Microdata and TableBuilder: National Aboriginal and Torres Strait Islander Health, Australia

The National Aboriginal and Torres Strait Islander Health Survey (NATSIHS) collects information on the health and wellbeing of Aboriginal and Torres Strait Islander people.

The 2018–19 NATSIHS is the fourth health survey of Aboriginal and Torres Strait Islander people conducted by the ABS. See National Aboriginal and Torres Strait Islander Health Survey, 2018–19 for summary results, methodology and other information.
You can use this data in: 

• DataLab - analyse detailed microdata in a remote environment
• TableBuilder - produce your own tables and graphs (will be available in early 2020)

Compare access options to see what's right for you or Apply for access.

The following table shows the levels available in each of the microdata products and the information contained on those levels:

Use the data item lists in the Data downloads section to confirm this dataset includes what you need for your research before purchasing your subscription.

Datasets from the National Aboriginal and Torres Strait Islander Health Survey are hierarchical in nature. A hierarchical data file is a way of presenting information which describes one to many, or many to many, relationships (e.g. a person may report multiple days on which alcohol was consumed and also multiple types of alcoholic beverages on each of these days).

The following table shows the hierarchical file structure and the relationship between each level:

The DataLab environment allows real time access to detailed microdata files from the 2018–19 National Aboriginal and Torres Strait Islander Health Survey through a portal to a secure ABS environment. The level of detail provided within the DataLab exceeds that previously provided in the Expanded CURF and provides researchers greater access to the data.

The DataLab is an interactive data analysis solution available for users to run advanced statistical analyses, for example, multiple regressions and structural equation modelling. The DataLab environment contains up-to-date versions of SPSS, Stata, SAS and R analytical languages. Controls in the DataLab have been put in place to protect the identification of individuals and organisations. All output from DataLab sessions is cleared by an ABS officer before it is released. 

For more information about the DataLab, please see the About the DataLab page.

The Expanded CURF and the use of the Remote Access Data Laboratory (RADL) have been discontinued for this collection. Users are encouraged to consider the DataLab as an improved mechanism to access detailed microdata.

Two test files have been created to enable researchers and analysts to become more familiar with the data structure and prepare code and programs before applying for or beginning a DataLab session.

The test files do not contain real data, and cannot be used for analysis. They mimic the structure of the detailed microdata in that they have the same data items and allowed values.

A Selected Person level and Household level test file are available as free downloads from the Data downloads section.

The following table shows the number of records on each level and the weighted counts, where applicable.

There are three weight variables on the detailed microdata files:

• Household weight (FINHHWT) - benchmarked to produce Aboriginal and Torres Strait Islander household estimates.
• Person weight (FINPERWT) - benchmarked to the total Aboriginal and Torres Strait Islander population.
• Hearing weight (FINHEAWT) - must be used whenever an item is used from the hearing level.

The weighted counts for the selected person and the hearing level are the same. This is because only a maximum of one hearing test was counted for each person. However, the hearing weight must be used whenever an item is used from the hearing level. If the correct weight is not used, the errors associated with the data will be incorrect.

There is no weight associated with the other levels. This is because the records are repeated for each person. If, for example, FINPERWT is merged onto the Conditions level, it will be attached to each condition record and therefore be repeated for each person where they have more than one condition.

Every record on each level of the file is uniquely identified. The identifiers ABSHIDD, ABSPID, ABSDID, ABSTID, ABSCID, ABSMID, ABSBID and ABSXID appear on all levels of the file. Where the information for the identifier is not relevant for a level, it has a value of 0. See the data item list for details on which identifier equates to which level.

Each household has a unique thirteen digit random identifier, ABSHIDD. This identifier appears on the household level and is repeated on each level on each record pertaining to that household. The combination of identifiers uniquely identifies a record at a particular level as shown below:

1. Household = ABSHIDD
2. Selected person = ABSHIDD + ABSPID
3. Alcohol - Day consumed = ABSHIDD + ABSPID + ABSDID
4. Alcohol - Type consumed = ABSHIDD + ABSPID + ABSDID + ABSTID
5. Binge - Type consumed = ABSHIDD + ABSPID + ABSBID
6. Conditions = ABSHIDD + ABSPID + ABSCID
7. Medications = ABSHIDD + ABSPID + ABSMID
8. Hearing = ABSHIDD + ABSPID + ABSXID

The household record identifier, ABSHIDD, assists with linking people from the same household, and also with household characteristics such as geography (located on the household level) to the person records.

The TableBuilder User Guide provides you with information about how to create basic tables, custom groups, graphs and large tables. It also includes practical examples and video tutorials.

When tabulating data in TableBuilder, person weights are automatically applied to the underlying sample counts. Weighting is the process of adjusting results from a sample survey to infer results for the total population. To do this, a 'weight' is allocated to each sample unit. The weight is the value that indicates how many population units are represented by the sample unit.

There are three weights on the file: 

• Household weight PAA - benchmarked to produce Aboriginal and Torres Strait Islander household estimates.
• Person weight PAA - benchmarked to the total Aboriginal and Torres Strait Islander population.
• Hearing test weight - must be used whenever an item is used from the hearing level.

When analysing a Household level item, you may want to change the summation weight from persons to households. For example, if you wanted to know the number of households in a state rather than the number of persons living in that state. In this case, you would add the weight called 'Household weight PAA' from the Summation Options heading. The default weight is 'Person weight PAA' - a count of persons.

The weighted counts for the selected person and the hearing level are the same, which is because only a maximum of one hearing test was counted for each person. When analysing any item from the Hearing level, you must use the 'Hearing test weight'. You would add the weight called 'Hearing' from the Summation Options heading. If the correct weight is not used, the errors associated with the data will be incorrect.

Most data items included in the TableBuilder file include a 'Not applicable' category. The classification values of these 'Not applicable' categories, where relevant, are shown in the data item list on the Downloads page. The 'Not applicable' category generally represents the number of people who were not asked a particular question or the number of people excluded from the population for a data item when that data was derived (e.g. Age started having daily insulin injections is not applicable for people who do not have daily insulin injections).

The population relevant to each data item is identified in the data item list and should be kept in mind when extracting and analysing data. The actual population estimate for each data item is equal to the total cumulative frequency minus the 'Not applicable' category.

TableBuilder includes a number of continuous variables:

• They can have a response value at any point along a continuum.
• Some continuous data items are allocated special codes for certain responses (e.g. 9999 = 'Not applicable)
• When creating ranges in TableBuilder for such continuous items, special codes will automatically be excluded. Therefore the total will show only 'valid responses' rather than all responses (including special codes).
• Continuous items with special codes have a corresponding categorical item on the Person level that provides the ability to display data for the special code.
• Any special codes for continuous data items are listed in the data item list in the Data downloads section.

A number of data items allow a respondent to have more than one response. For these items, a person is counted against each category they responded to and consequently the sum of the categories may be different to the total. The table below displays the multiple response data item 'Sources of personal income (NR only)' with the Person weight PAA. For example, a person who receives a 'wage or salary', 'superannuation or annuity' and an 'Other regular source' of income will be counted against each of these three categories.

Multiple response data items are identified in the Data Item List in the label using the indicator ‘<multiple response>’.

A confidentiality process called perturbation is applied to the data in TableBuilder to avoid releasing information that may lead to the identification of individuals, families, households, dwellings or businesses. See Confidentiality in the TableBuilder user guide.

This release previously used catalogue number 4715.0.55.001.