Parental investment and age at weaning in a Caribbean village

Predicting the timing of weaning in diverse environments is important because breast-feeding significantly contributes to child survival and overall health, especially in developing countries. This study examines associations between weaning and household demographic variables that test predictions derived from parental investment (PI) theory. Data were collected for 101 children (49 males and 52 females) in a rural community in Dominica. Analyses indicate that father absence is associated with early weaning. This was the only prediction from PI theory that was supported. The following results were contrary to expectations: (1) Availability of female alloparents and household wealth were negatively associated with age at weaning. (2) Number of dependent children in the household was positively associated with age at weaning. (3) Mother’s age at birth was not correlated with the timing of weaning. Lastly, (4) interaction terms for child’s sex by wealth and sex by maternal social support were not associated with age at weaning, indicating the lack of a Trivers–Willard effect on weaning in this population. We suggest that explanations of weaning from PI theory will benefit from including high opportunity costs of prolonged nursing, demands for reciprocal female labor, and the importance of investment in ‘‘embodied capital.’’ D 2002 Elsevier Science Inc. All rights reserved.