The 2003 SDAC response rate for private dwellings was 89.8 per cent, and for establishments was 91 per cent. SDAC data are weighted to account for non-response. The SDAC is conducted in all States and Territories including people in both private and non-private dwellings (including cared-accommodation establishments) but excluding those in gaols and correctional institutions and very remote areas. The exclusion of persons usually resident in very remote areas has a small impact on estimates, except for the Northern Territory, where such persons make up a relatively large proportion of the population. As a consequence of this exclusion, comparisons between Indigenous and non-Indigenous people in remote areas would not be available even if the same disability module was used in each survey. For more information on SDAC scope and coverage, see the Disability, Ageing and Carers User Guide on the ABS website. The NATSISS is conducted in all states and territories and includes remote and non-remote areas. The 2008 sample was 13 300 persons/6900 households, with a response rate of 82 per cent of households. There was a relatively large level of undercoverage in the 2008 NATSISS compared with other ABS surveys, therefore the analysis undertaken to ensure that results from the survey were consistent with other data sources was more extensive than usual. Potential bias due to undercoverage was addressed by adjustments to the initial weights and an adjustment to geographical areas based on the density of the Indigenous population. As undercoverage can result in variances across population characteristics, as well as across data items, caution should be exercised when interpreting the survey results. For more information, see the 2008 NATSISS Quality Declaration. Being drawn from sample surveys, data for this indicator is subject to sampling error. Sampling error occurs because a proportion of the population is used to produce estimates that represent the whole population. Sampling error can be reliably estimated as it is calculated based on the statistical methods used to design surveys. This indicator has acceptable levels of sampling error (relative standard errors less than 25 per cent) for most tables. RSEs increase for tables with a greater level of disaggregation, therefore the tables that are split by labour force status and state; age, sex and state; and disability status and state (tables NDA.2.14, NDA.2.24, NDA.2.26 and NDA.2.36) show some RSEs between 25 per cent and 50 per cent for Tasmania, ACT and NT. Data with these RSEs should be used with caution. |