The 'error of closure' is the population growth that cannot be accounted for either by natural increase or by quantifiable non-demographic factors. The term is somewhat misleading: since it incorporates all unquantifiable components of the increase in a population count, it is unlikely ever to be 'closed'. This study highlights the significance of variability of Indigenous population estimates by calculating standard errors, one of the conventional measures of reliability of statistics. Boyd Hunter and Mardi Dungey introduce the Dual System Estimator method for estimating the Indigenous population, and review the international literature on its strengths and weaknesses. Once Australia's Indigenous population has been estimated using this method, confidence intervals are compared to those produced using the traditional undercount method. While the paper concludes that Dual System Estimates of the Indigenous population are reasonably accurate at the national level, this may need to be revised when regionally disaggregated data are examined.