy at high school: evidence for genetic and twin specific shared environmental influences

4,835 twins and their non-twin siblings

Dutch adolescent twins and siblings, high school age

Threshold model decomposing variance in self-reported truancy frequency into additive genetic, shared environmental and non-shared environmental components. The sibling extension lets the authors test whether the apparent shared environment is genuinely familial or is an artefact of twins sharing a classroom, cohort and peer group.

Truancy heritability h2 = 45%. The remaining variance splits differently for twins and singletons: for twins, 25% shared environment and 30% non-shared; for singletons, ALL environmental variance is non-shared (55%). In other words the "shared family environment" component of school absence largely disappears once you compare ordinary siblings rather than twins - it is twin-specific (same class, same friends, same year) rather than household-level. Prevalence higher for boys (38% vs 29%) and rising with age.

This is the source that establishes the confound rather than suffering from it. School absence is roughly as heritable as many personality traits, and the household component that attendance policy implicitly targets is close to zero once twin-specific effects are removed. Any observational absence-achievement gradient is therefore contaminated by a trait that is 45% genetic and shares causes with achievement. Standard classical-twin caveats apply (equal environments assumption, self-report). Measures truancy specifically, not illness-driven or parent-permitted absence.