This cross-national study aimed at understanding variability in growth
among nations and its determinants. It examined the variability of gr
owth among nations, the factors most important for explaining this var
iability, how these factors jointly determine growth, and the consiste
ncy of the patterns and relationships. A conceptual model was adapted
from U. Jonsson. National data on height and weight of children, sex,
ethnicity, and 15 social, economic, health, education, and political f
actors were combined from three secondary sources. A sequence of multi
ple linear regression models was used with three age groups: 1-2, 3-5,
and 6-10 years. Substantial variability in growth among nations was s
een in comparison to that within nations. Regression models with sex,
ethnicity, food security, maternal and child care, and health services
and environment explained a large percentage of this variability. Ins
titutions, politics/ideology, economic structure, and potential resour
ces contributed little additionally to the models. Ethnic differences
among nations were observed and persisted even when models controlled
for other factors, but should not necessarily be interpreted as geneti
c differences. Determinants for height, weight, and weight adjusted fo
r height were somewhat different. Adjustment for only previous growth
and sex explained about 90% of variability in growth, supporting the a
ssumption that growth patterns are generally established early in life
. As policy decisions and programmes addressing mild-to-moderate malnu
trition are increasingly considered, the implications of cross-nationa
l variability in growth may assume greater importance. Future work sho
uld examine simultaneously the effects of national, community, family,
and individual characteristics on growth using data from individuals
in a number of nations.