AustLII Home | Databases | WorldLII | Search | Feedback

Indigenous Law Bulletin

Indigenous Law Bulletin
You are here:  AustLII >> Databases >> Indigenous Law Bulletin >> 2010 >> [2010] IndigLawB 26

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Biddle, Nicholas; Yap, Mandy --- "Closing the (Data) Gap" [2010] IndigLawB 26; (2010) 7(19) Indigenous Law Bulletin 17

Closing the (Data) Gap

Nicholas Biddle and Mandy Yap

In a forthcoming Research Monograph,[1] we use data from the 2006 Census to analyse the timing of key stages and events across the Indigenous lifecourse. The overarching aim of the paper was to consider whether there was something different about the Indigenous lifecourse compared to the non-Indigenous lifecourse (as observed in the 5% Sample File from the 2006 Census). The simple answer to this question was yes. We analysed 25 variables across the dimensions of fertility and family formation, migration and mobility, education participation, employment, housing, health and childhood outcomes. For all of these variables there was a significant difference between Indigenous and non-Indigenous Australians, either before or after controlling for other characteristics.

Importantly, not only were there differences between Indigenous and non-Indigenous Australians in terms of levels, there were also substantial differences in patterns across the lifecourse for a number of the dependent variables. For example, across the population, Indigenous Australians were more likely on average to have changed their place of usual residence over the previous five years compared to non-Indigenous Australians. However, between a person’s early 20s and early 40s, it is the non-Indigenous population who is more geographically mobile. While much of the analysis presented in the paper pointed to differences between the Indigenous and non-Indigenous populations, it also showed that in many cases there was as much variation within those two populations as there was between them. In a number of instances, the factors associated with the particular demographic or socioeconomic variable varied between the two populations. However, the relationship between demographic and socioeconomic variables was also shown to be vitally important.

There were a number of findings in the Monograph that have clear implications for Indigenous policy. Firstly, by bringing together so many demographic and socioeconomic variables, it is clear how big an influence education attainment has on a range of outcomes. For example, it was shown that those Indigenous Australians with reasonably high levels of education attainment had comparable levels of employment to similarly qualified non-Indigenous Australians. It was those with relatively low skills where there was a large employment gap. The relationship between education and employment is not surprising and confirms previous findings. However, we also showed a strong association between education and a number of other variables that have not previously been analysed in such a way. This includes volunteer work, marital status and residential mobility. Another finding from the analysis that has clear policy implications is that much of the difference between Indigenous and non-Indigenous females in terms of education participation and employment was explained by higher fertility rates.

Although we feel that these and other insights from the Monograph have important policy relevance, we were frustrated in our analysis by the reliance on a single cross-section (the 2006 Census). So, while we were able to show the difference between a 30–34 year old and a 50–54 year old in 2006 (for example), we weren’t able to show how the characteristics of an individual Indigenous Australian changed over a 20-year period. This was not an oversight or due to a poor choice of dataset. Rather, there is no dataset currently available that tracks a sufficient number of Indigenous Australians to make sensible comparisons, either within the population or between Indigenous and non-Indigenous Australians.

There are some datasets that have the potential to be used for limited longitudinal lifecourse analysis. The Household Income and Labour Dynamics in Australia (‘HILDA’)[2] survey has a small Indigenous sample. And, although the Longitudinal Study of Australian Children (‘LSAC’)[3] and the Longitudinal Survey of Australian Youth (‘LSAY’)[4] are restricted to particular age cohorts only, they do have a moderate Indigenous sample (at least in the first few waves). The Longitudinal Study of Indigenous Children (‘LSIC’)[5] has the potential to provide some information on the developmental pathways of two cohorts of children aged 6–18 months and 3½–4½ years respectively at the base year. However this study is only in its infancy, with the most relevant longitudinal information still a number of years away. Furthermore, there is no non-Indigenous comparison available in the LSIC.

In terms of large-scale surveys, there are three main sources of data for the Indigenous population: the Census (and the 5% Census Sample File – ‘CSF’) that was used in the Monograph; the National Aboriginal and Torres Strait Islander Social Survey (‘NATSISS’);[6] and the National Aboriginal and Torres Strait Islander Health Survey (‘NATSIHS’).[7] The NATSISS was most recently carried out in 2008, with data now available for analysis. The most recent NATSIHS was carried out in 2004–05, with the next survey scheduled for 2010–11.

Analysis of the Indigenous lifecourse alone may not be a sufficient reason to make changes to the overall approach to the collection of Indigenous statistics. However, the three surveys mentioned will provide the majority of the data that governments use to track progress in meeting their Closing the Gap targets. The evidence base used to track and meet these targets would be much more robust if it were possible to follow individuals through time, not just relying on repeated cross-sections of data. It is important that the NATSISS and NATSIHS are kept as nationally representative as possible, and hence replacing them with a longitudinal survey that is likely to suffer from significant sample attrition is not a viable alternative. However, the Indigenous population is also one of the most surveyed populations in Australia; adding an additional large-scale survey to the congested schedule may place too onerous a burden on the Indigenous community.

One alternative would be to implement a rolling-panel approach to the collection of national statistical datasets. Under this approach, households are retained in the sample for a fixed number of surveys but are eventually dropped out and replaced to keep the data representative of the nation as a whole. A hypothetical structure of a six-year collection cycle might begin with a NATSISS in 2012 (two years ahead of schedule), another in 2018, as well as a NATSIHS in 2016 (six years after the next survey) and 2021. In the intervening years, it is proposed that a reduced module of questions be asked that allow key lifecourse events to be tracked and the Government’s Closing the Gap targets to be analysed. This survey is referred to as the National Closing the Gap Survey (‘NCGS’) and, depending on costs, could be carried out on a subset of the original cohort only. This proposed structure is outlined below

Table 1 Hypothetical rolling panel for collecting longitudinal Indigenous information

Cohort 1
Cohort 2
Cohort 3
Cohort 4
















In essence, Cohort 1 would be given the NATSISS questionnaire in 2012, the NCGS in 2013 and 2014 and the NATSIHS questionnaire in 2015. Cohort 2 would be given the NATSIHS questionnaire in 2015, the NCGS in 2016 and 2017 and the NATSISS questionnaire in 2018. National estimates for the 2012 NATSISS would use Cohort 1, while national estimates from the 2015 NATSIHS would use Cohort 2. This would then be repeated using Cohort 3 and Cohort 4.

The benefits of the above structure are threefold. Firstly, it would be possible for the first time to undertake robust longitudinal analysis of a core set of Indigenous outcomes. This would be restricted to the core set of questions that are available on the NATSISS, the NATSIHS and the new NCGS. Major aspects of the Closing the Gap agenda like employment, education and health, as well as some of their determinants like housing, crime and mobility would therefore be priority data items. Individuals would also only be tracked for four years. Nonetheless, research on these surveys would likely yield vital policy-relevant findings.

The second benefit of the above structure (as opposed to a single longitudinal study) would be that the initial sample for the major surveys would still be nationally representative. The third major benefit is that, by overlapping the cohorts, the representativeness of the longitudinal aspects of the cohorts could be tested against the new cohorts that replace them. For example, the characteristics of Cohort 1 in 2015 could be tested against the characteristics of Cohort 2 in the same year. It may not be possible to maintain a sufficient sample to undertake robust through-time analysis for all jurisdictions. However, the Closing the Gap targets are set at the national level, hence it is vital that they be evaluated in these broad terms.

The above structure would clearly require a significant investment from all levels of government as well as (importantly) survey participants. It may not be possible for the Australian Bureau of Statistics to follow such an approach under its existing budget. However, compared to the investment governments have made, and will need to make, in order to substantially reduce Indigenous disadvantage, the investment in an adequate data collection is likely to be minimal in comparison.

In addition to properly conducted randomised controlled trials,[8] longitudinal information is the only way to truly analyse the Indigenous lifecourse, the determinants of Indigenous socioeconomic disadvantage and the types of social and economic policies that are likely to result in COAG’s Closing the Gap targets being met. Until such datasets are available, robust analyses of individual trajectories and the timing of key events remain elusive.

Dr. Nicholas Biddle is a Research Fellow at the Centre for Aboriginal Economic Policy Research at the Australian National University. He has a PhD in Public Policy, a Master of Education and a Bachelor of Economics.

Mandy Yap is a Research Officer at the Centre for Aboriginal Economic Policy Research. She has Honours in Applied Economics and a Bachelor of Business Administration.

[1] Nicholas Biddle and Mandy Yap, Demographic and Socioeconomic Outcomes across the Indigenous Lifecourse: Evidence from the 2006 Census (2010).

[2] See Melbourne Institute of Applied Economic and Social Research, The Household, Income and Labour Dynamics in Australia (HILDA) Survey <> .

[3] Australian Institute of Family Studies, Growing up in Australia: The Longitudinal Study of Australian Children <> .

[4] Department of Education, Employment and Workplace Relations, The Longitudinal Survey of Australian Youth <> .

[5] Department of Families, Housing and Community Services, Footprints in Time: The Longitudinal Study of Indigenous Children <> .

[6] Australian Bureau of Statistics, National Aboriginal and Torres Strait Islander Social Survey <>.

[7] Australian Bureau of Statistics, National Aboriginal and Torres Strait Islander Health Survey <> .

[8] Andrew Leigh, ‘Evidence-Based Policy: Summon the Randomistas?’ in Productivity Commission, Strengthening Evidence Based Policy in the Australian Federation, Volume 1: Proceedings, Roundtable Proceedings (2009).

AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback