On the accuracy of perceived parental height in a native Amazonian society.

Studies of secular trends in adult height in rural pre-literate societies are likely to show no change owing to random measurement error in age. In such societies, adults lack birth certificates and guess when estimating their age. We assess the accuracy of perceived height of the same-sex parent to estimate secular trends. We tested the method among the Tsimane', a native Amazonian society of farmers and foragers in Bolivia. Subjects included 268 women and 287 men >20 years of age. Over half the sample reported inaccurately the height of their same-sex living parent, with a tendency to report no difference when, in fact, differences existed. Results highlight the pitfalls of using perceived parental height to examine secular trends in adult height among the Tsimane', though the method might yield accurate information in other societies. We discuss possible reasons for the low accuracy of Tsimane' estimates.