Empirical evidence for quantity-quality trade-off is hardly ubiquitous, especially when quality is measured by child health outcomes. The paper offers a new explanation to this puzzle. It shows that the quantity-quality relationships are subject to occupational variation when quality is given by nutritional status, and occupations differ in their physical labor intensity. It embeds, in a simple household optimization model, a minimum consumption requirement that rises with physical work intensity of occupation. The occupational differences in subsistence consumption requirement generate variation in child nutritional status, and hence, in the shadow price of children. The nature of the quantity-quality relationship, therefore, varies with work intensity of occupations. The model yields an equilibrium relationship between the number and nutritional status of children that is positive for households in strenuous occupations and ambiguous for other households. These results help reconcile some inconsistent findings on quantity-quality trade-off, which may partly be explained by the omission of occupational variation in nutritional status.