Accurate data about Aboriginal and Torres Strait Islander people are needed to guide policy formulation, program development and service delivery, towards closing the gap in disadvantage between Aboriginal and Torres Strait Islander people and non-Indigenous Australians. Progress is difficult to measure accurately because many Aboriginal and Torres Strait Islander people are not consistently identified as such in key data sets.
Data linkage offers a cost-effective approach to enhancing the completeness and consistency of Indigenous status information on key data sets, for purposes of statistical reporting. There are, however, no nationally agreed approaches on how to deal with missing or inconsistent Indigenous status reporting across data sets. This leads to different methods being used and difficulties in interpreting findings, particularly when comparing results across studies.
To ensure that a consistent and informed program of data linkage work is carried out across Australia, the Council of Australian Governments (COAG) tasked the Australian Institute of Health and Welfare (AIHW) and the Australian Bureau of Statistics (ABS) to develop national best practice guidelines for linking data relating to Indigenous people.
To inform the preparation of the Guidelines, COAG also tasked the AIHW and ABS to conduct a review of past, ongoing and planned data linkage studies in Australia and overseas that have an Indigenous focus. The review studies are published separately, as attachments to these Guidelines. The AIHW and ABS not only worked closely but also consulted widely in the development of these Guidelines. The review of past, current and planned data linkage studies points to two main ways in which data linkage is used in studies related to Aboriginal and Torres Strait Islander people.
Data linkage has been used to enhance the quality of Indigenous status information on key data sets, especially where Indigenous status is missing or inconsistently reported across data sets. It has also been used to add value to data sets by bringing together data items from multiple data sets, with a view to carrying out analysis that would be impossible to undertake if the research were based on the individual data sets. In both cases, data linkage has not been used to alter Indigenous status information on source data sets.
The Guidelines focus on six key aspects of data linkage. These are the values and ethics in human research relating to Aboriginal and Torres Strait Islander people, transparency and accountability, the quality of the Indigenous status variable on key data sets, the quality of the variables that are used for the linkage, the quality of the linkage itself, and the methods and algorithms used to derive Indigenous status where Indigenous status varies across the individual data sets in the linked data set.Accurate data about Aboriginal and Torres Strait Islander people are needed to guide policy formulation, program development and service delivery, towards closing the gap in disadvantage between Aboriginal and Torres Strait Islander people and non-Indigenous Australians. Progress is difficult to measure accurately because many Aboriginal and Torres Strait Islander people are not consistently identified as such in key data sets.
Preliminary material: Preface; Acknowledgments; Abbreviations; The Principles and Guidelines
1 Background to the Guidelines
- Protocols of data linkage
- Key concepts in data linkage
- What is data linkage?
- Uses of data linkage
- Data linkage methods
- The data linkage process
- Statistical linkage keys
- Organisation of the Guidelines
- Context of the development of the Guidelines
- Purpose of the Guidelines
2 Values and ethics in Aboriginal and Torres Strait Islander research
- Aboriginal and Torres Strait Islander core values
- Project approval
3 Quality of Indigenous status information in data collections
- Factors affecting the quality of Indigenous statusinformation
- The data collection environment
- The respondent environment
- Checklist for understanding the quality of Indigenous status information
4 Quality of linkage variables
- Quality issues for linkage variables
- Name changes
- Date of birth
- Mobility and levels of geographic reporting
- Impact of quality of linkage variables on data linkage
5 Assessment of quality of data linkage
- Determinants of linkage quality
- Quality of blocking and linking variables
- Blocking and linking strategy
- Assessing the quality of data linkage
- Approaches to assessment of data linkage quality
- Measuring linkage quality through clerical assessment
- Other approaches to assessing the quality of linkage
- Measures of quality of data linkage
- Other measures of quality of linkage from clerical assessment
- Characteristics of unlinked records
- Edit checks that should be performed before analysing linked data
- Familiarisation with the data collections in the study
- Basic frequency analyses of individual data sets beforemerging data
- Knowledge of quality of input data sets and quality of linkeddata set
- Logic or internal consistency checks on the linked data set
6 Methods for deriving Indigenous status
- General issues to consider when choosing a method
- Algorithms for deriving Indigenous status
- Simple algorithms
- Complex algorithms
- Aggregate methods for deriving Indigenous status
- Quality checks for derived Indigenous status
- Why is transparency important?
- What should be disclosed, by whom and to whom?
- Before the project starts
- After the project has been completed
- Documentation considerations for data linkage institutions
End matter: Glossary; References; List of tables; List of figures; Forthcoming publications