This paper investigates, in a unified framework, the effects of sibship size, birth order and sibling sex composition on children's school enrolment in urban Turkey. We utilize an instrumental variable estimation method to address parents' joint fertility and schooling decisions using twin births as instruments. We find no causal impact of sibship size on school enrolment. However, there is evidence for a parabolic impact of birth order where middle-born children fare worse. Sex composition of siblings matters only for female children. Our finding that birth order and sibling sex composition matter more for poorer households suggests that scarce financial resources play an important role in bringing about the sibling composition effects.