Estimating a child’s future height

Traditionally it has been accepted that one could make a fairly accurate estimate of a child’s potential height from the parents’ stature but for some that is not good enough.

Sir Francis Galton was a noted scientist and a half-cousin to Charles Darwin. Galton was a Victorian polymath whose wide-ranging studies included geography and meteorology. He was a tropical explorer, the founder of differential psychology and a pioneer of fingerprint identification.

He wrote and campaigned extensively about what he called the improvement of the human stock and so is probably best remembered now for his theories on eugenics.

Galton was the first to apply statistical methods to the study of human differences. He published a formula that averaged the height of both parents and made adjustments for age and sex to estimate a child’s height.

More recently, Yurii Aulchenko at Erasmus University Medical Centre in Rotterdam devised a calculation based on gene variations linked to height. His researchers identified these mutations by scanning the genomes of thousands of Dutch volunteers and calculating how many gene variants linked to tall stature he or she possessed.

Aulchenko’s team then searched for single-letter changes shared by people of similar stature. They analysed 54 of these genetic variants and plotted them against each person’s height, and adjusted for age and sex.

Sadly, in spite of all this effort, Aulchenko’s team found only a minor correlation between a person’s genetic score and his or her actual height. Worse still, when they compared their results with Galton’s method it became clear that the old Victorian formula was about 10 times better at estimating the height of people than their new one.

Aulchenko noted that while some characteristics such as eye colour are easier to predict through genes, more research is needed to link other traits. I suppose in research terms that means some you win, some you lose.