Sex biases in bird populations, particularly an excess of males, can threaten small, isolated populations by reducing reproductive rates. A skew in sex-ratio can be caused by (i) differential mortality at conception or hatching, (ii) differential survival to adulthood, or (iii) stochastic population processes. However, an observed sex-ratio bias can also result from a bias in census methods rather than a realised bias. Distinguishing between the two possibilities is critical to identifying the underlying causes. An excess of males has long been observed in the Kangaroo Island population of the endangered South Australian Glossy Black-Cockatoo (Calyptorhynchus lathami halmaturinus) (Psittaciformes: Cacatuidae), and our census data confirm this. Here, we use DNA-based sexing to test the accuracy of morphologically based nestling sex assignments in C. l. halmaturinus. We find that in all but one case, the DNA-based sexing was concordant with the sex assigned as nestlings, thus confirming the validity of nestling plumage-based sexing in a larger sample from this population. Our analysis of nestling sex data suggests that the adult sex-ratio is a function of a skew at fledging and not caused by higher rates of female mortality post fledging.