COLLECTION AND INTERPRETATION OF ABS DISABILITY DATA
Collecting disability data poses particular challenges due to the complex nature of experiences that affect the physical, social and emotional well-being of individuals with disability, and their families. The subjective and variable nature of some disabilities may mean that responses are affected by factors such as:
- A person's energy levels, optimism, pain or depression at the time of the survey.
- The episodic or seasonal nature of some conditions (e.g. epilepsy, asthma) – this may result in respondents not reporting conditions as causing restrictions if they are not experiencing difficulties at the time of interview.
- Adaptation to limitations - restrictions or limitations may not be reported if the respondent has adapted so completely to their disability that they are no longer conscious of an inability to perform certain tasks.
- Sensitivities around some conditions - underreporting of disability may occur when there are sensitivities around conditions such as mental illness or mental deterioration, and the need for help with personal care activities.
The way in which the data is collected can also influence the data collected. For example:
- Respondents answering on behalf of other people in a household - while the person responding on behalf of the household might be more objective in assessing a household member’s difficulty and need for help, they may not be aware of all the conditions the other person has or how these affect their daily living.
- Different collection methodologies –for example, the use of prompt cards (which show numerous conditions at once as opposed to asking about conditions individually) reduce the number and repetition of questions. They can also lead to responses that differ from a methodology that asks separate, detailed questions about each condition. This is important to understand when comparing disability measures collected using the SDAC, and disability measures collected using the SDM or the Census ‘Need for Assistance’ questions. Questions collected by interviewers (as is the case in the SDAC and SDM) may elicit different responses to those in a ‘self-complete’ form, such as the Census. An explanation of this can be found at https://www.abs.gov.au/ausstats/abs@.nsf/Lookup/2901.0Chapter51852016.
In addition to the considerations outlined above, there are additional challenges when collecting disability information for the Aboriginal and Torres Strait Islander population:
- Perceptions of disability - for many Aboriginal and Torres Strait Islander people, the biomedical model and conceptualisation of disability may be culturally inappropriate1. The concept of disability may not translate in Aboriginal language and cultures2, and these conceptual differences might affect responses to questions concerning disability.
- Intersectional Inequality - some Aboriginal and Torres Strait Islander people may choose not to identify as having disability for fear of experiencing further discrimination. ‘Intersectional Inequality’, where inequality is compounded for people who are members of two or more marginalised groups2, can affect Aboriginal and Torres Strait Islander people with disability, particularly those with more profound or severe disability2.
MEASURING THE ABORIGINAL AND TORRES STRAIT ISLANDER POPULATION
Australia’s Aboriginal and Torres Strait Islander population is statistically measured across all ABS data collections, through self-identification. This means that the Aboriginal and Torres Strait Islander population is defined by individuals who identify as being Aboriginal and/or Torres Strait Islander, based on their own knowledge of their biological ancestry, and who choose to disclose that information3. There are many reasons why an individual may or may not identify as being Aboriginal and/or Torres Strait Islander, and these reasons can vary over time and across different data collections.
Self-identifying populations pose a statistical challenge, as changes to population estimates over time may be partly a result of changes in identification, outside of demographic causes such as births and deaths. The ABS consistently uses the Standard Indigenous Question (SIQ) as a basis across all data collections in order to minimise the statistical impact that self-identification can have on population estimates4.
COLLECTING DISABILITY DATA: BEST PRACTICE
The complex nature of the collection and interpretation of disability data can increase the variability of disability estimates obtained from ABS surveys. Every effort is made to minimise this effect in ABS collections through:
- Questionnaire design - questionnaires are designed and carefully tested to achieve objective and repeatable responses wherever possible.
- Avoiding terms that are emotive or prone to wide interpretation – terms such as ‘disability’ are not directly used in questionnaires. Disability measures are derived from a series of questions that avoid emotive terms and judgements, and instead focus on what a person can or cannot do.
In the development and collection of the NATSISS and NATSIHS, the ABS has included special measures designed to improve the cultural appropriateness and relevance of these surveys to the Aboriginal and Torres Strait Islander population. These measures include:
- Cultural awareness training - all ABS Interviewers are required to undergo cultural awareness training.
- Questionnaire testing – the NATSISS and NATSIHS questionnaires are tested and developed to ensure respondents have an understanding of, and are engaged with, the surveys.
- Special measures for engagement in remote areas – special measures are taken when conducting surveys in remote areas, and in Aboriginal and Torres Strait Islander Communities. These include:
- Contacting communities and health clinics by phone and through a Community Approach Letter in advance of any survey.
- Modifying how surveys are conducted to account for language and cultural differences.
- Interviewers in remote areas work in teams of two, one male and one female, to collect survey information.
- Interviewers are accompanied, wherever possible, by local facilitators (usually one male and one female).
- Local facilitators introduce interviewers to the community, assist in the conduct and completion of interviews, help to explain the purpose of the survey and assist respondents in understanding questions where needed.
In addition to this, the ABS has established a network of Aboriginal and Torres Strait Islander Engagement Managers
across ABS State and Territory offices. The role of Engagement Managers is to:
- Foster collaborative partnerships with Aboriginal and Torres Strait Islander communities
- Increase understanding and participation in ABS collections
- Facilitate the return of information to communities and
- Improve the quality and relevance of Aboriginal and Torres Strait Islander statistics to better meet the needs of Aboriginal and Torres Strait Islander communities and meet the policy needs of government.
The ABS has also implemented a Reconciliation Action Plan
that provides a framework for the whole Agency to build and improve connections, cultural understanding and increase the participation of Aboriginal and Torres Strait Islander people in the ABS through employment and greater statistical literacy.
1 Gilroy J, Donelly M, Colmar S, Parmenter T (2016) Twelve factors that can influence the participation of Aboriginal people in disability services. Australian Indigenous HealthBulletin 16(1). Retrieved 12/12/2018 from http://healthbulletin.org.au/articles/twelve-factors-that-can-influence-the-participation-of-aboriginal-people-in-disability.
2 Avery, S. (2018). Culture is Inclusion: A narrative of Aboriginal and Torres Strait Islander people with disability. First Peoples Disability Network (Australia). Sydney, Australia.
3 Australian Bureau of Statistics 2013, Information Paper: Perspectives on Aboriginal and Torres Strait Islander Identification in Selected Data Collection Contexts, 2012, cat. no. 4726.0, ABS, Canberra.
4 Petry, B, and Potts, E (2014) 'Measuring indigenous populations across nations: Challenges for methodological alignment'. Statistical Journal of the IAOS. V30, pp. 55-63.