The 2002 NATSISS was designed to provide information on the social circumstances of Indigenous people relevant to Indigenous stakeholders and policy researchers. Like the General Social Survey of the total population, the NATSISS is a multi-dimensional survey covering a wide range of areas of social concern, with the capacity to enable analysis of interactions among different topics, including the analysis of multiple disadvantage. The 2002 NATSISS was designed to measure selected changes over time (since 1994) and to allow for comparisons with the circumstances of the non-Indigenous population. To enable this, the survey questionnaire had about a 50 per cent content overlap with the 1994 NATSIS and considerable overlap with the 2002 GSS (ABS 1995, 2003b). It also included new material, for example, on disability, incarceration and age at first formal charge.
The 2002 NATSISS was designed to provide reliable estimates at the national level and for each of the eight Australian States and Territories. In addition, the Torres Strait Islander population was over-sampled in order to produce data for the Torres Strait area and the remainder of Queensland. The sample was spread across the States and Territories in order to produce estimates that would have a relative standard error (RSE) of no greater than 20 per cent for characteristics that are relatively common in the Indigenous population; for example, characteristics that at least 10 per cent of the population would possess.
The 2002 NATSISS incorporated two broad samples that together comprised about 9400 people in 5900 households and covered all areas of Australia. The first was a random sample of about 2100 people in discrete Indigenous communities, predominantly in remote areas. Discrete communities in Queensland, South Australia, Western Australia and the Northern Territory were included in this sample. The second was a random sample of about 7300 people in so-called ‘non-community’ areas drawn from major cities, regional and remote areas in all States and Territories. The two samples were designed separately, with each involving a multi-stage sampling process. Overall, the 2002 NATSISS sampled one in 30 of the total Indigenous population (see Table 3.1).
Table 3.1. Sample size, 2002 NATSISS
|
Non-remotea |
Remoteb |
Total |
Estimate (15 years or over) |
Sample fraction |
|
|
New South Wales |
1137 |
365 |
1502 |
83 800 |
1 in 56 |
|
Victoria |
806 |
– |
806 |
17 400 |
1 in 22 |
|
Queensland |
1018 |
847 |
1865 |
76 000 |
1 in 41 |
|
South Australia |
605 |
405 |
1010 |
15 800 |
1 in 16 |
|
Western Australia |
465 |
1097 |
1562 |
39 600 |
1 in 25 |
|
Tasmania |
713 |
23 |
736 |
10 900 |
1 in 15 |
|
Northern Territory |
168 |
1380 |
1548 |
36 200 |
1 in 23 |
|
Australian Capital Territory |
330 |
– |
330 |
2600 |
1 in 8 |
|
Australia |
5242 |
4117 |
9359 |
282 200 |
1 in 30 |
a. Comprises major cities, inner regional and outer regional areas.
b. Comprises remote and very remote areas.
Source: The 2002 NATSISS MURF
In 2002, survey content and methodology were specifically designed to take account of the different circumstances of Indigenous people in remote communities and non-remote areas. In the community sample, the standard household survey approaches were modified as a result of pre-testing, to accommodate language and other issues associated with the geographic remoteness of these communities. ABS interviewers were accompanied by local Indigenous facilitators who assisted in the conduct and completion of the interviews. In the community sample, interviewers used a pen and paper interview (PAPI) questionnaire, while in the non-community sample interviews were conducted predominantly using Computer Assisted Interviewing (CAI).
While PAPI was only used in the four jurisdictions (Queensland, South Australia, Western Australia and the Northern Territory) that had households in the community sample, about half of the national sample of 4100 people surveyed in remote/very remote areas were interviewed using CAI and half using the PAPI form. Of the 1600 people surveyed in the remote component, just over two-thirds were interviewed using CAI, whereas of the 2500 people in the very remote component, two-thirds were interviewed using PAPI (see Fig. 3.1). Use of PAPI varied from 30 per cent of the total remote samples in South Australia and Western Australia to nearly 50 per cent in Queensland and almost 100 per cent in the Northern Territory.
While wording was modified in the PAPI questionnaire, most underlying concepts remained consistent with those in the CAI questionnaire. Data items or output categories that were specific to either PAPI or CAI, such as some output associated with disability status, are identified in the main ABS publication and in the documentation for the CURF.
