Print

National Disability/ CaLD & NESB Datacube Project 2013 - 2014

“This publication has been prepared by the National Ethnic Disability Alliance for the Australian Government, represented by the Department of Social Services. The views expressed in this publication are those of National Ethnic Disability Alliance (NEDA) and do not necessarily represent the views of the Australian Government”

Project Proposal and Outcomes:

This proposal of the NEDA disability data cube was in essence the foundation of the NEDA Activity Work Plan 2013-2014; this project formed the key requirements of NEDA to the Department over the 12 month period 2013 - 2014.

It should be noted that the scope of this project was vast and required many resources both in house and from external sources. It was hoped that this project would be completed by the due date and this was realised; but in saying this it should be noted that this is a project with ongoing requirements for it to be maintained and will require the updating of data both going in and the collation of data going out.

It is thought that this project will continue to evolve over the coming months and years growing and in-turn offering greater insight not just into the disability demographic but that of the CaLD and NESB community that live with disability in Australia. This project will require ongoing resources for future data outcomes to be realised.  

It should be noted that the department has already benefited from the data gathered via several conversations held by NEDA staff and Departmental representatives with regard to the NDIS tier 3 numbers, these numbers were revised after the data collected from the project showed a short fall in the numbers purported by the Government.

The numbers were revised from 411K to 460K people; this was based on early data from the NEDA data cube that suggested a short fall in the government numbers with regard to the CaLD/ NESB people living with disability, suggesting a future unmet need with this cohort.

NEDA has also presented data from the project to the DES conference held in SA late last year and has also supplied several member organisations with data collected ie: AFDO, FECCA, NDIA, DES, two NGO’s that operate at a national level as well as the ACT Human Rights Commission and our member organisation in Queensland - AMPARO.

In the last twelve months numbers collated form the data cube have already been incorporated in several position statements; ie:

  1. NEDA’s response to: The Inquiry into the Migration Amendment (Regaining Control over Australia’s Protection Obligations) Bill 2013
  2. AFDO for their response to Australian Education Reforms -2014
  3. Submission for the National Disability Strategy 2014 Progress Report to COAG - NEDA
  4. Inquiry into the Migration Legislation Amendment Bill (No. 1) 2014 - NEDA
  5. Joint submission to Affordable housing inquiry – DANA/ NEDA- 2014
  6. Review of equal recognition before the law and legal capacity for people with disability – AFDO/ NEDA 2014
  7. Disability Employment Services (DES) Consumer Engagement Project 2014
  8. NDIS CaLD/ NESB reference group data presentation NEDA – 2014

 

NEDA has also continued with its other core business during the production of the Data Cube, with regards to the many Government endorsed reference groups and committees we attend, offering advice to other Government Departments and the community sector with regard to the needs of our consumer base.

At these forums NEDA has often discussed the Data Cube project and the need for relevant data to be collected on the NEDA cohort, encouraging providers to use the data cube, asking for feedback.

It was proposed that the National Ethnic Disability Alliance commission the creation of a national disability data cube which provided estimates of disability for persons from a Non English Speaking Background at local government area or its equivalent geography for each jurisdiction.

The data cube has integrated where possible information from the 2011 and 2006 ABS Census of Population and Housing and the 2009 and 2003 Survey of Disability, Ageing and Carers. Information from the Home and Community Care Minimum Data Set and National Disability Agreement Minimum Data Set were also summarised.

There will be estimates of settlement by visa type, local government area and jurisdiction.

Suggested Tables

 

  • Migration Stream by Birthplace
  • Visa Type by Birthplace
  • Migration Stream by Religion
  • Main identified language spoken at by English Level Proficiency in Spoken English/Language by need for assistance by estimates for disability
  • Age by Language Spoken by Need for Assistance by Birthplace by estimates for disability
  • Religion by Language[1] Spoken by Age by Need for Assistance by estimates for disability
  • Religion by Place of Birth by Age by Need for Assistance by estimates for disability
  • Religion by Language Spoken by Need for Assistance by estimates for disability
  • Religion by Place of Birth by Need for Assistance by estimates for disability
  • Ancestry by Language Spoken by Need for Assistance by estimates for disability
  • Ancestry by Place of Birth by Need for Assistance by estimates for disability
  • Ancestry by Religion by Need for Assistance by estimates for disability
  • Ancestry by Age by Need for Assistance by estimates for disability
  • Service usage Home and Community Care by language spoken and Place of Birth

 

 

There are some limits to the validity of the estimates as smaller communities will have a higher error associated with the calculations used (Appendix A). The estimates for disability will be based on ABS methodology. For further discussion of the limitations of the approach adopted see Disability, Ageing and Carers, Australia: Summary of Findings 2009

 

Project Structure

 

The project will consist of two phases:

 

  1. 1.The development of the data cubes which has taken approximately 26 to 30 weeks to complete, aspects of it are on-going as mentioned previously.

 

  1. 2.The establishment of a process to ensure that end users have sufficient understanding to obtain the maximum benefit from the data cubes, NEDA has coopted its membership to use the tool and offer advice on its useability and the information contained concerning the quality of data produced. This is an ongoing process of “peer review”.

Datacube Methodology:

The NEDA Datacube was developed from information provided from the settlement database, the 2011 Census and the Survey of Disability, Ageing and Carers2009. The approach used is called IPF (Iterative Proportional Fit) ‘Iterative Proportional Fitting’ (IPF) is a mathematical procedure originally developed to combine the information from two or more data sets. IPF is a well-established technique.

The tables are displayed in a Tableau format which allows for easy manipulation of several variables. There is a limit to the amount of data which could be displayed in any one database. The tables included in the current Datacube are Disability Support Payment for 2010[2], Age profile by country of birth from ABS, Religion, Humanitarian Visa recipients (Settlement Database) and from the Survey of Disability, Ageing and Carers 2009) Severe/Profound[3], Mild, Moderate and General Disability estimates.  

The assumption used in the development of the estimates of CaLD and NESB disabilities is that the age of disability[4] will be similar in all communities and would compensate for the under estimations produced by the Survey of Disability, Ageing and Carers (SDAC)[5].

The settlement database has information on visa arrivals in Australia. The Settlement Database is an internal database developed by the former Department of Immigration and Citizenship (DIAC) to provide statistical support for the department that also enables the distribution of immigration, multicultural and population data to other government agencies and the general community.

The database brings together data from various sources including the Settlement Details form (Form 886), and departmental systems used to process migration applications both in Australia and at overseas posts. Geographic location data is also captured when the settler enrols for services such as Adult Migrant English Program (AMEP) classes in Australia or Medicare.

DIAC uses the data for 'Settlement Planning', a process by which the department identifies which new arrivals are in most need of assistance to settle successfully in Australia, what their particular needs are, and how to best target policy and programmes to meet those needs. DIAC also uses the database to assist in the settlement planning of humanitarian entrants to regional Australia.

The small area estimates produced for the now disbanded Disability Policy and Research Working Group by ABS provides the local area estimates by local government areas and statistical local areas for most of Australia. The small area estimates are derived from the 2009 Survey of Disability Ageing and Carers.

For national consistency, the geography for the Datacube was based on local government areas, with the exception of the ACT, NT and Brisbane where the geographies were based on SRS.

What has been produced is the actual tool located on the NEDA website, this project was demonstrated at the DSS office on: May 27th 2014, in attendance from NEDA were Suresh Rajan NEDA President, Dwayne Cranfield NEDA CEO and Brian Cooper the project architect and several members of the DSS team.

 

Access to the data cube can be found on the NEDA website at:      www.neda.org.au and found under the tab: Statistics.

 

Appendix A

Table 1 - Disability by Degree of Limitation or Restriction

 

 

INTRODUCTION

  1. 1.Request 1 consists of predictions of the proportion of persons in private dwellings having a disability by degree of limitation or restriction. Each person with a disability is classified to one of the following limitation or restriction categories:
  • Profound/Severe core activity limitation Moderate core activity limitation
  • Mild core activity limitation
  • Any Disability (Includes profound, severe, moderate, mild limitations as well as schooling or employment restriction and disability with no restriction or limitation)

2. To maintain the quality and usability of estimates, some of the categories of restriction or limitation were collapsed (the collapsed categories being those listed above). For more details on the full (un-collapsed) categories of restriction or limitation see the Explanatory Notes and ABS Catalogue No. 4430.0.  

 

 

MEASURES OF ACCURACY

3. The relative root mean squared error (RRMSE) provides an indication of the mean squared deviation of the estimate from the true value. As a general rule of thumb, estimates with Relative Root Mean Squared Errors (RRMSEs) greater than 25% should be used with caution and estimates with RRMSEs greater than 50% are considered too unreliable for general use. These cases have been indicated by asterisks in the following tables.

 

4. However, it is important to recognise that for very small estimated proportions, the RRMSEs may be inflated. The RRMSEs have been used to construct 95% confidence intervals around the estimates of proportion. Where inflated RRMSEs are an issue, we recommend also using confidence intervals to judge the accuracy of the estimates.

 

5. For more explanation of RRMSEs and confidence intervals, see the Explanatory Notes and Appendices.

 

LEVEL OF ESTIMATES

6. For NSW and WA, estimates are provided at the Local Government Area (LGA) level. Estimates for the rest of Australia are provided at the Statistical Local Area (SLA) level, except for ACT, NT, and Brisbane estimates. The SLA level of geography was considered to be too small to produce reliable estimates for these regions. Instead, the ACT and NT estimates have been released by Statistical Sub Division (SSD), and Brisbane estimates have been released by Statistical Region Sector (SRS).

 

SOME GUIDELINES

7. Estimated private dwelling populations for each SLA are for the year 2009.

 

8. The small area estimates are for private dwellings only, and exclude remote and sparsely settled parts of Australia, non-Australian diplomatic personnel and members of non-Australian defence forces (and their dependents) stationed in Australia.

 

9. Use the small area estimates for getting a picture of the likely distribution of people with disabilities across regions.

 

10. Do not expect the proportions to be appropriate for every region. Some regions will differ from predictions because of local effects that are not captured by the model.

 

11. The small area estimates are a tool. Used in conjunction with an understanding of local area characteristics and their quality limitations, they should assist in making sensible decisions on issues involving the regional distribution of disability.

 

12. Aggregation of the small area estimates to larger regions (such as regional planning areas) will improve the accuracy of the estimates. See the aggregation worksheet for some worked through examples.

 

13. The estimated disability proportions are given so that they may be directly applied to the private dwelling population counts for that SLA cell.

Source: Table 1 - Disability by Degree of Limitation or Restriction (National Disability Administrators Small Area Estimates)

 

“This publication has been prepared by the National Ethnic Disability Alliance for the Australian Government, represented by the Department of Social Services. The views expressed in this publication are those of National Ethnic Disability Alliance (NEDA) and do not necessarily represent the views of the Australian Government”



[1] Currently polylingualism is not recorded in the Australian Census or in most administrative data sets. Many Australian residents and citizens are articulate in several languages, these being the national language, local languages and trade languages. As an example a person born of Chinese heritage in Vietnam will speak their local dialect (Teochew, Hokkien or Haka as examples), with either Cantonese or Mandarin (trade languages) or both and Vietnamese (national language). The census will only require that one language be identified.

[2] Currently Centrelink data are only available as aggregated data with a limited choice of field for aggregation.

[3] The Severe/Profound grouping is the Tier 3 group for the National Disability Insurance Scheme(NDIS)

[4] Age as recorded in the SDAC, for further explanation see Disability, Ageing and Carers, Australia: Summary of Findings 2009