Genetic and environmental influences on writing and their relations to language and reading

540 twins (identical and fraternal), aged 8-18

US twin sample (Colorado Learning Disabilities Research Center), ages 8 to 18

Classical twin design (Annals of Dyslexia 63(1), 25-43) testing three writing measures - Woodcock-Johnson III Writing Samples, Woodcock-Johnson III Writing Fluency, and Handwriting Copy from the Group Diagnostic Reading and Aptitude Achievement Tests - alongside three language skills (phonological awareness, rapid naming, vocabulary) and three reading skills (word recognition, spelling, reading comprehension). Standardised instruments throughout, which is unusual for this subject. Multivariate genetic analyses decompose the covariance between writing measures. The equal-environments assumption is the standard caveat.

Substantial genetic influence on writing samples and on handwriting copy, and on all the language and reading measures. SHARED ENVIRONMENT WAS GENERALLY NOT SIGNIFICANT except for vocabulary - the same pattern the archive records for reading and for physical capacity, and it means the household does not explain who writes well. Genetic influences among the writing measures were significantly CORRELATED, most strongly between the two speeded measures (writing fluency and handwriting copy), but there were also significant INDEPENDENT genetic influences distinguishing handwriting copy from writing samples and writing fluency from writing samples - so transcription speed and composition quality are not genetically the same thing, even though they covary. That partial genetic independence is the strongest available reason to treat transcription and composition as separate teaching targets rather than assuming one drives the other.

This IS the genetically informative source for the subject. Its limits are the equal-environments assumption, an age range starting at 8 (so it says nothing directly about the 5-7 band), and the absence of any instructional manipulation - it partitions variance in existing differences and licenses no prediction about any individual child or any response to teaching.