Sample selection biases and the historical growth pattern of children

Bodenhorn et al. (2017) have recently sparked considerable controversy by arguing that the fall in adult stature observed in military samples in the United States and Britain during industrialisation was a figment of sample selection bias. While subsequent papers have questioned the extent of the bias (Komlos and A’Hearn 2016; Zimran 2017), there is renewed concern about selection bias in historical anthropometric datasets. This paper extends Bodenhorn et al.’s discussion of selection bias on unobservables to sources of children’s growth, specifically focussing on biases that could distort the age pattern of growth. Understanding how the growth pattern of children has changed is important since these changes underpinned the secular increase in adult stature and are related to child stunting observed in developing countries today. However, there is potential for selection on unobservables in historical datasets containing children’s and adolescents’ height, so scholars must be aware of these biases before analysing these sources. This paper highlights, among others, three common sources of bias: 1) positive selection of children into secondary school in the late nineteenth and early twentieth centuries; 2) distorted height by age profiles created by age thresholds for enlistment in the military; and 3) changing institutional ecology which determines to which institutions children are sent. Accounting for these biases weakens the evidence of a strong pubertal growth spurt in the nineteenth century and raises doubts on some long run analyses of changes in children’s growth, especially for Japan.