This article raises some interesting and important points, in particular the need for consistent and comprehensive data to be collected on Australia’s Aboriginal residents.

I take issue with a few minor aspects, however.

The author says: “The key requirement of a Census or survey question on Indigenous status is to identify respondents who meet the officially set requirements for Aboriginality.” I disagree. A census can only collect data on people’s self-descriptions, not cross-check those descriptions with objective data. Race, religion, ethnic origin – all of these are fuzzy areas in which a person’s self-description may not accord with “official” or independent criteria. And, of course, “official” definitions of ethnicity may not be accepted by all members of an ethnic group – defining aboriginality is notoriously controversial.

A second quibble is that, in his efforts to highlight genuine deficiencies in the data, Brendan may downplay the significance of the official explanations for the rising census count of Aboriginal people.

Under-counting has been a serious problem. Here in Western Australia, there was an under-count of Indigenous people in the 2006 census that caused headaches for service delivery agencies.

The tendency for children of mixed marriages to be identified as indigenous may largely explain why the number of indigenous children exceeds what the fertility rate among Aboriginal women would suggest. But it cannot explain the reported growth in people identified as indigenous WITHIN age cohorts. While changes to the census questions may account for some of this change, I suspect that the primary official explanation really does account for most of this increase - the increasing tendency of people to self-identify as Aboriginal.

Perhaps there are other, more effective ways to collect data on Australia’s Aboriginal people than through the census

Rhian

Both your points are broadly valid.

The Census can indeed only collect data based on the self-assessments of respondents, and (as you correctly indicate) has no means of cross-checking those responses with objective data. My point is that the Census question on Aboriginality does not itself facilitate correct answers because the question itself (in asking about Indigenous Origin but not addressing Identity) is inadequate.

You are correct in noting that a lot of improvements have been introduced by the ABS over the years, both in-the-field and in terms of the overall tabulations (to correct problems such as under-count and certain specific types of errors). On the other hand, I am not convinced that all the changes made to the Indigenous Census question have been for the better.

I am not at all denying that growth within Indigenous age cohorts over time is partly due to increased rates of self-identification as Indigenous. My point is that this important influence has tended to be used as a blanket explanation leading to the exclusion of other likely contributory explanations. Growth within Indigenous age cohorts is also likely to be due to a growing proportion of those ticking the Indigenous origin box, who may have an identity other than Indigenous as their main identity.

The solution is for the Census question to ask about main identity as well as Indigenous origin.

Brendan

Hi Brendan

I think we broadly agree. The census is not great for collecting detailed info on a single racial group, but we rely on it heavily for time series and comparative profiles. Probably a mix of better-crafted and consistent census questions and a more detailed consistent and periodic sample survey (itself easier said than done) would be the way to go.

There’s a trade-off, too, between getting the best possible questions and maintaining consistency for the sake of meaningful time series. The worst of all possible worlds is when we get neither!

There is another reason for the large increase in Australians identifying as indigenous. The benefits that are available to indigenous Australians in identifying as indigenous.
Scholarships, public service jobs and places in university courses are available to those who identify as indigenous. Indigenous programs which should be aimed to lift indigenous Australians out of poverty are now been used by middle class Australians to benefit themselves as they are partly (often a small part) indigenous.