Summary

This paper provides a summary of who uses data about and for the Aboriginal and Torres Strait Islander population of Australia and how these data are used:

  • indigenous community and other organisations use data for reflection and advocacy;
  • government and other policy makers use data for allocating resources, identifying needs, monitoring, evaluation and planning;
  • the wider Australian community uses data for accountability (of politicians and policies) and to provide more information about the circumstances of non-Indigenous people;
  • academics and researchers use data for understanding and explaining.

This paper discusses the availability, strengths and weaknesses of 5 types of data: cross-sectional surveys, longitudinal data, qualitative data, administrative data and community-controlled collections.

The extent to which the data sets and types of data meet the uses of the data varies considerably. While the needs of governments and researchers are generally well provided for, the needs of Indigenous community and other organisations who work with and for Indigenous Australians are not as well met due to their reliance on community-controlled collections that are relatively poorly resourced.

Other gaps and issues include:

  • complexities around identification, where the same individual could be recorded as having a different Indigenous status across collections or through time;
  • the lack of a data set with longitudinal information on Indigenous Australians across the life course (that is, from childhood through to adulthood);
  • the use of a range of data sets for a single purpose, meaning that scope and definitions may not be consistent;
  • the reliance on the Australian Census, a data set that is the only source of data for some of the uses discussed above (in particular allocation and planning) and is collected only every 5 years;
  • difficulties in accessing administrative and other data by those outside government, although large tables of appendices from some of those data collections are published;
  • with some exceptions, the shortage of data at the right level of geography—by socially meaningful groupings (for example women and men or across the life course), for the right population and for the right concepts—to be useful for Indigenous community and other organisations;
  • the lack of trial and program data that will enable the evaluation of the effectiveness of programs and policies.

With these limitations in mind, this paper discusses 4 suggestions for improving data about and for Indigenous Australians:

  • considering the inclusion of Indigenous specific information in the Census;
  • expanding the availability of longitudinal data, ideally through the linking of existing cross-sectional data sets;
  • enhancing evaluations (including of service availability and effectiveness) with funding built into the design of programs at the initiation phase and making unit-record data publicly available;
  • increasing support for Indigenous quantitative researchers through targeted scholarships and a national training program for Indigenous data collectors, data managers and analysts.

What we know

  • The Indigenous population have a greater likelihood of being involved in quantitative and qualitative research than most other Australians.
  • We have reasonable information about where Indigenous people live and the distribution of their socioeconomic outcomes, albeit with some error and uncertainty.
  • We know that there is a large gap between Indigenous and non-Indigenous Australians in health and socioeconomic status, and that the gap holds across Australia.
  • A consistent finding from data about Indigenous Australians is that there are different levels and determinants of wellbeing depending on the measure used. For example, on some measures of wellbeing, those in remote areas do relatively poorly, in others they do reasonably well.

What works

  • From both a rights-based and pragmatic perspective, the inclusion of Indigenous people and Indigenous standpoints in the design and analysis of Indigenous-specific data collections and surveys enhances their relevance and acceptance within the Indigenous community.
  • The consistent use of Indigenous identification questions in administrative and other data allows for comparability across data collections.
  • Combining qualitative and quantitative data gives a more nuanced view of Indigenous outcomes and aspirations.

What doesn’t work

  • Government program and policy evaluations that are designed after the policy has been implemented are likely to be more expensive and usually do not give useful information about the effectiveness of the policy.
  • If we assume that a particular policy or program will be effective, then we are unlikely to learn much from its evaluation.
  • Assuming that Indigenous people will respond to surveys or stay in longitudinal surveys at the same rate or for the same reasons as the non-Indigenous population can lead to high rates of non-response and sample attrition.

What we don’t know

  • We rarely consider or discuss how to balance the competing priorities of the different uses and users of data on Indigenous people.
  • Due to the lack of longitudinal information, we know little about the specific causes or consequences of changes in Indigenous outcomes.
  • There is a consistent lack of information and data on the specific policies and programs that will lead to sustained improvement in Indigenous wellbeing or what some of the unintended consequences of particular programs might be.