Behaviour genetics: an unsuitable model for fertility theory

Atam Vetta, Oxford Brookes University
Daniel Courgeau, Institut National d'Études Démographiques (INED)

Some fertility researchers recommend the use of behaviour genetics (BG) for the study of fertility. BG is based on the false assumption that genes segregate independently. There is assortative mating for fertility. Behaviour geneticists invariably use wrong formulae when they fit assortative mating models. Fertility genes must be ‘fixed’ by now. If so, the variation in fertility is ‘environmental’. Kohler & Rogers (2003) use the ACE model. They did not have sufficient data to estimate parameters of this model. Their use of Cholsky’s decomposition does not help. We obtain parameter values by eliminating E (environment) from the ACE model. Heritability estimates are not robust and depend on correlations used in the model. There are no genes for the trait ‘years of tertiary education’ and its heritability has no value. BG is of no use to demographers. They need to look to molecular and genomic researches.