Academic journal article Australian Journal of Social Issues

Evidence for the Use of an Algorithm in Resolving Inconsistent and Missing Indigenous Status in Administrative Data Collections

Academic journal article Australian Journal of Social Issues

Evidence for the Use of an Algorithm in Resolving Inconsistent and Missing Indigenous Status in Administrative Data Collections

Article excerpt

Introduction

Administrative data collections maintained by Australian State, Territory, and Commonwealth human services agencies provide the basis for a range of key measures of social progress used in official reporting in Australia. These include measures used in the Council of Australian Governments (COAG) Closing the Gap reporting framework, designed to monitor the progress of Indigenous (1) Australians, relative to non-Indigenous Australians in living standards, life expectancy, education, health and employment. The Closing the Gap National Indigenous Reform Agreement has six major targets, including: closing the life expectancy gap within a generation; halving the gap in mortality rates for Indigenous children under five within a decade; and halving the gap for Indigenous students in reading, writing, and numeracy within a decade (COAG 2008). Measurement of progress against these targets depends upon the provision of administrative data.

Using administrative data for government reporting makes use of existing information that is a by-product of the day-to-day activities of human services agencies, rather than incurring the additional costs, resources and respondent burden that traditional survey methods impose (Holman et al. 2008). Despite the accepted benefits of using administrative data as a measurement tool, there are limitations, especially regarding consistent and reliable recording of the Indigenous status of individuals over time. Although Indigenous status in administrative collections in Australia is typically determined by a process of self-identification, Indigenous identification occurs in a complex cultural and historical context. For example, some individuals have reported that their willingness to self-identify as Indigenous is affected by perceptions of possible discrimination, and by the purpose of the data collection (ABS 2012b). Indigenous identification is also voluntary, so 'missing' or 'not stated' can be considered a valid response.

Historically, administrative data has been collected for the business needs of the respective custodian agencies, not for COAG reporting, and the quality and type of data collected has reflected this. Data collection practices can also vary between locations and across time, leading to further variation in data quality. In some settings the actual question regarding Indigenous identification may not be asked, and is either presumed by the collector or simply left unknown and missing. Another factor is the willingness of respondents, when asked, to provide personal information in a setting where they may feel uncomfortable about doing so. It remains the right of any individual to choose which Indigenous status they wish at the point of data collection, and that status may legitimately change from one identification occasion to the next. As a consequence of these issues, Indigenous status reporting on administrative data collections can be affected by missing, inconsistent and incorrect data. When multiple administrative data collections are statistically linked, these problems further increase the complexity of reporting, especially when an individual has differing Indigenous status recorded both within and across multiple data collections.

COAG Closing the Gap indicators are measures of the disparity between Indigenous and non-Indigenous Australians across a range of measures in living standards, life expectancy, education, health and employment. Without consistency in measurement of Indigenous status it becomes increasingly difficult to compare outcomes accurately across multiple time points, and misguided conclusions can be drawn from the data (Council 2012). Compositional changes in Indigenous and non-Indigenous populations can affect estimates of the magnitude of the gap. For example, although not an administrative data collection, results from the Australian Bureau of Statistics (ABS) 2011 Census of Population and Housing show substantial increases since the 2006 Census in the number of Indigenous persons in the Australian jurisdictions of Victoria (26. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.