FOI Request LS7888, Schedule of Released Documents [PDF 777kb] (pdf)
Download cached file | Download from AEC--- Page 1 --- Request for: documents relating to the design, testing and implementation of the AEC's new methodology during 2018 & 2019 for its Indigenous Electoral Participation Program (IEPP), namely: 1. 2. 3. 4. 5. 6. Copies of documentations from 1 Jan 2018-31 Dec 2019 (inclusive) outlining planning for the new methodology. This should include any documents relating to the nature of catalyst for the new methodology starting to be planned (e.g. instructions from relevant Ministers to undertake a new methodology); Copies of any progress reports filed between 1 Jan 2018-31 Dec 2019 (inclusive) about the progress of the design, testing and implementation of the new methodology—this can include internal emails between staffers if necessary; Copies of documentation from 1 Jan 2018-31 Dec 2019 (inclusive) outlining how implementation of the new methodology was tracked by the AEC, including any comparisons of results conducted with prior methodologies for assessing Indigenous enrolment rates; Copies of final summary reports relating to the outcomes of the new methodology from 1 Jan 2018-31 Dec 2019 (inclusive); Documentation from 1 Jan 2018-31 Dec 2019 (inclusive) outlining the number of Indigenous people involved by the AEC in the planning, design and implementation of the new methodology; and Documentation from 1 Jan 2018-31 Dec 2019 (inclusive) outlining the final cost of the design, testing and implementation of the new methodology—this should also include details of the number of AEC staffers assigned to the task. FOI REQUEST LS7888 SCHEDULE OF RELEASED DOCUMENTS Document No. Description Email from Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to from Warrick Poyser, Consultant, Systems Delivery & Modernisation Section, Operations Branch, AEC re Indigenous data Subsumed in Document No. 4. Date 8/03/2018 3:38 PM 1 of 7 Released under the Freedom of Information Act 1982 --- Page 2 --- Document No. Description Date Email from Warrick Poyser, Consultant, Systems Delivery & Modernisation Section, Operations Branch, AEC to Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Indigenous data Subsumed in Document No. 4. Email dated 1/06/2018 12:00 PM from Angelo Rojas, Senior Project Officer, Roll Program & Community Engagement Section, Roll Management & Community Engagement Branch, AEC to Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Indigenous data with a thread comprising: Subsumed in Document No. 4. Email from Angelo Rojas, Senior Project Officer, Roll Program & Community Engagement Section, Roll Management & Community Engagement Branch, AEC to Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Indigenous data with a thread comprising: Minute from Guy Lewis, Assistant Director, Roll Management Design, Roll Program and Community Engagement Section, Roll Management and Community Engagement Branch to Andrew Gately, First Assistant Commissioner, Network and Election Operations, AEC re Using DHS Indigeneity data to enhance Indigenous enrolment. Paper titled Indigenous Franchise KPIs using Centrelink data Email from Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to Michael Roden, Assistant Director, Business Intelligence, Research & Electoral Integrity, Operations Branch AEC re Publication of Indigenous enrolment and another. Subsumed in Document No. 12. 8/03/2018 5:12 PM 1/06/2018 12:00 PM 4/06/2018 3:45 PM 2/07/2018 undated 24/08/2018 12:58 PM Paper titled Indigenous Franchise KPIs using Centrelink data undated Released under the Freedom of Information Act 1982 --- Page 3 --- Document No. Description Date Email from Michael Roden, Assistant Director, Business Intelligence, Research & Electoral Integrity, Operations Branch AEC to Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Publication of Indigenous enrolment and another. 26/08/2018 1:37 PM Subsumed in Document No. 12. Email from Michael Roden, Assistant Director, Business Intelligence, Research & Electoral Integrity, Operations Branch AEC to Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Publication of Indigenous enrolment and another. Subsumed in Document No. 12. Email dated 27/08/2018 9:47 AM from Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to Michael Roden, Assistant Director, Business Intelligence, Research & Electoral Integrity, Operations Branch AEC re Publication of Indigenous enrolment and another. Subsumed in Document No. 12. Email from Michael Roden, Assistant Director, Business Intelligence, Research & Electoral Integrity, Operations Branch AEC to Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Publication of Indigenous enrolment and another. Email dated 4/12/2018 9:03 AM from Paul Oglethorpe, Assistant Director Parliamentary Advisor, Parliamentary and Party Registration Section, Disclosure, Assurance and Engagement Branch to the AEC’s Executive Leadership Team Mailbox and others re Parliamentary review 3 December 2018. Subsumed in Document No. 18. Email dated 4/12/2018 9:12 AM from Delyse Paliaga, Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to Guy Lewis and others re Parliamentary review 3 December 2018. Subsumed in Document No. 18 26/08/2018 1:42 PM 27/08/2018 9:47 AM 19/09/2018 8:59 AM 4/12/2018 9:03 AM 4/12/2018 9:12 AM Released under the Freedom of Information Act 1982 --- Page 4 --- Document No. Description Email from Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to Michael Roden, Assistant Director, Business Intelligence, Research & Electoral Integrity, Operations Branch AEC re Parliamentary review 3 December 2018. Subsumed in Document No. 18 Email from Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to Delyse Paliaga, Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Indigenous missing. Email dated 5/12/2018 8:20 AM from Michael Roden, Assistant Director, Business Intelligence, Research & Electoral Integrity, Operations Branch AEC to Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Parliamentary review 3 December 2018. Subsumed in Document No. 18 Email from Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to Michael Roden, Assistant Director, Business Intelligence, Research & Electoral Integrity, Operations Branch AEC re Parliamentary review 3 December 2018 Ministerial Submission from Tom Rogers, Electoral Commissioner, AEC to Alex Hawke, Special Minister of State re Publication of Indigenous enrolment rates. Estimates Brief – Hot Issue from Robyn Legg, Assistant Commissioner, Roll Management & Community Engagement Branch, AEC to the Special Minister of State re Indigenous Date 4/12/2018 9:17 AM 4/12/2018 1:32 PM 5/12/2018 8:20 AM 5/12/2018 8:43 AM 14/02/2019 15/02?2019 Email from Daniel Clode, Department of Finance to Jeff Pope, Deputy Electoral Commissioner, AEC re Indigenous population statistics 5/08/2019 9:21 AM Subsumed in Document No. 28. Released under the Freedom of Information Act 1982 --- Page 5 --- Document No. Description Date Email dated 5/08/2019 9:28 AM from Jeff Pope to Andrew Gately, First Assistant Commissioner, Network and Election Operations, AEC re Indigenous population statistics. Subsumed in Document No. 28. Email dated 6/08/2019 1:47 PM from Robyn Legg, Assistant Commissioner, Roll Management & Community Engagement Branch, AEC to Jeff Pope, Deputy Electoral Commissioner, AEC re Indigenous population statistics. Subsumed in Document No. 28. Email dated 12/08/2019 8:52 AM from Jeff Pope, Deputy Electoral Commissioner, AEC to Robyn Legg re Indigenous population statistics Email from Robyn Legg, Assistant Commissioner, Roll Management & Community Engagement Branch, AEC to from Delyse Paliaga, Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Indigenous population statistics Subsumed in Document No. 28. Paper titled Analysis of Indigenous Enrolment. Email dated 12/08/2019 9:46 AM from Delyse Paliaga, Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Indigenous population statistics. Subsumed in Document No. 28. 5/08/2019 9:28 AM 6/08/2019 1:47 PM 12/08/2019 8:52 AM 12/08/2019 9:03 AM undated 12/08/2019 9:46 AM Email from Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to Stephanie Maxwell, Senior Project Manager, Roll Program and Community Engagement Section, Roll Management & Community Engagement Branch, AEC and another re Indigenous population statistics. 12/08/2019 11:19 AM Paper titled Analysis of Indigenous Enrolment undated Released under the Freedom of Information Act 1982 --- Page 6 --- Document No. Description Paper titled Indigenous Franchise KPIs using Centrelink data Email dated 12/08/2019 2:49 PM from Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to Tita Tabije, Data Analyst, Business Intelligence, Research & Electoral Integrity Section, Operations Branch, AEC re Indigenous population data Subsumed in Document No. 32. Email from Tita Tabije, Data Analyst, Business Intelligence, Research & Electoral Integrity Section, Operations Branch, AEC to Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Indigenous population data. Email from Tita Tabije, Data Analyst, Business Intelligence, Research & Electoral Integrity Section, Operations Branch, AEC to Gus Verzosa, BI Manager, Business Intelligence, Research and Electoral Integrity Section, Operations Branch, AEC and others re Indigenous KPI. Subsumed in Document No. 37. Email from Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to Tita Tabije, Data Analyst, Business Intelligence, Research & Electoral Integrity Section, Operations Branch, AEC re Indigenous KPI Subsumed in Document No. 37. Date undated 12/08/2019 2:49 PM 12/08/2019 4:45 PM 30/8/2019 2:29 PM 30/08/2019 2:42 PM Email from Tita Tabije to Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Indigenous KPI. 2/09/2019 10:54 AM Subsumed in Document No. 37. Email dated 3/09/2019 8:57 AM from Angelo Rojas, Senior Project Officer, Roll Program and Community Engagement Section, Roll Management & Community Engagement Branch, AEC to Tita Tabije, Data Analyst, Business Intelligence, Research & Electoral Integrity Section, Operations Branch, AEC and another re Indigenous KPI. 3/09/2019 8:57 AM Subsumed in Document No. 37. Released under the Freedom of Information Act 1982 --- Page 7 --- Document No. Description Email from Tita Tabije, Data Analyst, Business Intelligence, Research & Electoral Integrity Section, Operations Branch, AEC to Angelo Rojas, Senior Project Officer, Roll Program and Community Engagement Section, Roll Management & Community Engagement Branch, AEC re Indigenous KPI Email from Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC to Sean Ferrari, Director, National Enrolment Services, Roll Management & Community Engagement Branch, AEC and others re Indigenous enrolment rates . Subsumed in Document No. 40. Email dated 11/09/2019 1:21 PM from Sean Ferrari, Director, National Enrolment Services, Roll Management & Community Engagement Branch, AEC to Guy Lewis, Assistant Director, Roll Management and Community Engagement, Roll Management & Community Engagement Branch, AEC re Indigenous enrolment rates Subsumed in Document No. 39. Email from Angelo Rojas, Senior Project Officer, National Enrolment Services, Roll Management & Community Engagement Branch, AEC to Sean Ferrari, Director, National Enrolment Services, Roll Management & Community Engagement Branch, AEC and others re Indigenous enrolment rates Estimates Brief -Hot Issue – Supplementary Budget Estimates, October 2019 Current Enrolment figures, trends Redaction code 19. End Date 3/09/2019 2:18 PM 11/09/2019 12:20 PM 11/09/2019 1:21 PM 12/09/2019 10:48 AN 22/10/2019 Released under the Freedom of Information Act 1982
This document is a "SCHEDULE OF RELEASED DOCUMENTS" for FOI Request LS7888, listing 40 documents released by the Australian Electoral Commission (AEC) spanning from March 2018 to October 2019. It details various internal communications, minutes, and papers concerning the AEC's new methodology for estimating Indigenous electoral enrolment rates, primarily utilising Centrelink (DHS) data.
Relevance to FOI Request LS7888:
The schedule directly addresses several aspects of the FOI request:
- Planning for the new methodology and catalyst (point 1): The schedule lists a "Minute... re Using DHS Indigeneity data to enhance Indigenous enrolment" (Doc 4, July 2018) and multiple "Paper titled Indigenous Franchise KPIs using Centrelink data" (undated), which indicate the catalyst (Centrelink data) and planning efforts.
- Progress reports (point 2): Numerous emails between AEC staff, particularly those regarding "Indigenous data," "Indigenous population statistics," and "Parliamentary review" (e.g., Docs 12-18, Docs 21-28), serve as progress updates on the design, testing, and implementation.
- Tracking implementation and comparisons (point 3): The inclusion of "Paper titled Analysis of Indigenous Enrolment" (undated, e.g., Doc 27, Doc 29) and "Indigenous Franchise KPIs using Centrelink data" suggests documents detailing how implementation was tracked and likely include comparisons to prior methodologies, supporting the increased enrolment rates noted in the FOI overview.
- Final summary reports on outcomes (point 4): A "Ministerial Submission... re Publication of Indigenous enrolment rates" (Doc 19, Feb 2019) and "Estimates Brief – Hot Issue... re Indigenous" (Doc 20, Feb 2019; Doc 40, Oct 2019) are listed, which serve as high-level summaries of the methodology's outcomes.
However, the schedule does not list any documents that explicitly address the following requested items, aligning with the FOI overview's findings:
- Number of Indigenous people involved (point 5): No document description indicates details on Indigenous involvement in the methodology's development.
- Final cost and dedicated AEC staffing (point 6): No document description indicates details on the financial costs or the number of AEC staffers specifically assigned to the task.
LS7888 documents [ZIP 35.12mb] (zip)
Download cached ZIP | Download from AECZIP Contents
- 7888-012.pdf (pdf)
- 7888-016.pdf (pdf)
- 7888-018.pdf (pdf)
- 7888-019.pdf (pdf)
- 7888-020.pdf (pdf)
- 7888-026.pdf (pdf)
- 7888-028.pdf (pdf)
- 7888-029.pdf (pdf)
- 7888-030.pdf (pdf)
- 7888-032.pdf (pdf)
- 7888-037.pdf (pdf)
- 7888-04.pdf (pdf)
- 7888-040.pdf (pdf)
- 7888-041.pdf (pdf)
- 7888-05.pdf (pdf)
- 7888-06.pdf (pdf)
- 7888-08.pdf (pdf)
7888-012.pdf (pdf)
Download file--- Page 1 --- Released under the Freedom of Information Act 1982 --- Page 2 --- Released under the Freedom of Information Act 1982 --- Page 3 --- Released under the Freedom of Information Act 1982 --- Page 4 --- Released under the Freedom of Information Act 1982 --- Page 5 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
This document consists of five pages. The first four pages are administrative boilerplate, each stating "Released under the Freedom of Information Act 1982". The fifth page details a list of "REDACTION CODES" (numbered 1-28), outlining various categories of information that may have been redacted from other documents (e.g., Personal Information, Business information, Legal Professional Communication, Deliberative material, Irrelevant material).
Relevance to FOI Request LS7888:
This document provides no substantive information directly related to FOI request LS7888's inquiries about the Australian Electoral Commission's (AEC) new methodology for its Indigenous Electoral Participation Program (IEPP). It does not contain details regarding the planning, implementation, outcomes, Indigenous involvement, or costs of the new methodology, nor does it provide data on enrolment rates or tracking efforts. Its content is purely administrative, serving as an introductory or explanatory piece regarding the FOI release process itself and the types of redactions applied to other released materials. Therefore, this specific document does not address any of the core questions posed by the FOI request.
7888-016.pdf (pdf)
Download file--- Page 1 --- Owen Jones From: Sent: To: Cc: Subject: Guy Lewis Tuesday, 4 December 2018 1:32 PM Delyse Paliaga Michael Roden; Stephanie Maxwell Indigenous missing [SEC=UNCLASSIFIED] Delyse, just a note the 42% Indigenous missing. The Indigenous Electoral Participation Program 2016‐17 National Program Plan, 22 August 2016 (page 4) states that internal AEC research suggests around 58 per cent are enrolled to vote, on enrolment (2011) and AEC participation statistics in June 2014. . This data was sourced from Internal agency research conducted by SRA based on ABS Indigenous population data This aligns with the Snowden statement in Hansard ‘In addition, in 2014 the AEC had already identified that 42% of Indigenous people were not enrolled to vote’. Regards Guy Lewis | Asst. Director Roll Program Design Roll Program and Community Engagement Section Roll Management and Community Engagement Branch Australian Electoral Commission T: Int: 21312 | Ext: (02) 6271 4758 | M: 1 LS7888 Released Document No. 16.71919Released under the Freedom of Information Act 1982 --- Page 2 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
This internal AEC email, dated 4 December 2018, from Guy Lewis to Delyse Paliaga, clarifies the historical baseline for Indigenous electoral enrolment rates. It references the 2016-17 Indigenous Electoral Participation Program (IEPP) National Program Plan, stating that internal AEC research (based on 2011 ABS data and 2014 AEC statistics) estimated approximately 58% of Indigenous Australians were enrolled to vote, meaning 42% were "missing." This figure is corroborated by a 2014 Hansard statement.
Relevance to FOI Request LS7888:
This document is relevant as it establishes the historical context and the "old 2014 methodology" baseline for Indigenous enrolment rates (58% enrolled, or 42% unenrolled) that the FOI request seeks to contrast with the AEC's new methodology. While not detailing the new methodology itself, it provides the prior understanding of Indigenous enrolment that the new approach, with its significantly higher estimated rates (76.4% in June 2018, 76.6% in June 2019), was intended to update and improve upon.
7888-018.pdf (pdf)
Download file--- Page 1 --- Owen Jones From: Sent: To: Subject: Guy Lewis Wednesday, 5 December 2018 8:43 AM Michael Roden RE: Parliamentary review 3 December 2018 [SEC=UNCLASSIFIED] Yes I found it cited in the 2015 ANAO report Guy From: Michael Roden Sent: Wednesday, 5 December 2018 8:20 AM To: Guy Lewis Subject: RE: Parliamentary review 3 December 2018 [SEC=UNCLASSIFIED] Guy, I think that reflects the first cut of the Indigenous enrolment rates we tentatively trialled 4 (?) years ago, solely using regression methods. Many methodological changes since then, plus a general improvement in enrolment rates. Cheers m. From: Guy Lewis <Guy.Lewis@aec.gov.au> Sent: Tuesday, 4 December 2018 9:17 AM To: Michael Roden <Michael.Roden@aec.gov.au> Subject: FW: Parliamentary review 3 December 2018 [SEC=UNCLASSIFIED] Michael, are you able to shed light on the following statement by Warren Snowden (Lingiari) in parliament yesterday? Do you recall where this statement might have come from? ‘in 2014 the AEC had already identified that 42 per cent of Indigenous people were not enrolled to vote.’ Regards guy From: Delyse Paliaga <Delyse.Paliaga@aec.gov.au> Sent: Tuesday, 4 December 2018 9:12 AM To: Guy Lewis <Guy.Lewis@aec.gov.au>; Darryl Miller <darryl.miller@aec.gov.au> Cc: Jaime Garrido <Jaime.Garrido@aec.gov.au>; Sarah Gowty <Sarah.Gowty@aec.gov.au>; Robyn Legg <Robyn.Legg@aec.gov.au> Subject: FW: Parliamentary review 3 December 2018 [SEC=UNCLASSIFIED] Hi Guy and Darryl Guy are you able to check the figures re enrolment that Mr Snowden quotes in his speech below? Darryl do you have any comments on Mr Snowdens speech. Delyse Delyse Paliaga | Director Roll Program and Community Engagement Roll Management & Community Engagement Branch Australian Electoral Commission 1 LS7888 Released Document No. 18Released under the Freedom of Information Act 1982 --- Page 2 --- Released under the Freedom of Information Act 1982 --- Page 3 --- Released under the Freedom of Information Act 1982 --- Page 4 --- Released under the Freedom of Information Act 1982 --- Page 5 --- Released under the Freedom of Information Act 1982 --- Page 6 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Document Summary
This document is an email exchange from December 2018 between several Australian Electoral Commission (AEC) staff members, including Guy Lewis, Michael Roden, and Delyse Paliaga. The core discussion revolves around a parliamentary statement made by MP Warren Snowden on 3 December 2018, claiming that "in 2014 the AEC had already identified that 42 per cent of Indigenous people were not enrolled to vote" (implying 58% enrolment).
- Michael Roden confirms that this 2014 figure likely represents the "first cut" of Indigenous enrolment rates, which was tentatively trialled around four years prior (circa 2014) and "solely using regression methods."
- He explicitly states that "Many methodological changes since then" have occurred, leading to a "general improvement in enrolment rates."
- Guy Lewis confirms that this 2014 figure was cited in a 2015 Australian National Audit Office (ANAO) report.
- The exchange highlights the AEC's internal discussion and clarification regarding historical Indigenous enrolment rate estimates derived from an earlier methodology.
Relevance to FOI Request LS7888
This document is highly relevant to FOI request LS7888, which sought information on the AEC's new methodology for the Indigenous Electoral Participation Program (IEPP) in 2018-2019.
- Context for Methodological Shift: The FOI overview states that the AEC "transitioned to a more direct method for estimating Indigenous electoral enrolment rates," increasing estimates from "approximately 58% (under the old 2014 methodology)" to over 76% by 2018-2019. This document directly corroborates and provides internal AEC confirmation for the "old 2014 methodology" figure of 58% enrolment (42% unenrolled).
- Acknowledgement of Methodological Evolution: Michael Roden's comment about "many methodological changes since then" explicitly acknowledges the evolution of the AEC's approach to estimating Indigenous enrolment rates, which is central to the FOI request's focus on the "new methodology."
- Historical Data Point: By confirming the source and nature of the 2014 enrolment estimate (regression-based and cited in a 2015 ANAO report), the document provides crucial historical context against which the impact and outcomes of the new methodology (Centrelink data-driven) can be understood.
- While it does not detail the planning, implementation, Indigenous involvement, or specific costs of the new methodology, it solidifies the understanding of the baseline from which the AEC began its improvements in Indigenous electoral participation estimation.
7888-019.pdf (pdf)
Download file--- Page 1 --- LS7888 Released Document No. 1945 NOTEDReleased under the Freedom of Information Act 1982 --- Page 2 --- LS7888 Released Document No. 19444Released under the Freedom of Information Act 1982 --- Page 3 --- LS7888 Released Document No. 19Released under the Freedom of Information Act 1982 --- Page 4 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Summary of Document LS7888 Released Document No. 19/19444/1945
This document, identified as part of FOI release LS7888, primarily consists of administrative information pertaining to the Freedom of Information Act 1982. The initial pages display standard FOI release numbering and disclaimers. The substantive content is a list of "REDACTION CODES" (Page 4), which enumerates and defines various categories of information that may be withheld or redacted from documents released under the FOI Act, such as personal information, business information, deliberative material, and electoral roll data. The document itself contains no direct content, data, or discussion related to the Australian Electoral Commission's (AEC) Indigenous Electoral Participation Program (IEPP) new methodology.
Relevance to FOI Request LS7888
This document's relevance to FOI request LS7888 is purely administrative. It does not contain any direct information or detail concerning the AEC's new methodology for its Indigenous Electoral Participation Program (IEPP), its planning, implementation, outcomes, Indigenous involvement, or costs, which are the core subjects of the request. Instead, it serves as a key or legend for understanding the types of information that may have been redacted from other documents released as part of the broader LS7888 FOI disclosure. While it confirms its inclusion in the overall FOI release, it provides no substantive contribution to addressing the specific queries regarding the IEPP methodology.
7888-020.pdf (pdf)
Download file--- Page 1 --- ESTIMATES BRIEF - Hot Issue Additional Estimates – February 2019 INDIGENOUS ENROLMENT RATE ESTIMATES Subject/Issue The Australian Electoral Commission (AEC) has prepared Indigenous enrolment rate estimates that show an overall increase on previous estimates. Key points AEC Indigenous enrolment rate estimates, show an overall increase on previous published data and shows that the Indigenous enrolment rate has grown at a faster pace than that of the Roll as a whole. As at 30 June 2018 the AEC estimates that the national Indigenous enrolment was 76.4 per cent. This new estimate based on more robust data reveals that the estimated number of Indigenous Australians of voting age, who are enrolled continues to rise. In the 12 months from 30 June 2017 to 30 June 2018 the AEC estimates there has been: o a 5.1 per cent increase in the estimated number of enrolled Indigenous Australians as a proportion of the estimated Indigenous population, rising from 364,631 to 383,078; and o a corresponding 4 per cent fall in the estimated number of non-enrolled Indigenous Australians from 123,517 to 118,581. In the same 12 month period the estimated Indigenous voting age population only increased by 2.8 per cent (13,511 electors) from 488,148 to 501,659. Any gap between non-Indigenous and Indigenous enrolment rates is disappointing, however it is encouraging to see that Indigenous enrolment rates have improved considerably since the last time estimates were prepared. This new estimates will further inform the strategies that the AEC is delivering to further increase electoral participation by Aboriginal and Torres Strait Islander people. SB19-000015 1 LS7888 Released Document No. 20 --- Page 2 --- Background Indigenous enrolment rate estimates Previous Indigenous enrolment rate estimates of 58 per cent were based on the best information we had at the time. These new figures are based on more recent and accurate data. Specifically, the AEC has been provided with self-identified Indigeneity data by the Department of Human Services, and with population data by the Australian Bureau of Statistics. The AEC will be using this improved methodology from here on, and we’ll be updating the data every year as part of our annual roll-integrity review. Indigenous enrolment rate has grown at a faster pace than that of the Roll as a whole. The table below shows the estimated Indigenous enrolment rates at 30 June 2017 and 30 June 2018 by state and territory. A comparison column shows the rate of improvement across all states/territories and nationally. State Enrolment Rate as at 30 June 2017 Enrolment Rate as at 30 June 2018 Change in estimated Indigenous enrolled 30 June 2017 to 30 June 2018 NSW VIC QLD WA SA TAS ACT NT National 85.9% 73.6% 70.0% 62.8% 67.9% 83.3% 76.2% 67.1% 74.7% 87.3% 76.0% 72.5% 63.8% 70.2% 84.1% 79.4% 67.5% 76.4% +1.4% +2.4% +2.5% +1.0% +2.3% +0.8% +3.2% +0.4% +1.7% State of the electoral roll As at 31 December 2018, the estimated enrolment rate of the eligible population was 96.2 per cent, which is 1.2 percentage points higher than on election day for the 2016 federal election. An estimated 95 per cent of the eligible population was enrolled to vote at the 2016 federal election (compared with 92.4 per cent at the 2013 election). Date: Cleared by (SES): Telephone No: Branch: Contact Officer: Telephone No: Consultation: PDR Number: 15/02/2019 Tom Rogers 02 6271 4506 Roll Management and Community Engagement Robyn Legg (02) 6271 4786 Nil SB19-000015 SB19-000015 2 LS7888 Released Document No. 20
This document is an "Estimates Brief" from February 2019, providing updated Indigenous enrolment rate estimates.
Summary of Document
The document, an "Estimates Brief" prepared for Additional Estimates in February 2019, details the Australian Electoral Commission's (AEC) updated methodology and findings regarding Indigenous electoral enrolment rates. Key points include:
- Increased Enrolment: The national Indigenous enrolment rate was estimated at 76.4% as of 30 June 2018, a significant increase from previous estimates of 58%. This indicates a faster growth rate for Indigenous enrolment compared to the overall electoral roll.
- New Methodology: The revised estimates are based on an "improved methodology" utilising more robust data, specifically self-identified Indigeneity data provided by the Department of Human Services (Centrelink) and population data from the Australian Bureau of Statistics (ABS). This new methodology will be updated annually.
- Positive Trends (2017-2018): Between 30 June 2017 and 30 June 2018, the estimated number of enrolled Indigenous Australians increased by 5.1% (from 364,631 to 383,078), while the estimated number of non-enrolled Indigenous Australians decreased by 4% (from 123,517 to 118,581).
- Strategic Use: The AEC states that these new estimates will inform future strategies aimed at further increasing electoral participation by Aboriginal and Torres Strait Islander people.
- State-by-State Data: The document includes a table showing Indigenous enrolment rates for each state and territory as of June 2017 and June 2018, demonstrating a positive change across all jurisdictions.
Relevance to FOI Request LS7888
This document is highly relevant to FOI request LS7888 as it directly addresses several core aspects:
- New Methodology: It explicitly details the "new methodology" for estimating Indigenous enrolment rates, confirming the shift from the 2014 methodology (58% estimate) and the utilisation of Centrelink's self-identified Indigeneity data, a key finding of the FOI request.
- Implementation & Outcomes: The brief serves as a formal communication of the implementation of this new methodology and its immediate "outcomes," presenting the significantly increased estimated enrolment rate of 76.4% by June 2018, which aligns precisely with the figures highlighted in the FOI overview.
- Planning & Context: Although not detailing the planning process itself, the document is an output of that planning, reflecting the AEC's internal reporting on the results of its updated approach to Indigenous electoral participation.
- Timeframe: Dated February 2019, the document falls squarely within the 2018-2019 period specified by the FOI request.
- Program Information: It states that these new estimates will "further inform the strategies that the AEC is delivering to further increase electoral participation by Aboriginal and Torres Strait Islander people," directly linking the methodology's results to the ongoing Indigenous Electoral Participation Program (IEPP).
7888-026.pdf (pdf)
Download file--- Page 1 --- Analysis of Indigenous Enrolment Population Estimates The ABS data provided at table 1 represents the ABS estimated resident Indigenous population 1=8 years and over by state as at 2016 and projected to 30 June 2019. This data has not been adjusted by the ABS to calculate estimated Indigenous enrolment eligible population, however, table 2 provides the AEC’s estimated Indigenous voting age population as at 30 June 2018. Table 1 – ABS estimated resident and projected population - Indigenous Australians - 18 years and over Estimated resident and projected population, Aboriginal and Torres Strait Islander Australians, Series B, 18 years and over - 30 June 2016 to 2019 (see note a) State/territory Estimated Resident Population Projected Population 2016 (see note b) 2017 2018 2019 ACT NSW NT(b) QLD SA Tas Vic WA Australia (see note c) 4,670 156,172 48,180 127,492 24,991 16,991 34,289 60,733 473,710 4,886 160,403 49,282 131,166 25,719 17,471 35,457 62,544 487,121 5,089 164,510 50,464 135,141 26,507 17,986 36,645 64,451 500,988 5,273 168,896 51,674 139,306 27,207 18,489 37,839 66,335 515,215 Source: Australian Bureau of Statistics, Estimates and Projections, Aboriginal and Torres Strait Islander Australians, 2006 to 2031 (cat. No. 3238.0) (a) Projection Series B has been used for the period 2017-2019. (b) An age heaping adjustment was applied to derive final 30 June 2016 Aboriginal and Torres Strait Islander population estimates for the Northern Territory. This information was published in Estimates of Aboriginal and Torres Strait Islander Australians, June 2016 (cat. no. 3238.0.55.001) and has been used as the base population for compiling estimates and projections for this release. (c) Includes Other Territories. LS7888 Released Document No. 26Released under the Freedom of Information Act 1982 --- Page 2 --- Estimated Indigenous Enrolment Rates Table 2 - Estimated Indigenous Enrolment Rates Estimated Indigenous Enrolment Rates as at 30 June 2018 State Estimated Indigenous voting age population Estimated enrolled Estimated unenrolled Estimated enrolment rate NSW VIC QLD WA SA TAS ACT NT 164,845 36,538 135,642 64,545 26,522 18,004 5,000 50,563 143,891 27,775 98,357 41,182 18,622 15,150 3,971 34,130 National 501,659 383,078 20,954 8,763 37,285 23,363 7,900 2,854 1,029 16,433 118,581 87.3% 76.0% 72.5% 63.8% 70.2% 84.1% 79.4% 67.5% 76.4% As Australia’s federal electoral roll does not have a personal Indigenous identifier, estimating the enrolment rates for Indigenous persons is problematic, however, having this key franchise KPI for the Indigenous population is important for Closing the Gap purposes and the AECs focus on how the franchise could be better delivered. In 2014 the AEC’s Research Section commenced indirect estimation of these figures, using a regression model of the relationship between the Indigenous population and each of the three franchise KPIs at a small area level, such as polling place catchment or the ABS’ SA2 geography. While the models did not fit the data particularly well, the resulting Indigenous enrolment rates were statistically based on real data, rather than anecdote. Yet the resulting enrolment and turnout rates in particular were alarmingly low, and hopefully represented a worst case picture. As part of the Annual Roll Integrity Review (ARIR), the person-level matching between the Roll and Centrelink now includes Centrelink’s Indigenous indicator. However, as Centrelink customers represent only around half of the Roll, the matching (even if perfect) would be far from assigning all enrolments an Indigenous flag. Nevertheless, for the first time the AEC has a very large number of enrolled persons, over 6 million, with an explicit Indigenous status. The enrolment rate, even for all persons regardless of indigeneity, is difficult to calculate reliably due to conceptual, collection and methodological differences between the Roll (numerator) and the ABS- based enrolment-eligible population (denominator, “EEP”). In the case of the Indigenous population LS7888 Released Document No. 26Released under the Freedom of Information Act 1982 --- Page 3 --- these problems are magnified by having no Indigenous identifier on the Roll and less reliable estimates of Indigenous EEP. However, matching the Roll to Centrelink data allows a more direct method of estimating the Indigenous enrolment rate. It is still subject to quite courageous assumptions, but as fewer are required it likely represents an improved approach. In summary: For all Centrelink customers with an identified Indigenous or non-Indigenous status, calculate the proportion on the Roll. This is the raw enrolment rate of Centrelink clients. Calculate the ratio of the actual AEC enrolment rate versus the Centrelink-based rate, for the same point in time. This is the overall factor to adjust any Centrelink-based rates to align with AEC enrolment rates. Apply the adjustment factor to raw Centrelink-based enrolment rates by Indigenous status to give Indigenous and non-Indigenous enrolment rates. This process can be done at the State level, though results will have lower reliability. It should be noted that while this method avoids indirectly modelling Indigenous enrolment rates via geographic association, it does assume that any bias in Centrelink clients being enrolled is equal for Indigenous and non-Indigenous, and that there are no substantive Indigenous-status biases inherent in the Roll-to-Centrelink person matching. The Centrelink client file with Indigenous indicator newly available to the AEC has presented an excellent opportunity, via matching to the Roll and mark-off data, to improve estimation of Indigenous franchise KPIs. While the results are subject to a number of non-trivial assumptions, the more direct estimation methods lead to a higher likelihood of robust results. The figures should nevertheless still be regarded as indicative only. LS7888 Released Document No. 26Released under the Freedom of Information Act 1982 --- Page 4 --- Centrelink Data The number of Indigenous Australians that are identified both on the Centrelink file sent to the AEC in July 2018 and the electoral Roll does not represent all Indigenous enrolments. It is limited to those Indigenous Australians who both receive a Centrelink benefit and who have self-identified as Indigenous to Centrelink. After considerable data cleansing and analysis the AEC identified 72,451 potentially unenrolled Indigenous Australians. This represents around 61 per cent of the 118,581 Indigenous Australians the AEC estimate were unenrolled as at 30 June 2018 (see table 2 above). The enrolment stimulation event conducted in February/March 2019 resulted in the enrolment of 2,646 Indigenous Australians, reducing the estimated number of unenrolled Indigenous Australians identified in the Centrelink dataset to 69,143. Table 3 sets out the distribution of this cohort across electoral divisions. It is important to note that the potential unenrolled under 65 Indigenous Australians as at the 2019 close of rolls represents those people that are on the Centrelink file sent to the AEC in July 2018 and who were included in the February 2019 enrolment stimulation event but had not enrolled as at the 2019 close of rolls (18 April 2019). Analysis of the Centrelink potentially unenrolled Indigenous Australians 65 years and older identified significant matching error, due to the quality of data input and the nature of the residential status of this cohort. Work will be undertaken on the 2018-19 dataset to better understand this phenomenon to facilitate better targeting of potentially unenrolled indigenous Australians in the future. Table 3 - Enrolled and potentially unenrolled Indigenous Australians identified on the 2017-18 Centrelink file by electoral division as at the 2019 close of rolls Division Potential Unenrolled under 65 Indigenous Australians as at the 2019 close of rolls Number of Indigenous Australians in the Centrelink file and enrolled as at 30 June 2018 BEAN CANBERRA FENNER BANKS BARTON BENNELONG BEROWRA BLAXLAND BRADFIELD CALARE CHIFLEY COOK COWPER CUNNINGHAM DOBELL 168 157 145 50 63 23 14 52 4 494 439 21 743 164 210 702 468 631 365 415 206 133 364 74 3,676 2,838 307 4,382 1,229 2,386 LS7888 Released Document No. 26Released under the Freedom of Information Act 1982 --- Page 5 --- Division Potential Unenrolled under 65 Indigenous Australians as at the 2019 close of rolls Number of Indigenous Australians in the Centrelink file and enrolled as at 30 June 2018 EDEN-MONARO FARRER FOWLER GILMORE GRAYNDLER GREENWAY HUGHES HUME HUNTER KINGSFORD SMITH LINDSAY LYNE MACARTHUR MACKELLAR MACQUARIE MCMAHON MITCHELL NEW ENGLAND NEWCASTLE NORTHSYDNEY PAGE PARKES PARRAMATTA PATERSON REID RICHMOND RIVERINA ROBERTSON SHORTLAND SYDNEY WARRINGAH WATSON WENTWORTH WERRIWA WHITLAM LINGIARI SOLOMON BLAIR BONNER 308 733 87 439 160 102 19 91 253 153 257 387 218 14 100 68 11 805 234 13 837 2,162 103 247 27 328 451 123 159 376 4 41 11 177 197 5,483 1,395 898 224 1,658 2,641 532 2,839 659 710 370 1,133 3,017 1,010 2,159 3,088 1,911 160 1,182 567 155 5,527 2,033 101 3,856 10,143 493 2,955 208 1,961 3,337 1,308 1,831 1,231 120 307 124 1,077 1,832 16,473 3,659 2,331 832 LS7888 Released Document No. 26Released under the Freedom of Information Act 1982 --- Page 6 --- Division Potential Unenrolled under 65 Indigenous Australians as at the 2019 close of rolls Number of Indigenous Australians in the Centrelink file and enrolled as at 30 June 2018 BOWMAN BRISBANE CAPRICORNIA DAWSON DICKSON FADDEN FAIRFAX FISHER FLYNN FORDE GRIFFITH GROOM HERBERT HINKLER KENNEDY LEICHHARDT LILLEY LONGMAN MARANOA MCPHERSON MONCRIEFF MORETON OXLEY PETRIE RANKIN RYAN WIDE BAY WRIGHT ADELAIDE BARKER BOOTHBY GREY HINDMARSH KINGSTON MAKIN MAYO SPENCE STURT BASS 286 234 991 971 217 195 216 201 1,114 484 247 748 2,504 731 4,221 3,537 379 551 1,131 123 202 258 640 467 749 112 818 343 459 610 221 1,931 412 278 242 131 705 154 326 945 636 2,537 2,687 818 978 866 911 2,715 1,464 691 2,035 4,808 2,178 7,587 10,021 1,077 1,888 3,025 751 885 704 1,499 1,536 1,653 408 1,886 1,155 1,038 1,207 620 3,938 1,141 810 773 443 1,936 416 1,199 LS7888 Released Document No. 26Released under the Freedom of Information Act 1982 --- Page 7 --- Division Potential Unenrolled under 65 Indigenous Australians as at the 2019 close of rolls Number of Indigenous Australians in the Centrelink file and enrolled as at 30 June 2018 BRADDON CLARK FRANKLIN LYONS ASTON BALLARAT BENDIGO BRUCE CALWELL CASEY CHISHOLM COOPER CORANGAMITE CORIO DEAKIN DUNKLEY FLINDERS FRASER GELLIBRAND GIPPSLAND GOLDSTEIN GORTON HIGGINS HOLT HOTHAM INDI ISAACS JAGAJAGA KOOYONG LATROBE LALOR MACNAMARA MALLEE MARIBYRNONG MCEWEN MELBOURNE MENZIES MONASH NICHOLLS 364 279 255 399 29 179 176 104 116 75 21 225 56 186 39 120 62 77 74 489 23 103 25 61 20 184 26 72 9 25 75 68 662 61 66 115 10 157 576 2,491 964 1,231 1,521 172 625 761 303 361 327 125 571 302 576 187 404 393 251 295 1,121 75 393 126 301 118 717 141 253 69 256 446 222 1,631 206 391 264 73 497 1,570 LS7888 Released Document No. 26Released under the Freedom of Information Act 1982 --- Page 8 --- Division Potential Unenrolled under 65 Indigenous Australians as at the 2019 close of rolls Number of Indigenous Australians in the Centrelink file and enrolled as at 30 June 2018 SCULLIN WANNON WILLS BRAND BURT CANNING COWAN CURTIN DURACK FORREST FREMANTLE HASLUCK MOORE O'CONNOR PEARCE PERTH STIRLING SWAN TANGNEY NO FIXED ADDRESS TOTAL 105 251 63 430 675 332 347 131 6,885 424 444 583 72 1,684 505 343 380 602 139 1,166 69,805 415 764 291 1,371 1,718 1,025 953 339 9,241 1,274 1,012 1,420 303 3,260 1,437 730 886 1,484 330 224,554 Table 4 - Enrolled and potentially unenrolled Indigenous Australians identified on the 2017- 18 Centrelink file by electoral division State Potential Unenrolled under 65 Indigenous Australians as at the 2019 close of rolls Number of Indigenous Australians in the Centrelink file and enrolled as at 30 June 2018 ACT NSW NT Qld SA Tas Vic WA 470 11,972 6,878 23,792 5,143 1,623 4,123 13,976 1,801 78,610 20,132 61,507 12,322 7,406 15,993 26,783 LS7888 Released Document No. 26Released under the Freedom of Information Act 1982 --- Page 9 --- No Fixed Address Total 1,166 69,143 - 224,554 LS7888 Released Document No. 26Released under the Freedom of Information Act 1982
This document, titled "Analysis of Indigenous Enrolment Population Estimates," details the Australian Electoral Commission's (AEC) updated methodology for estimating Indigenous electoral enrolment rates and its initial outcomes.
Summary of Document Content:
The document outlines the AEC's shift from an indirect, statistically-based regression model (used in 2014, which yielded "alarmingly low" enrolment rates) to a "more direct method" for estimating Indigenous enrolment. This new methodology leverages self-identified Indigeneity data from Centrelink, matched against the electoral roll.
Key information provided includes:
* Estimated Indigenous Enrolment Rate: As at 30 June 2018, the estimated national Indigenous enrolment rate using this new methodology was 76.4%.
* Methodological Details: The document explains the process of using Centrelink data, involving calculating raw enrolment rates for Centrelink clients (both Indigenous and non-Indigenous), applying an adjustment factor derived from the overall AEC enrolment rate, and then applying this factor to Indigenous and non-Indigenous Centrelink-based rates. It acknowledges that the method involves "courageous assumptions" and potential biases, yet is considered an "improved approach."
* Identification of Unenrolled: Through the Centrelink data matching, the AEC identified 72,451 potentially unenrolled Indigenous Australians as of mid-2018 (representing approximately 61% of the total estimated unenrolled Indigenous population at that time).
* Enrolment Stimulation Outcomes: A subsequent enrolment stimulation event in February/March 2019 resulted in the enrolment of 2,646 Indigenous Australians, reducing the identified potentially unenrolled cohort to 69,143.
* Data Granularity: The document includes detailed tables (Table 3 and 4) providing a breakdown of potentially unenrolled Indigenous Australians (under 65) and Indigenous Australians identified on the Centrelink file who were enrolled, presented by electoral division and state as at the 2019 close of rolls.
* Limitations: It notes difficulties in calculating reliable enrolment rates due to the lack of an Indigenous identifier on the Roll and less reliable Indigenous population estimates, and specifically highlights issues with data matching for Indigenous Australians aged 65 and over.
Relevance to FOI Request LS7888:
This document is highly relevant to FOI request LS7888, which sought information on the AEC's new methodology for its Indigenous Electoral Participation Program (IEPP), particularly concerning its planning, implementation, and outcomes.
- New Methodology (Planning & Implementation): The document directly addresses the "new methodology" by detailing the strategic shift from the previous regression model to the Centrelink data-matching approach. It explains the core mechanics, assumptions, and improvements of this updated method, illustrating its "planning" (conceptual shift) and "implementation" (how the data is processed).
- Outcomes: It provides a key quantitative outcome of the new methodology: the significantly increased estimated Indigenous enrolment rate (76.4% by June 2018). It also details a direct outcome of leveraging this new data: the identification of a large cohort of potentially unenrolled Indigenous Australians and the results of a subsequent enrolment stimulation event (2,646 new enrolments), directly addressing the "outcomes" aspect of the FOI.
- Indigenous Involvement/Costs/Staffing: Consistent with the FOI overview, the document does not explicitly detail the number of Indigenous people involved in the methodology's development or its final costs and dedicated AEC staffing.
7888-028.pdf (pdf)
Download file--- Page 1 --- Owen Jones From: Sent: To: Subject: Attachments: Guy Lewis Monday, 12 August 2019 11:19 AM Stephanie Maxwell; Angelo Rojas FW: Indigenous population statistics [DLM=For-Official-Use-Only] Indigenous data for Min meeting Aug 2019.docx I have updated the Indigenous data we prepared last week as requested by Delyse. I would be grateful if you could both review asap and add comments as you wish. I need to get Tita’s sign off by cob today if possible regards Guy Lewis | Assistant Director Roll Program & Community Engagement Section | Roll Management & Community Engagement Branch Australian Electoral Commission T: (02) 6271 4758 X: 21312 M: From: Delyse Paliaga Sent: Monday, 12 August 2019 9:46 AM To: Guy Lewis Cc: Robyn Legg ; Julia Moscaritolo Subject: FW: Indigenous population statistics [DLM=For‐Official‐Use‐Only] Hi Guy Can you update this doc with Jeffs comments below. Can I see it first thing in the morning to give Robyn time to clear it. Thanks Delyse Delyse Paliaga | Director Roll Program & Community Engagement Section | Roll Management & Community Engagement Branch Australian Electoral Commission T: (02) 6271 4651 X: 21049 M: From: Robyn Legg <Robyn.Legg@aec.gov.au> Sent: Monday, 12 August 2019 9:03 AM To: Delyse Paliaga <Delyse.Paliaga@aec.gov.au>; Michael Lynch <Michael.Lynch@aec.gov.au> Cc: Julia Moscaritolo <Julia.Moscaritolo@aec.gov.au> Subject: FW: Indigenous population statistics [DLM=For‐Official‐Use‐Only] Hi Delyse Can your team please work with BIREI to amend as requested? I need to get it to Jeff tomorrow COB. Thanks 1 LS7888 Released Document No. 2877Released under the Freedom of Information Act 1982 --- Page 2 --- Robyn Legg | Assistant Commissioner Roll Management & Community Engagement Branch Executive | Roll Management & Community Engagement Branch Australian Electoral Commission T: (02) 6271 4786 X: 21473 M: From: Jeff Pope <Jeff.Pope@aec.gov.au> Sent: Monday, 12 August 2019 8:52 AM To: Robyn Legg <Robyn.Legg@aec.gov.au> Cc: Andrew Gately <Andrew.Gately@aec.gov.au>; Delyse Paliaga <Delyse.Paliaga@aec.gov.au>; Tom Rogers <Tom.Rogers@aec.gov.au>; Tim Courtney <Tim.Courtney@aec.gov.au>; Joanne Reid <Joanne.Reid2@aec.gov.au> Subject: RE: Indigenous population statistics [DLM=For‐Official‐Use‐Only] Hi Robyn The work you and your team are continuing to do on the analysis of indigenous enrolment is excellent and extremely valuable. As you can see from our discussion with the Assistant Minister on Thursday last week, this data is extremely important (and as you know we already have a request from the Finance Minister’s Office for this data). I would like to provide both Minister’s this data within the next 2‐3 business days but before doing so I think we need to do a couple of things with the documents produced: ‐ ‐ ‐ Firstly, and most critically, at the start of the document we need to outline the methodology used but in particular we need to emphasise that this data is based on estimates on top of estimates so they clearly understand that this data is not and cannot be considered 100% accurate. This document will need to be a stand alone document if we are going to provide it into the Office’s and whilst I can verbally convey the caveats I don’t want those to be lost over time so can we please embed a paragraph or two at the top of the document that provides that critical context (a fair bit of which you have in your email below). I think table 3 should precede table 2 because table 3 looks like the high level summary of table 2 and I think we need a few sentences that outlines the differences between the high level data in table 1 and the high level data in table 3 (which I am proposing becomes table 2) – which you also have in your email below. I also note there is a slight discrepancy in the numbers of unenrolled in tables 2 and 3 – of about 700 electors. We will need to reconcile that or explain it as that will be quickly picked up by the advisors. Grateful for an updated draft of the document by cob tomorrow please so I can look at providing to the Minister’s Offices. Regards JP Jeff Pope APM | Deputy Electoral Commissioner National Executive Australian Electoral Commission T: (02) 6271 4619 | M: 2 LS7888 Released Document No. 2877Released under the Freedom of Information Act 1982 --- Page 3 --- From: Robyn Legg <Robyn.Legg@aec.gov.au> Sent: Tuesday, 6 August 2019 1:47 PM To: Jeff Pope <Jeff.Pope@aec.gov.au> Cc: Andrew Gately <Andrew.Gately@aec.gov.au>; Delyse Paliaga <Delyse.Paliaga@aec.gov.au> Subject: RE: Indigenous population statistics [SEC=UNCLASSIFIED] Hi Jeff In the attachment we have provided: Table 1 ‐ our published estimated Indigenous Enrolment rates based on self‐identified Indigenous people on Centrelink data as compared to the Roll and taken as a percentage of estimated Indigenous eligible population Table 2 – the Divisional breakdown of the 72K or so Indigenous people identified from that original Centrelink data who were contacted as part of the Feb stimulation event (data was cleansed to remove over 65s and to removed those who we thought were on the roll – minimising mis‐matches) and excludes those who enrolled before the federal election (ie. those identified as unenrolled). It also details the number of people that we were able to match from the Centrelink data (those identified as enrolled). Table 3 ‐ the same data as above by State. Table 4 ‐ ABS estimated Indigenous population, over 18yo – not eligible population. This is not available broken down any further. I preface all of it by saying that only Table 1 and the ABS data at Table 4 are published externally. While there are multiple caveats due the nature of the data it certainly gives us much to work on. Aside from the ABS data, the rest is all based on June 18 so getting older now (although not irrelevant) and we are working on the new data sets now. We have done some internal breakdowns for IEPP officers but we wouldn’t feel comfortable sharing this as the further we break it down, the greater risk of error. They remain indicative and we have had to make some assumptions from the data. As I mentioned in my previous email we are working through the latest June 2019 Centrelink data with BIREI at the moment and as this will be our second time around, hope to have increasing confidence in our estimates, albeit still limited by some of the quality issues with their data. Robyn Legg | Assistant Commissioner Roll Management & Community Engagement Branch Executive | Roll Management & Community Engagement Branch Australian Electoral Commission T: (02) 6271 4786 X: 21473 M: From: Jeff Pope <Jeff.Pope@aec.gov.au> Sent: Monday, 5 August 2019 9:28 AM To: Andrew Gately <Andrew.Gately@aec.gov.au>; Robyn Legg <Robyn.Legg@aec.gov.au> Cc: Tom Rogers <Tom.Rogers@aec.gov.au> Subject: FW: Indigenous population statistics [SEC=UNCLASSIFIED] UNCLASSIFIED Hi Andrew and Robyn 3 LS7888 Released Document No. 287Released under the Freedom of Information Act 1982 --- Page 4 --- Can you please advise how long it will take for us to pull together our data estimates to respond to the questions below from the Minister’s Office. I am assuming we have something close to ready (noting our data limitations on this issue) as part of our preparation for the Assistant Minister’s visit on Thursday. I would like to get something to Daniel within the next 48 hours. Regards JP Sent with BlackBerry Work (www.blackberry.com) UNCLASSIFIED From: Clode, Daniel <Daniel.Clode@finance.gov.au> Date: Monday, 05 Aug 2019, 09:21 To: Jeff Pope <Jeff.Pope@aec.gov.au> Subject: Indigenous population statistics [SEC=UNCLASSIFIED] UNCLASSIFIED Hi Jeff Could I please get some statistics on Indigenous population distributions, insofar as this data is held by the AEC? For instance Indigenous enrolment by electorate, and/or Indigenous enrolment by suburb/ town? I also wonder what data we have on under‐enrolment of Indigenous Australians by location? For instance the AEC certainly has the ABS estimate of the Census undercount in the NT, but I wonder if there is similar data for other locations. Thanks Daniel UNCLASSIFIED ________________________________ Finance Australian Business Number (ABN): 61 970 632 495 Finance Web Site: www.finance.gov.au IMPORTANT: This transmission is intended only for the use of the addressee and may contain confidential or legally privileged information. If you are not the intended recipient, you are notified that any use or dissemination of this communication is strictly prohibited. If you have received this transmission in error, please notify us immediately by telephone on 61-2-6215-2222 and delete all copies of this transmission together with any attachments. If responding to this email, please send to the appropriate person using the suffix .gov.au. ________________________________ 4 LS7888 Released Document No. 28Released under the Freedom of Information Act 1982 --- Page 5 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Document Summary
This document is an email chain from August 2019 among senior Australian Electoral Commission (AEC) staff, including the Deputy Electoral Commissioner, Assistant Commissioner, and directors/assistant directors of the Roll Management & Community Engagement Branch. The correspondence centers on the preparation and refinement of an "Indigenous data" report ("Indigenous population statistics") for presentation to Australian Government Ministers (Assistant Minister, Finance Minister).
Key points from the document include:
* Data Content: The report compiles various Indigenous electoral statistics, including:
* Published estimated Indigenous enrolment rates based on self-identified Indigenous people from Centrelink data compared to the Electoral Roll (Table 1).
* Divisional and state breakdowns of approximately 72,000 Indigenous people identified from Centrelink data (June 2018) who were contacted as part of a "Feb stimulation event" (Tables 2 & 3). This data was cleansed to remove certain groups (e.g., over 65s, those already on the roll).
* Australian Bureau of Statistics (ABS) estimated Indigenous population (over 18 years old) (Table 4).
* Methodology and Caveats: Senior staff emphasize the need to explicitly state the methodology used and stress that the data is "based on estimates on top of estimates" and "cannot be considered 100% accurate" due to the nature and quality of the underlying data (Centrelink data quality issues, further breakdown leading to greater error risk).
* Ministerial Interest: There is high-level ministerial interest in this data, with requests from both the Assistant Minister and the Finance Minister's Office.
* Ongoing Work: The AEC was actively working on incorporating the latest June 2019 Centrelink data with their Business Intelligence, Research and Electoral Integrity (BIREI) unit to increase confidence in their estimates.
Relevance to FOI Request LS7888
This document is highly relevant to FOI request LS7888 as it provides direct insight into the AEC's new methodology for its Indigenous Electoral Participation Program (IEPP) and associated data.
* New Methodology & Data Use: The document explicitly confirms the AEC's reliance on "self-identified Indigenous people on Centrelink data" to estimate Indigenous enrolment rates and identify potentially unenrolled Indigenous Australians. This directly aligns with the FOI request's overview stating the AEC's transition to a more direct method using Centrelink data.
* Implementation & Outcomes: The discussion of the "Feb stimulation event" where ~72,000 Indigenous people were identified and contacted for enrolment stimulation demonstrates a concrete implementation activity and its associated outcomes (e.g., numbers identified). The document details how the outcomes of the new methodology (estimated enrolment rates, figures of unenrolled Indigenous Australians) are being tracked and communicated internally and to ministers.
* Planning & Refinement: The email chain illustrates the internal planning and refinement process for communicating the results of the new methodology, including the critical discussion around presenting the data with appropriate caveats regarding its estimation basis.
* Consistency with FOI Findings: The mention of "over 72,000 potentially unenrolled Indigenous Australians" in mid-2018 (or identified from June 2018 data) directly supports and adds detail to the FOI overview's finding about identifying such a group.
* Gaps in FOI Request: The document, consistent with the FOI request's noted gaps, does not explicitly detail the number of Indigenous people involved in the methodology's development or its final costs and dedicated AEC staffing.
7888-029.pdf (pdf)
Download file--- Page 1 --- Analysis of Indigenous Enrolment Population estimates Table 1 represents the ABS estimated resident Indigenous population 18 years and over by state as at 2016 and projected to 30 June 2019. The projected population for Indigenous Australians has been recently updated and adjusted by the ABS. Table 2 provides the AEC’s estimated Indigenous voting age population as at 30 June 2018, before the more recent ABS Indigenous populations adjustments for 2018. Table 1 – ABS estimated resident and projected population – Indigenous Australians – 18 years and over State/territory Estimated Resident Population Projected Population (a) 2016 (b) 2017 2018 2019 NSW VIC QLD WA SA TAS ACT NT (b) Australia (c) 156,172 34,289 127,492 60,733 24,991 16,991 4,670 48,180 160,403 35,457 131,166 62,544 25,719 17,471 4,886 49,282 164,510 36,645 135,141 64,451 26,507 17,986 5,089 50,464 168,896 37,839 139,306 66,335 27,207 18,489 5,273 51,674 473,710 487,121 500,988 515,215 Source: Australian Bureau of Statistics, Estimates and Projections, Aboriginal and Torres Strait Islander Australians, 2006 to 2031 (cat. No. 3238.0) (a) Projection Series B has been used for the period 2017-2019. (b) An age heaping adjustment was applied to derive final 30 June 2016 Aboriginal and Torres Strait Islander population estimates for the Northern Territory. This information was published in Estimates of Aboriginal and Torres Strait Islander Australians, June 2016 (cat. no. 3238.0.55.001) and has been used as the base population for compiling estimates and projections for this release. (c) Includes Other Territories. Indigenous enrolment rate estimates As Australia’s federal electoral roll does not have a personal Indigenous identifier, estimating the enrolment rates for Indigenous persons is problematic, however, having this key franchise KPI for the Indigenous population is important for Closing the Gap purposes and the AECs focus on how the franchise could be better delivered. In 2014 the AEC’s Research Section commenced indirect estimation of these figures, using a regression model of the relationship between the Indigenous population and each of the three franchise KPIs at a small area level, such as polling place catchment or the ABS’ SA2 geography. LS7888 Released Document No. 29Released under the Freedom of Information Act 1982 --- Page 2 --- While the models did not fit the data particularly well, the resulting Indigenous enrolment rates were statistically based on real data, rather than anecdote. Yet the resulting enrolment and turnout rates in particular were alarmingly low, and hopefully represented a worst case picture. Current method The Centrelink client file with Indigenous indicator available to the AEC annually since 2017 has presented an excellent opportunity, via matching to the Roll and mark-off data, to improve estimation of Indigenous franchise KPIs. While the results are subject to a number of non-trivial assumptions, the more direct estimation methods lead to a higher likelihood of robust results. The figures should nevertheless still be regarded as indicative only. As part of the Annual Roll Integrity Review (ARIR), the person-level matching between the Roll and Centrelink now includes Centrelink’s Indigenous indicator. However, as Centrelink customers represent only around half of the Roll, the matching (even if perfect) would be far from assigning all enrolments an Indigenous flag. Nevertheless, for the first time the AEC has a very large number of enrolled persons, over 6 million, matched to Centrelink records with an explicit Indigenous status. The enrolment rate, even for all persons regardless of indigeneity, is difficult to calculate reliably due to conceptual, collection and methodological differences between the Roll (numerator) and the ABS-based enrolment-eligible population (denominator, “EEP”). In the case of the Indigenous population these problems are magnified by having no Indigenous identifier on the Roll and less reliable estimates of Indigenous EEP. However, matching the Roll to Centrelink data allows a more direct method of estimating the Indigenous enrolment rate. Results for 30 June 2018 are provided in Table 2. Table 2 - Estimated Indigenous Enrolment Rates, 30 June 2018 State Estimated Indigenous voting age population Estimated enrolled Estimated not enrolled Estimated enrolment rate NSW VIC QLD WA SA TAS ACT NT 164,845 36,538 135,642 64,545 26,522 18,004 5,000 50,563 143,891 27,775 98,357 41,182 18,622 15,150 3,971 34,130 Australia 501,659 383,078 20,954 8,763 37,285 23,363 7,900 2,854 1,029 16,433 118,581 87.3% 76.0% 72.5% 63.8% 70.2% 84.1% 79.4% 67.5% 76.4% It is still subject to quite courageous assumptions, but as fewer are required it likely represents an improved approach. In summary: For all Centrelink customers with an identified Indigenous or non-Indigenous status, calculate the proportion on the Roll. This is the raw enrolment rate of Centrelink clients. Calculate the ratio of the actual AEC enrolment rate versus the Centrelink-based rate, for the same point in time. This is the overall factor to adjust any Centrelink-based rates to align with AEC enrolment rates. Apply the adjustment factor to raw Centrelink-based enrolment rates by Indigenous status to give Indigenous and non-Indigenous enrolment rates. LS7888 Released Document No. 29Released under the Freedom of Information Act 1982 --- Page 3 --- This process can be done at the State level, though results will have lower reliability. It should be noted that while this method avoids indirectly modelling Indigenous enrolment rates via geographic association, it does assume that any bias in Centrelink clients being enrolled is equal for Indigenous and non-Indigenous, and that there are no substantive Indigenous-status biases inherent in the Roll-to-Centrelink person matching. Centrelink data The number of Indigenous Australians that are identified both on the Centrelink file sent to the AEC in July 2018 and the electoral Roll does not represent all Indigenous enrolments. It is limited to those Indigenous Australians who both receive a Centrelink benefit and who have self-identified as Indigenous to Centrelink. After considerable data cleansing and analysis the AEC identified 72,451 potentially unenrolled Indigenous Australians. This represents around 61 per cent of the 118,581 Indigenous Australians the AEC estimate were not enrolled as at 30 June 2018 (see Table 2 above). The enrolment stimulation event conducted in February/March 2019 using 2018 Centrelink data resulted in the enrolment of 2,646 people who indicated they were Indigenous. This reduced the estimated number of unenrolled Indigenous Australians identified in the Centrelink dataset to 69,143. Table 3 sets out the distribution of this cohort across states and Table 4 provides the data for electoral divisions. It is important to note that the potential unenrolled under 65 Indigenous Australians as at the 2019 close of rolls represents those people that are on the Centrelink file sent to the AEC in July 2018 and who were included in the February 2019 enrolment stimulation event but had not enrolled as at the 2019 close of rolls (18 April 2019). Analysis of the Centrelink potentially unenrolled Indigenous Australians 65 years and older identified significant matching error, due to the quality of data input and the nature of the residential status of this cohort. Work will be undertaken on the 2019 dataset to better understand this phenomenon to facilitate better targeting of potentially unenrolled indigenous Australians in the future. Table 3 – Indigenous Australians identified on the 2017-18 Centrelink file by state and territory – enrolled as at 30 June 2018 and potentially not enrolled as at the 2019 close of rolls State Number of Indigenous Australians in the Centrelink file and enrolled as at 30 June 2018 Potential not enrolledunder 65 Indigenous Australians as at the 2019 close of rolls NSW VIC QLD WA SA TAS ACT NT No fixed address Total 78,610 15,993 61,507 26,783 12,322 7,406 1,801 20,132 - 224,554 11,972 4,785 23,792 13,976 5,143 1,623 470 6,878 1,166 69,805 LS7888 Released Document No. 29Released under the Freedom of Information Act 1982 --- Page 4 --- Table 4 – Indigenous Australians identified on the 2017-18 Centrelink file by electoral division – enrolled as at 30 June 2018 and potentially not enrolled as at the 2019 close of rolls State Division Indigenous Australians in the Centrelink file and enrolled as at 30 June 2018 Potential not enrolled under 65 Indigenous Australians as at the 2019 close of rolls NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW NSW Banks Barton Bennelong Berowra Blaxland Bradfield Calare Chifley Cook Cowper Cunningham Dobell Eden-Monaro Farrer Fowler Gilmore Grayndler Greenway Hughes Hume Hunter Kingsford Smith Lindsay Lyne Macarthur Mackellar Macquarie McMahon Mitchell New England Newcastle North Sydney Page Parkes Parramatta Paterson Reid Richmond Riverina Robertson Shortland Sydney 365 415 206 133 364 74 3,676 2,838 307 4,382 1,229 2,386 1,658 2,641 532 2,839 659 710 370 1,133 3,017 1,010 2,159 3,088 1,911 160 1,182 567 155 5,527 2,033 101 3,856 10,143 493 2,955 208 1,961 3,337 1,308 1,831 1,231 50 63 23 14 52 4 494 439 21 743 164 210 308 733 87 439 160 102 19 91 253 153 257 387 218 14 100 68 11 805 234 13 837 2,162 103 247 27 328 451 123 159 376 LS7888 Released Document No. 29Released under the Freedom of Information Act 1982 --- Page 5 --- State Division Indigenous Australians in the Centrelink file and enrolled as at 30 June 2018 Potential not enrolled under 65 Indigenous Australians as at the 2019 close of rolls NSW NSW NSW NSW NSW VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC VIC QLD QLD Warringah Watson Wentworth Werriwa Whitlam Aston Ballarat Bendigo Bruce Calwell Casey Chisholm Cooper Corangamite Corio Deakin Dunkley Flinders Fraser Gellibrand Gippsland Goldstein Gorton Higgins Holt Hotham Indi Isaacs Jagajaga Kooyong Latrobe Lalor Macnamara Mallee Maribyrnong McEwen Melbourne Menzies Monash Nicholls Scullin Wannon Wills Blair Bonner 120 307 124 1,077 1,832 172 625 761 303 361 327 125 571 302 576 187 404 393 251 295 1,121 75 393 126 301 118 717 141 253 69 256 446 222 1,631 206 391 264 73 497 1,570 415 764 291 2,331 832 4 41 11 177 197 29 179 176 104 116 75 21 225 56 186 39 120 62 77 74 489 23 103 25 61 20 184 26 72 9 25 75 68 662 61 66 115 10 157 576 105 251 63 898 224 LS7888 Released Document No. 29Released under the Freedom of Information Act 1982 --- Page 6 --- State Division Indigenous Australians in the Centrelink file and enrolled as at 30 June 2018 Potential not enrolled under 65 Indigenous Australians as at the 2019 close of rolls QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD QLD WA WA WA WA WA WA WA WA WA WA WA WA WA WA WA WA SA Bowman Brisbane Capricornia Dawson Dickson Fadden Fairfax Fisher Flynn Forde Griffith Groom Herbert Hinkler Kennedy Leichhardt Lilley Longman Maranoa McPherson Moncrieff Moreton Oxley Petrie Rankin Ryan Wide Bay Wright Brand Burt Canning Cowan Curtin Durack Forrest Fremantle Hasluck Moore O'Connor Pearce Perth Stirling Swan Tangney Adelaide 945 636 2,537 2,687 818 978 866 911 2,715 1,464 691 2,035 4,808 2,178 7,587 10,021 1,077 1,888 3,025 751 885 704 1,499 1,536 1,653 408 1,886 1,155 1,371 1,718 1,025 953 339 9,241 1,274 1,012 1,420 303 3,260 1,437 730 886 1,484 330 1,038 286 234 991 971 217 195 216 201 1,114 484 247 748 2,504 731 4,221 3,537 379 551 1,131 123 202 258 640 467 749 112 818 343 430 675 332 347 131 6,885 424 444 583 72 1,684 505 343 380 602 139 459 LS7888 Released Document No. 29Released under the Freedom of Information Act 1982 --- Page 7 --- State Division Indigenous Australians in the Centrelink file and enrolled as at 30 June 2018 Potential not enrolled under 65 Indigenous Australians as at the 2019 close of rolls SA SA SA SA SA SA SA SA SA TAS TAS TAS TAS TAS ACT ACT ACT NT NT Total Barker Boothby Grey Hindmarsh Kingston Makin Mayo Spence Sturt Bass Braddon Clark Franklin Lyons Bean Canberra Fenner Lingiari Solomon No fixed address 1,207 620 3,938 1,141 810 773 443 1,936 416 1,199 2,491 964 1,231 1,521 702 468 631 16,473 3,659 224,554 610 221 1,931 412 278 242 131 705 154 326 364 279 255 399 168 157 145 5,483 1,395 1,166 69,805 LS7888 Released Document No. 29Released under the Freedom of Information Act 1982
This document, titled "Analysis of Indigenous Enrolment," details the Australian Electoral Commission's (AEC) updated methodology for estimating Indigenous electoral enrolment rates.
Summary of Document:
The document outlines the AEC's shift from an indirect, statistically-based regression model (used in 2014, yielding "alarmingly low" rates) to a more direct estimation method. This new approach, implemented since 2017, leverages Centrelink client files that include a self-identified Indigenous indicator. By matching these Centrelink records to the electoral roll, the AEC can more accurately estimate Indigenous enrolment rates.
Key findings and details presented:
* New Methodology: The AEC's "current method" involves person-level matching between the electoral roll and Centrelink data, including Centrelink's Indigenous indicator. An adjustment factor is applied to raw Centrelink-based enrolment rates to align them with overall AEC enrolment rates.
* Estimated Enrolment Rates: As of 30 June 2018, the estimated national Indigenous enrolment rate using this new methodology was 76.4%. State-level breakdowns are also provided.
* Enrolment Stimulation: Through data cleansing and analysis of the Centrelink file, the AEC identified 72,451 potentially unenrolled Indigenous Australians (under 65) as of mid-2018.
* Outcomes of Stimulation: An enrolment stimulation event conducted in February/March 2019, using the 2018 Centrelink data, resulted in the enrolment of 2,646 individuals who indicated they were Indigenous. This reduced the identified unenrolled cohort to 69,143.
* Data Limitations: The document acknowledges that Centrelink data represents only about half of the total electoral roll and is limited to Indigenous Australians receiving Centrelink benefits who have self-identified. It also notes issues with matching data for Indigenous Australians aged 65 and over.
* Detailed Breakdown: Tables provide granular data on enrolled and potentially unenrolled Indigenous Australians by state and electoral division.
Relevance to FOI Request LS7888:
This document is highly relevant to FOI request LS7888 as it directly addresses:
* New Methodology: It provides comprehensive details on the AEC's new methodology for its Indigenous Electoral Participation Program (IEPP), specifically its reliance on Centrelink's self-identified Indigenous data.
* Planning and Implementation: It describes the transition from the old method, the integration of Centrelink data into the Annual Roll Integrity Review (ARIR), and the execution of the enrolment stimulation event.
* Outcomes: It presents key outcomes such as the increased estimated Indigenous enrolment rate (76.4% by June 2018), the identification of potentially unenrolled Indigenous Australians, and the number of new enrolments resulting from the stimulation event.
Consistent with the FOI overview, the document does not explicitly detail the number of Indigenous people involved in the methodology's development or its final costs and dedicated AEC staffing. However, it significantly clarifies the technical planning, implementation, and initial outcomes of the new IEPP methodology.
7888-030.pdf (pdf)
Download file--- Page 1 --- Indigenous Franchise KPIs using Centrelink data Background As Australia’s federal electoral roll does not have a personal Indigenous identifier, estimating the enrolment, rates for Indigenous persons is problematic. Having these key franchise KPIs for the Indigenous population is important for Closing the Gap purposes and the AECs focus on how the franchise could be better delivered. In 2014 the AEC’s Research Section commenced indirect estimation of these figures, using a regression model of the relationship between the Indigenous population and each of the three franchise KPIs at a small area level, such as polling place catchment or the ABS’ SA2 geography. While the models did not fit the data particularly well, the resulting Indigenous enrolment, real data, rather than anecdote. Yet the resulting enrolment and turnout rates in particular were alarmingly low, and hopefully represented a worst case picture. rates were statistically based on Centrelink Data Around the same time, the Research Section informally proposed that person-level Centrelink extracts that the AEC received occasionally for other Roll validation work, could perhaps be extended to include Centrelink’s Indigenous indicator. This came to fruition in 20171. As part of the Annual Roll Integrity Review (ARIR), the person-level matching between the Roll and Centrelink now included Centrelink’s Indigenous indicator. However as Centrelink customers represent only around half of the Roll, the matching (even if perfect) would be far from assigning all enrolments an Indigenous flag. Nevertheless, for the first time the AEC has a very large number of enrolled persons, over 6 million, with an explicit Indigenous status. Enrolment Rate 1. The enrolment rate, even for all persons regardless of indigeneity, is difficult to calculate reliably due to conceptual, collection and methodological differences between the Roll (numerator) and the ABS-based enrolment-eligible population (denominator, “EEP”). In the case of the Indigenous population these problems are magnified by having no Indigenous identifier on the Roll and less reliable estimates of Indigenous EEP. However, matching the Roll to Centrelink data allows a more direct method of estimating the Indigenous enrolment rate. It is still subject to quite courageous assumptions, but as fewer are required it likely represents an improved approach. In summary: For all Centrelink customers with an identified Indigenous or non-Indigenous status, calculate the proportion on the Roll. This is the raw enrolment rate of Centrelink clients. Calculate the ratio of the actual AEC enrolment rate versus the Centrelink-based rate, for the same point in time. This is the overall factor to adjust any Centrelink-based rates to align with AEC enrolment rates. Apply the adjustment factor to raw Centrelink-based enrolment rates by Indigenous status to give Indigenous and non-Indigenous enrolment rates. This process can be done at the State level, though results will have lower reliability. NOTE: While this method avoids indirectly modelling Indigenous enrolment rates via geographic association, it does assume that any bias in Centrelink clients being enrolled is equal for Indigenous and non-Indigenous, and that there are no substantive Indigenous-status biases inherent in the Roll-to-Centrelink person matching. 1 Acknowledge with appreciation the work of the former Electoral Integrity Unit and the IT Branch. LS7888 Released Document No. 301919Released under the Freedom of Information Act 1982 --- Page 2 --- Conclusion and Results The Centrelink client file with Indigenous indicator newly available to the AEC has presented an excellent opportunity, via matching to the Roll and mark-off data, to improve estimation of Indigenous franchise KPIs. While the results are subject to a number of non-trivial assumptions, the more direct estimation methods lead to a higher likelihood of robust results. The figures should nevertheless still be regarded as indicative only. LS7888 Released Document No. 301919Released under the Freedom of Information Act 1982 --- Page 3 --- Released under the Freedom of Information Act 1982 --- Page 4 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
This document, titled "Indigenous Franchise KPIs using Centrelink data," details the Australian Electoral Commission's (AEC) updated methodology for estimating Indigenous electoral enrolment rates. It highlights the challenge of tracking Indigenous enrolment due to the absence of a specific identifier on the electoral roll.
The document explains the transition from an "indirect estimation" method (a regression model from 2014 that yielded "alarmingly low" and unreliable rates) to a "more direct method" developed in 2017. This new approach leverages person-level Centrelink data, specifically the Indigenous indicator included in Centrelink extracts. Although Centrelink data covers only about half of the electoral roll, it allows the AEC to directly associate Indigenous status with over 6 million enrolled persons for the first time.
The new methodology for calculating the Indigenous enrolment rate involves:
1. Calculating the raw enrolment rate for Centrelink clients (Indigenous and non-Indigenous) who are on the electoral roll.
2. Deriving an overall adjustment factor by comparing the actual AEC enrolment rate for all persons with the Centrelink-based rate.
3. Applying this adjustment factor to the raw Centrelink-based Indigenous and non-Indigenous enrolment rates to produce adjusted, indicative estimates.
While acknowledging that the results are based on "courageous assumptions" and should be considered "indicative only," the document concludes that this Centrelink-based method offers a "higher likelihood of robust results" compared to the previous model. The work of the former Electoral Integrity Unit and IT Branch is acknowledged for their contributions.
Relevance to FOI Request LS7888:
This document is highly relevant to FOI request LS7888 as it directly addresses the AEC's new methodology for its Indigenous Electoral Participation Program (IEPP) and provides insights into its development and impact.
- New Methodology: It comprehensively describes the "new methodology" for estimating Indigenous electoral enrolment rates, central to the FOI request, detailing the shift from indirect modelling to the direct use of Centrelink data.
- Planning & Implementation: The document outlines the genesis of the new approach (an "informal proposal" by the Research Section) and its "fruition in 2017" as part of the Annual Roll Integrity Review (ARIR), addressing the planning and implementation aspects.
- Outcomes: While not providing specific enrolment rate percentages, the document details the methodological improvement that led to the "significantly increased estimated Indigenous enrolment rates" noted in the FOI overview, moving away from "alarmingly low" figures to more "robust" (albeit indicative) results.
- Indigenous Involvement & Costs: Consistent with the FOI overview, this document does not explicitly detail the number of Indigenous people involved in the methodology's development nor does it provide information on the final costs or dedicated AEC staffing for this initiative.
7888-032.pdf (pdf)
Download file--- Page 1 --- Owen Jones From: Sent: To: Cc: Subject: Angelo Rojas Thursday, 12 September 2019 10:48 AM Sean Ferrari; Guy Lewis Stephanie Maxwell RE: Indigenous enrolment rates [SEC=UNCLASSIFIED] The estimated enrolled NSW Indigenous rate decreased only 1 percentage point from 87% in 2018 to 86% in 2019. This decrease is not statistically significant and falls within an estimation error for these calculations (being this difference actually marginal). This 1 percentage point difference represents in absolute numbers an increase of the estimated unenrolled NSW Indigenous population of 2,285 Indigenous (from 20,954 to 23,239) which then represents an increase of 10.9 per cent in unenrolled Indigenous when compared to previous year. However, note that this apparent b g increase is because the state size (the biggest in the country in relation to enrolment eligible population) where a decrease in 1 percentage point in enrolled indigenous then represents a big number in relation to the total estimated unenrolled indigenous. For example, if the estimated enrolled QLD indigenous population decreased 1 percentage point too from 73% in 2018 to 72% in 2019 (instead of the observed 1 percentage point increase in this period) then this decrease will represent a 6.3 per cent increase (2,362) of the estimated unenrolled QLD Indigenous population from 37,285 to 39,647 Indigenous. In this example, the increase (6.3 per cent) in QLD s lower than the actual increase observed in NSW (10.9 per cent) because the total estimated unenrolled Indigenous population in QLD (37,285 in 2018) is larger than that estimate in NSW (20,954 in 2018) – NSW has the largest estimated enrolled Indigenous rate in the Country at 86 per cent compared to QLD at only 74 per cent. Regards, Angelo Rojas | Senior Project Officer National Enrolment Services | Roll Management & Commun ty Engagement Branch Australian Electoral Commission T: (02) 6271 4507 X: 21336 M: From: Sean Ferrari Sent: Wednesday, 11 September 2019 1:21 PM To: Guy Lewis ; Delyse Paliaga ; Robyn Legg Cc: Angelo Rojas ; Stephanie Maxwell Subject: RE: Indigenous enrolment rates [SEC=UNCLASSIFIED] Guy, How did the non‐enrolled Indigenous population in NSW increase by 11%? That seems to be the only real outlier to me. In most cases it stayed pretty much the same. I think we will need to be prepared to answer the question about NSW. Thanks. Sean Ferrari | Director National Enrolment Services | Roll Management & Community Engagement Branch Australian Electoral Commission T: (02) 6271 4494 X: 21323 M: 1 LS7888 Released Document No. 3277Released under the Freedom of Information Act 1982 --- Page 2 --- From: Guy Lewis <Guy.Lewis@aec.gov.au> Sent: Wednesday, 11 September 2019 12:20 PM To: Sean Ferrari <Sean.Ferrari@aec.gov.au>; Delyse Paliaga <Delyse.Paliaga@aec.gov.au>; Robyn Legg <Robyn.Legg@aec.gov.au> Cc: Angelo Rojas <Angelo.Rojas@aec.gov.au>; Stephanie Maxwell <Stephanie.Maxwell@aec.gov.au> Subject: Indigenous enrolment rates [SEC=UNCLASSIFIED] We have the Indigenous enrolment rates for 2019 and we are preparing the paper work (Minute, speaking notes, QTB, draft letters etc.) covering them but hear is some early advice. National NSW NT WA Qld There was a small increase nationally from 76.4 per cent to 76.6 per cent, not really significant but better than a reduction. This continues the upward trend over the last few years with he 2017 rate at 74.7 per cent. The estimated Indigenous eligible enrolment population (EEP) grew by 13,360 people (or 2.7 per cent) while the estimated number of Indigenous people enrolled grew by 11,455 (3 per cent) The estimated unenrolled Indigenous population grew from 118,581 to 120,486 (1.6 per cent increase). This compares unfavourably with the change between 2017 and 2018 where the estimated unenrolled Indigenous population fell from 123,517 to 118,581 (4 per cent decrease). There was a decrease in NSW from 87.3 per cent to 86.2 per cent. While falling over the last two years the rate is up from the 2017 rate of 85.9 per cent. The estimated Indigenous EEP for NSW grew by 4,051 people (2.5 per cent) while the estimated number of Indigenous people enrolled in the NSW grew by 1,766 (1.2 per cent) The estimated unenrolled NSW Indigenous population grew from 20,954 to 23,239 (10.9 per cent increase). This compares unfavourably with the change between 2017 and 2018 where the estimated unenrolled Indigenous population fell from 22,605 to 20,954 (7.3 per cent decrease). There was a small increase in the NT from 67.5 per cent to 68.2 per cent. This continues the upward trend over the last few years with the 2017 rate at 67.1 per cent. The estimated Indigenous EEP for NT grew by 1,111 people (or 2.2 per cent) while the estimated number of Indigenous people enrolled in the NT grew by 1,132 (3.3 per cent) The estimated unenrolled NT Indigenous population fell from 16,433 to 16,412 (0.1 per cent decrease). This compares favourably with the change between 2017 and 2018 where the estimated unenrolled Indigenous population grew from 16,275 to 16,433 (1 per cent increase). There was a small increase in the WA from 63.8 per cent to 65 per cent. This continues the upward trend over the last few years with the 2017 rate at 62.8 per cent. The estimated Indigenous EEP for WA grew by 1,790 people (or 2.8 per cent) while the estimated number of Indigenous people enrolled in WA grew by 1,905 (4.6 per cent) The estimated unenrolled WA Indigenous population fell from 23,363 to 23,248 (0.1 per cent decrease). This compares favourably with the change between 2017 and 2018 where the estimated unenrolled Indigenous population fell from 23,367 to 23,363 (0.02 per cent decrease). There was a small increase in the Qld from 72.5 per cent to 73.7 per cent. This continues the upward trend over the last few years with the 2017 rate at 70 per cent. The estimated Indigenous EEP for Qld grew by 1,790 people (or 2.8 per cent) while the estimated number of Indigenous people enrolled in Qld grew by 1,905 (4.6 per cent) 2 LS7888 Released Document No. 32Released under the Freedom of Information Act 1982 --- Page 3 --- LS7888 Released Document No. 32 --- Page 4 --- Guy Lewis | Assistant Director National Enrolment Services | Roll Management & Community Engagement Branch Australian Electoral Commission T: (02) 6271 4758 X: 21312 M: 4 LS7888 Released Document No. 327Released under the Freedom of Information Act 1982 --- Page 5 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Document Summary
This document is an internal email exchange within the Australian Electoral Commission (AEC) from September 2019, discussing the latest Indigenous electoral enrolment rates for 2019, with comparisons to 2018 and 2017. Key points include:
- National Trends: The national Indigenous enrolment rate saw a small increase from 76.4% in 2018 to 76.6% in 2019, continuing an upward trend from 74.7% in 2017. While the estimated number of enrolled Indigenous people grew by 3%, the estimated unenrolled Indigenous population also increased by 1.6% (from 118,581 to 120,486), which was noted as "unfavourable" compared to the previous year's decrease.
- State-level Analysis:
- NSW: Experienced a slight decrease in its Indigenous enrolment rate from 87.3% (2018) to 86.2% (2019). Despite this small percentage drop, the estimated unenrolled NSW Indigenous population significantly increased by 10.9% (from 20,954 to 23,239), due to the large overall eligible population in the state. This increase in unenrolled numbers was a point of internal discussion, noting the need to be prepared to explain it.
- NT, WA, QLD: All showed small increases in their Indigenous enrolment rates, continuing upward trends from 2017. The NT and WA saw marginal decreases in their estimated unenrolled Indigenous populations. (Detailed QLD unenrolled figures are not fully present in the provided snippet but are implied in the discussion).
- Internal Discussion: The emails show AEC staff analyzing the data, particularly the "outlier" trend in NSW, and preparing documentation (minutes, speaking notes, QTB) for reporting on these figures.
Relevance to FOI Request LS7888
This document is highly relevant to FOI request LS7888, which sought information on the AEC's new Indigenous Electoral Participation Program (IEPP) methodology, specifically its planning, implementation, and outcomes for 2018-2019.
- Outcomes: The document directly addresses the "outcomes" aspect of the FOI request by providing precise Indigenous enrolment rates and detailed statistics on enrolled and unenrolled populations for 2018 and 2019, both nationally and for key states (NSW, NT, WA, QLD). These figures corroborate the overall national rate increases cited in the FOI request overview (76.4% by June 2018 and 76.6% by June 2019), demonstrating the tangible results of the new methodology.
- Implementation/Impact: By presenting the actual enrolment rates derived from the AEC's ongoing data collection (presumably using the Centrelink data-driven methodology), the document illustrates the impact and ongoing implementation of the new approach during the specified timeframe.
- Planning: The email exchange demonstrates internal AEC analysis and "paper work" preparation related to reporting on these outcomes, indicating ongoing planning and management of the IEPP and its metrics.
- Limitations: Consistent with the FOI overview, the document does not explicitly detail the number of Indigenous people involved in the methodology's development or its final costs and dedicated AEC staffing.
7888-037.pdf (pdf)
Download file--- Page 1 --- Owen Jones From: Sent: To: Cc: Subject: Hi Angelo, Tita Tabije Tuesday, 3 September 2019 2:18 PM Angelo Rojas; Gus Verzosa Guy Lewis RE: Indigenous KPI [DLM=For-Official-Use-Only] Yeah, good pick‐up by Gus. Thanks Gus. The revised numbers are now available and Gus is happy with the numbers. Angelo, when you’re free please come over so we can discuss. I created a file to note the number of records T:\Operations\BIREI\BI_SAS\04_ttabije\data\Number of records.xlsx from the input files to output datasets, for checking. Thanks, Tita Tita Tabije | Data Analyst Business Intelligence, Research & Electoral Integrity Section | Operations Branch Australian Electoral Commission T: (02) 6271 4473 X: 21901 From: Angelo Rojas Sent: Tuesday, 3 September 2019 8:57 AM To: Tita Tabije ; Gus Verzosa Subject: RE: Indigenous KPI [DLM=For‐Official‐Use‐Only] Good guys that you found that problem! I always assumed that the Federal Enrolment and Enrolment Eligible Population for June 2019 was already been updated on the workbook. Regards, Angelo Rojas | Senior Project Officer Roll Program & Community Engagement Section | Roll Management & Community Engagement Branch Australian Electoral Commission T: (02) 6271 4507 X: 21336 M: From: Tita Tabije <Tita.Tabije@aec.gov.au> Sent: Monday, 2 September 2019 10:54 AM To: Guy Lewis <Guy.Lewis@aec.gov.au> Cc: Andrew Trainer <Andrew.Trainer@aec.gov.au>; Sean Ferrari <Sean.Ferrari@aec.gov.au>; Gus Verzosa <Gus.Verzosa@aec.gov.au>; Sam Hannan‐Morrow <Sam.Hannan‐Morrow@aec.gov.au>; Angelo Rojas <Angelo.Rojas@aec.gov.au> Subject: RE: Indigenous KPI [DLM=For‐Official‐Use‐Only] Hi Guy, 1 LS7888 Released Document No. 377Released under the Freedom of Information Act 1982 --- Page 2 --- We are reviewing the numbers and we suggest you wait until we provide the final numbers before you write the Minute to ELT. Note that we just noticed that the comparison between 2019 and 2018 looks strange and we just fixed this. We will provide you with the revised numbers once our team has completed the review. Thanks, Tita Tita Tabije | Data Analyst Business Intelligence, Research & Electoral Integrity Section | Operations Branch Australian Electoral Commission T: (02) 6271 4473 X: 21901 From: Guy Lewis <Guy.Lewis@aec.gov.au> Sent: Friday, 30 August 2019 2:42 PM To: Tita Tabije <Tita.Tabije@aec.gov.au> Cc: Andrew Trainer <Andrew.Trainer@aec.gov.au>; Sean Ferrari <Sean.Ferrari@aec.gov.au>; Gus Verzosa <Gus.Verzosa@aec.gov.au>; Sam Hannan‐Morrow <Sam.Hannan‐Morrow@aec.gov.au>; Angelo Rojas <Angelo.Rojas@aec.gov.au> Subject: RE: Indigenous KPI [DLM=For‐Official‐Use‐Only] Thanks Tita, we will prepare a Minute to ELT re this and pass it through you for comment before we send it up. regards Guy Lewis | Assistant Director National Enrolment Services | Roll Management & Community Engagement Branch Australian Electoral Commission T: (02) 6271 4758 X: 21312 M: From: Tita Tabije <Tita.Tabije@aec.gov.au> Sent: Friday, 30 August 2019 2:29 PM To: Gus Verzosa <Gus.Verzosa@aec.gov.au>; Sam Hannan‐Morrow <Sam.Hannan‐Morrow@aec.gov.au>; Angelo Rojas <Angelo.Rojas@aec.gov.au> Cc: Guy Lewis <Guy.Lewis@aec.gov.au>; Andrew Trainer <Andrew.Trainer@aec.gov.au> Subject: Indigenous KPI [DLM=For‐Official‐Use‐Only] Hi All, Below are the results for 30 June 2019 ATSI enrolment rates for your review for endorsement to RMCE. Angelo and I have looked at these figures and are both happy with the results. If you want to see the working file, please see . T:\Strategic Capability\Strategic Research & Analysis\Research\Polling Places\IEPP & other requests\KPIs\Using 2016 Census Data & Methods\ARIR‐Centrelink\Indigenous Enrolment Rate 2019 BENCHMARKED.xlsb Angelo, as discussed, the only thing we need to check and make sure that is current is the below file. I will have a look and see if I can find a SAS program that outputs "Informality modelling.xlsb" next week. Note that any change in this file will have no effect on the results below. 2 LS7888 Released Document No. 377Released under the Freedom of Information Act 1982 --- Page 3 --- Non‐Indigenous 5066630 4065778 3101863 1181392 1573353 365800 103459 286429 15744704 144552 28434 101926 19003 42547 15609 35122 4225 391418 Enrolment Rates Indigenous 86% 75% 73% 70% 64% 84% 68% 80% 76% Non‐Indigenous 98% 96% 96% 97% 97% 98% 91% 99% The 'Missing' Indigenous 24344 9405 37380 8204 23788 2880 16552 1048 Non‐Indigenous 101348 162933 129077 32891 56613 9217 9985 2882 97% 123601 504946 Non‐Enrolment Rates Indigenous Non‐Indig 14% 25% 27% 30% 36% 16% 32% 20% 24% Indigenous Status Enrolment Rates, 30 June 2019 Enrolment Eligible Population Indigenous Non‐Indigenous Enrolments Indigenous 168896 37839 139306 27207 66335 18489 51674 5273 515019 5167978 4228711 3230940 1214283 1629966 375017 113444 289311 16249650 Estimated Indigenous voting age population as at 30 June 2019 Estimated enrolled as at 30 June 2019 Estimated ‘missing’ as at 30 June 2019 Change in estimated ‘missing’ between 30 June 2018 and 30 June 2019 168,896 144,552 24,344 3,390 NSW VIC QLD SA WA TAS NT ACT AUST State NSW VIC QLD WA SA TAS ACT NT 37,839 28,434 9,405 139,306 101,926 37,380 66,335 42,547 23,788 27,207 19,003 18,489 15,609 5,273 4,225 8,204 2,880 1,048 51,674 35,122 16,552 National 515,019 391,418 123,601 Differences between 30 June 2019 and 30 June 2018 COMPARISON Enrolment Eligible Population Indigenous NSW VIC QLD SA WA TAS 4051 1301 3664 685 1790 485 Non‐Indigenous ‐4051 ‐1301 ‐3664 ‐685 ‐1790 ‐485 642 95 425 304 26 19 119 5,020 Enrolments Indigenous 661 659 3569 381 1365 459 Non‐Indigenous ‐661 ‐659 ‐3569 ‐381 ‐1365 ‐459 Enrolment Rates Indigenous ‐1.7% ‐0.9% 0.7% ‐0.4% 0.3% 0.3% Non‐Indigenous 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 3 The 'Missing' Indigenous 3390 642 95 304 425 26 Non‐Indigenous ‐3390 ‐642 ‐95 ‐304 ‐425 ‐26 Non‐Enrolment Rates Indigenous Non‐Indigenous ‐0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 1.7% 0.9% ‐0.7% 0.4% ‐0.3% ‐0.3% LS7888 Released Document No. 37Released under the Freedom of Information Act 1982 --- Page 4 --- 1111 273 13360 ‐1111 ‐273 ‐13360 992 254 8340 ‐992 ‐254 ‐8340 0.5% 0.7% ‐0.4% 0.0% 0.0% 0.0% 119 19 5020 ‐119 ‐19 ‐5020 ‐0.5% ‐0.7% 0.4% 0.0% 0.0% 0.0% NT ACT AUST Thanks, Tita Tita Tabije | Data Analyst Business Intelligence, Research & Electoral Integrity Section | Operations Branch Australian Electoral Commission T: (02) 6271 4473 X: 21901 4 LS7888 Released Document No. 37Released under the Freedom of Information Act 1982 --- Page 5 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Summary of Document LS7888 Released Document No. 377
This document consists of an email exchange from late August to early September 2019 between various AEC staff members (Tita Tabije, Angelo Rojas, Gus Verzosa, Guy Lewis, Andrew Trainer, Sean Ferrari, Sam Hannan-Morrow) from the Business Intelligence, Research & Electoral Integrity (BIREI) and Roll Management & Community Engagement (RMCE) sections. The correspondence centers on the review and finalisation of "Indigenous KPI" results, specifically the estimated Indigenous electoral enrolment rates for 30 June 2019.
The emails reveal an internal process of data validation and refinement, including the identification and correction of an issue that made the comparison between 2018 and 2019 numbers appear "strange." Staff confirm that the revised figures are satisfactory and ready for endorsement by the Executive Leadership Team (ELT). A key detail is the reference to a working file path, T:\Strategic Capability\Strategic Research & Analysis\Research\Polling Places\IEPP & other requests\KPIs\Using 2016 Census Data & Methods\ARIR-Centrelink\Indigenous Enrolment Rate 2019 BENCHMARKED.xlsb
, which explicitly links the calculations to the Indigenous Electoral Participation Program (IEPP) and indicates the use of Centrelink data (ARIR-Centrelink
) in conjunction with 2016 Census data for Key Performance Indicators (KPIs).
The document includes two tables presenting the finalised data:
1. Indigenous Status Enrolment Rates, 30 June 2019: This table provides detailed figures by state and nationally for both Indigenous and Non-Indigenous populations. Key national figures for Indigenous Australians are:
* Estimated voting age population: 515,019
* Estimated enrolled: 391,418
* Estimated 'missing' (unenrolled): 123,601
* Enrolment Rate: 76%
2. Differences between 30 June 2019 and 30 June 2018: This table shows the changes in estimated populations, enrolments, and enrolment rates over the preceding year. Nationally, the estimated 'missing' Indigenous population decreased by 5,020 between June 2018 and June 2019.
Relevance to FOI Request LS7888
This document is highly relevant to FOI request LS7888, which sought information on the AEC's new methodology for its Indigenous Electoral Participation Program (IEPP) concerning its planning, implementation, and outcomes during 2018-2019:
- New Methodology (Implementation & Refinement): The email exchange and the referenced file path clearly demonstrate the ongoing implementation and internal review of the new methodology for calculating Indigenous enrolment rates. The explicit mention of
ARIR-Centrelink
data and2016 Census Data & Methods
directly confirms the application of the Centrelink-based approach for IEPP KPIs. The discussion about addressing data discrepancies between 2018 and 2019 highlights the meticulous refinement efforts within the methodology. - Outcomes: The attached tables provide concrete outcomes of the new methodology. Specifically, they present the national estimated Indigenous enrolment rate of 76% as of June 2019, which closely aligns with the FOI overview's stated outcome (76.6% for June 2019). The document also details the estimated number of unenrolled Indigenous Australians ("the missing") as 123,601 for June 2019 and indicates a reduction of 5,020 in this figure compared to June 2018, implying positive progress in enrolment efforts or data accuracy.
- Timeline: The document's dates (late August to early September 2019) and the data it presents (as of 30 June 2019) fall squarely within the 2018-2019 timeframe specified by the FOI request.
While highly relevant to the methodology's implementation and outcomes, the document does not explicitly detail the number of Indigenous people involved in the methodology's development or its final costs and dedicated AEC staffing, as noted as gaps in the FOI overview.
7888-04.pdf (pdf)
Download file--- Page 1 --- From: Angelo Rojas <Angelo.Rojas@aec.gov.au> Sent: Monday, 4 June 2018 3:45 PM To: Guy Lewis <Guy.Lewis@aec.gov.au> Cc: Stephanie Maxwell <Stephanie.Maxwell@aec.gov.au> Subject: FW: Indigenous data [SEC=UNCLASSIFIED] Guy - as discussed, the number of new enrolments for Indigenous not on the roll could be predicted to a fair extent by the proportion of Email/SMSs provided by State. Further analysis of this shows that the strong correlation is for people aged 18-39. The proportion of people aged 18-39 in NT, WA and SA who provided Email/SMS is low and so is their proportion of new enrolments. This analysis will be more direct when Mike provides me with the extract that includes Email and/or SMS sent to confirm two hypothesis: 1. We are sending less Email/SMS to these states. 2. Additionally, enrolment response rates are also lower for these states than for the rest (higher level of disengagement). LS7888 Released Document No. 4Released under the Freedom of Information Act 1982 --- Page 2 --- Released under the Freedom of Information Act 1982 --- Page 3 --- Released under the Freedom of Information Act 1982 --- Page 4 --- 60-69 70+ 1.4% 0.0% 100.0% 1.2% 0.2% 100.0% 1.8% 0.4% 100.0% 1.0% 0.2% 100.0% 3.6% 0.2% 100.0% 0.8% 0.0% 100.0% 1.0% 0.0% 100.0% 1.9% 0.4% 100.0% 1.4% 0.2% 100.0% Total % Angelo Rojas | Senior Project Officer Roll Program & Community Engagement Section | Roll Management & Community Engagement Branch LS7888 Released Document No. 4Released under the Freedom of Information Act 1982 --- Page 5 --- Australian Electoral Commission T: (02) 6271 4507 X: 21336 M: From: Warrick Poyser Sent: Thursday, 8 March 2018 5:12 PM To: Guy Lewis <Guy.Lewis@aec.gov.au> Cc: Delyse Paliaga <Delyse.Paliaga@aec.gov.au>; Gina Dario <Gina.Dario@aec.gov.au>; Stephanie Maxwell <Stephanie.Maxwell@aec.gov.au>; Angelo Rojas <Angelo.Rojas@aec.gov.au>; Michael Roden <Michael.Roden@aec.gov.au> Subject: RE: Indigenous data [SEC=UNCLASSIFIED] Thanks Guy – this data is very interesting. I suspect we may have to do something additional/special with the BDM data to establish eligibility prior to being able to do any meaningful direct enrolment effort. Warrick Poyser Systems Delivery & Modernisation Section | Operations Branch Australian Electoral Commission T: (02) 6271 4452 X: 21319 M: From: Guy Lewis Sent: Thursday, 8 March 2018 3:38 PM To: Warrick Poyser <Warrick.Poyser@aec.gov.au> Cc: Delyse Paliaga <Delyse.Paliaga@aec.gov.au>; Gina Dario <Gina.Dario@aec.gov.au>; Stephanie Maxwell LS7888 Released Document No. 4777Released under the Freedom of Information Act 1982 --- Page 6 --- <Stephanie.Maxwell@aec.gov.au>; Angelo Rojas <Angelo.Rojas@aec.gov.au>; Michael Roden <Michael.Roden@aec.gov.au> Subject: Indigenous data [SEC=UNCLASSIFIED] Warrick, the age breakdown of the 107,607 Indigenous people that are only on the Centrelink file is. Year of Birth 1900 – 1920 1924 – 1930 1931 – 1940 1941 – 1950 1951 – 1960 1961 – 1970 1971 – 1980 1981 – 1990 1991 – 2000 2001 Number 23 239 944 2,667 6,201 12,345 16,216 23,863 41,597 3,512 Total 107,607 % 0.02% 0.2% 0.9% 2.5% 5.8% 11.5% 15.1% 22.2% 38.7% 3.3% 100% Youth Group (18 to 24) 33,008 30.7% Mike is also rerunning the data against the deceased persons file and deleted records to remove those who are deceased and identify those who have been enrolled in the past. I’ll let you know the outcome when I get it. Guy Lewis | Asst. Director Roll Design Roll Progarm and Community Engagement | Roll Management and Community Engagement Branch Australian Electoral Commission T: Int: 21312 | Ext: (02) 6271 4758 | M: LS7888 Released Document No. 47Released under the Freedom of Information Act 1982 --- Page 7 --- LS7888 Released Document No. 4Released under the Freedom of Information Act 1982 --- Page 8 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
This document, dated March-June 2018, consists of internal emails between Australian Electoral Commission (AEC) staff (Angelo Rojas, Guy Lewis, Warrick Poyser, Michael Roden, Stephanie Maxwell, Delyse Paliaga, Gina Dario) concerning data analysis for Indigenous electoral enrolment.
Summary of Document Content:
- Centrelink Data Utilisation: The document prominently features the AEC's use of Centrelink data to identify potentially unenrolled Indigenous Australians. Specifically, an email from Guy Lewis (March 2018) details an age breakdown of "107,607 Indigenous people that are only on the Centrelink file," noting that 30.7% (33,008) are in the 18-24 age group.
- Data Refinement and Eligibility: Discussions include the need to cross-reference this Centrelink data with other sources (e.g., Births, Deaths, and Marriages (BDM) data, deceased persons files, and deleted records) to establish eligibility and remove those who are deceased or previously enrolled, prior to any direct enrolment efforts.
- Enrolment Stimulation and Outreach Analysis: An email from Angelo Rojas (June 2018) discusses analysis linking new Indigenous enrolments to the proportion of Email/SMS messages provided by state, especially for the 18-39 age group. It identifies lower Email/SMS provision and new enrolments in NT, WA, and SA, hypothesising lower outreach efforts or higher disengagement in these states.
Relevance to FOI Request LS7888:
This document is highly relevant to FOI request LS7888 as it provides direct insight into the planning and implementation of the AEC's new methodology for its Indigenous Electoral Participation Program (IEPP) during the 2018-2019 period.
- New Methodology: It explicitly details the AEC's reliance on self-identified Indigeneity data from Centrelink as a core component of the "more direct method for estimating Indigenous electoral enrolment rates" and identifying potential unenrolled voters, which was a key aspect of the new methodology. The 107,607 figure directly relates to the efforts to identify "over 72,000 potentially unenrolled Indigenous Australians" mentioned in the FOI overview (this document provides a specific number used in early stages).
- Planning and Implementation: The discussions about data refinement (BDM, deceased persons) illustrate the intricate planning and technical implementation steps involved in validating and utilising this new Centrelink-derived dataset.
- Outcomes and Enrolment Stimulation: The analysis concerning Email/SMS outreach and its correlation with new enrolments demonstrates the AEC's efforts to track and refine its strategies for "enrolment stimulation," and identify areas of lower engagement, directly addressing another aspect of the FOI request.
The document does not explicitly detail the number of Indigenous people involved in the methodology's development, its final costs, or dedicated AEC staffing, which were identified as gaps in the overall FOI release.
7888-040.pdf (pdf)
Download file--- Page 1 --- Owen Jones From: Sent: To: Cc: Subject: Angelo Rojas Thursday, 12 September 2019 10:48 AM Sean Ferrari; Guy Lewis Stephanie Maxwell RE: Indigenous enrolment rates [SEC=UNCLASSIFIED] The estimated enrolled NSW Indigenous rate decreased only 1 percentage point from 87% in 2018 to 86% in 2019. This decrease is not statistically significant and falls within an estimation error for these calculations (being this difference actually marginal). This 1 percentage point difference represents in absolute numbers an increase of the estimated unenrolled NSW Indigenous population of 2,285 Indigenous (from 20,954 to 23,239) which then represents an increase of 10.9 per cent in unenrolled Indigenous when compared to previous year. However, note that this apparent b g increase is because the state size (the biggest in the country in relation to enrolment eligible population) where a decrease in 1 percentage point in enrolled indigenous then represents a big number in relation to the total estimated unenrolled indigenous. For example, if the estimated enrolled QLD indigenous population decreased 1 percentage point too from 73% in 2018 to 72% in 2019 (instead of the observed 1 percentage point increase in this period) then this decrease will represent a 6.3 per cent increase (2,362) of the estimated unenrolled QLD Indigenous population from 37,285 to 39,647 Indigenous. In this example, the increase (6.3 per cent) in QLD s lower than the actual increase observed in NSW (10.9 per cent) because the total estimated unenrolled Indigenous population in QLD (37,285 in 2018) is larger than that estimate in NSW (20,954 in 2018) – NSW has the largest estimated enrolled Indigenous rate in the Country at 86 per cent compared to QLD at only 74 per cent. Regards, Angelo Rojas | Senior Project Officer National Enrolment Services | Roll Management & Commun ty Engagement Branch Australian Electoral Commission T: (02) 6271 4507 X: 21336 M: From: Sean Ferrari Sent: Wednesday, 11 September 2019 1:21 PM To: Guy Lewis ; Delyse Paliaga ; Robyn Legg Cc: Angelo Rojas ; Stephanie Maxwell Subject: RE: Indigenous enrolment rates [SEC=UNCLASSIFIED] Guy, How did the non‐enrolled Indigenous population in NSW increase by 11%? That seems to be the only real outlier to me. In most cases it stayed pretty much the same. I think we will need to be prepared to answer the question about NSW. Thanks. Sean Ferrari | Director National Enrolment Services | Roll Management & Community Engagement Branch Australian Electoral Commission T: (02) 6271 4494 X: 21323 M: 1 LS7888 Released Document No. 4077Released under the Freedom of Information Act 1982 --- Page 2 --- From: Guy Lewis <Guy.Lewis@aec.gov.au> Sent: Wednesday, 11 September 2019 12:20 PM To: Sean Ferrari <Sean.Ferrari@aec.gov.au>; Delyse Paliaga <Delyse.Paliaga@aec.gov.au>; Robyn Legg <Robyn.Legg@aec.gov.au> Cc: Angelo Rojas <Angelo.Rojas@aec.gov.au>; Stephanie Maxwell <Stephanie.Maxwell@aec.gov.au> Subject: Indigenous enrolment rates [SEC=UNCLASSIFIED] We have the Indigenous enrolment rates for 2019 and we are preparing the paper work (Minute, speaking notes, QTB, draft letters etc.) covering them but hear is some early advice. National NSW NT WA Qld There was a small increase nationally from 76.4 per cent to 76.6 per cent, not really significant but better than a reduction. This continues the upward trend over the last few years with he 2017 rate at 74.7 per cent. The estimated Indigenous eligible enrolment population (EEP) grew by 13,360 people (or 2.7 per cent) while the estimated number of Indigenous people enrolled grew by 11,455 (3 per cent) The estimated unenrolled Indigenous population grew from 118,581 to 120,486 (1.6 per cent increase). This compares unfavourably with the change between 2017 and 2018 where the estimated unenrolled Indigenous population fell from 123,517 to 118,581 (4 per cent decrease). There was a decrease in NSW from 87.3 per cent to 86.2 per cent. While falling over the last two years the rate is up from the 2017 rate of 85.9 per cent. The estimated Indigenous EEP for NSW grew by 4,051 people (2.5 per cent) while the estimated number of Indigenous people enrolled in the NSW grew by 1,766 (1.2 per cent) The estimated unenrolled NSW Indigenous population grew from 20,954 to 23,239 (10.9 per cent increase). This compares unfavourably with the change between 2017 and 2018 where the estimated unenrolled Indigenous population fell from 22,605 to 20,954 (7.3 per cent decrease). There was a small increase in the NT from 67.5 per cent to 68.2 per cent. This continues the upward trend over the last few years with the 2017 rate at 67.1 per cent. The estimated Indigenous EEP for NT grew by 1,111 people (or 2.2 per cent) while the estimated number of Indigenous people enrolled in the NT grew by 1,132 (3.3 per cent) The estimated unenrolled NT Indigenous population fell from 16,433 to 16,412 (0.1 per cent decrease). This compares favourably with the change between 2017 and 2018 where the estimated unenrolled Indigenous population grew from 16,275 to 16,433 (1 per cent increase). There was a small increase in the WA from 63.8 per cent to 65 per cent. This continues the upward trend over the last few years with the 2017 rate at 62.8 per cent. The estimated Indigenous EEP for WA grew by 1,790 people (or 2.8 per cent) while the estimated number of Indigenous people enrolled in WA grew by 1,905 (4.6 per cent) The estimated unenrolled WA Indigenous population fell from 23,363 to 23,248 (0.1 per cent decrease). This compares favourably with the change between 2017 and 2018 where the estimated unenrolled Indigenous population fell from 23,367 to 23,363 (0.02 per cent decrease). There was a small increase in the Qld from 72.5 per cent to 73.7 per cent. This continues the upward trend over the last few years with the 2017 rate at 70 per cent. The estimated Indigenous EEP for Qld grew by 1,790 people (or 2.8 per cent) while the estimated number of Indigenous people enrolled in Qld grew by 1,905 (4.6 per cent) 2 LS7888 Released Document No. 40Released under the Freedom of Information Act 1982 --- Page 3 --- LS7888 Released Document No. 40 --- Page 4 --- Guy Lewis | Assistant Director National Enrolment Services | Roll Management & Community Engagement Branch Australian Electoral Commission T: (02) 6271 4758 X: 21312 M: 4 LS7888 Released Document No. 407Released under the Freedom of Information Act 1982 --- Page 5 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Document Summary: AEC Indigenous Enrolment Rates (September 2019)
This document is an email exchange from September 2019 between key Australian Electoral Commission (AEC) staff in the National Enrolment Services Branch, primarily Angelo Rojas (Senior Project Officer), Sean Ferrari (Director), and Guy Lewis (Assistant Director). The correspondence discusses the estimated Indigenous electoral enrolment rates for 2019, comparing them to previous years (2017-2018) and anticipating questions from stakeholders.
Key Points:
- National Enrolment Trends (2017-2019):
- The national Indigenous enrolment rate saw a small increase from 76.4% in 2018 to 76.6% in 2019, continuing an upward trend from 74.7% in 2017.
- The estimated Indigenous eligible enrolment population (EEP) grew by 2.7% (13,360 people), while the estimated number of Indigenous people enrolled grew by 3% (11,455 people).
- The estimated unenrolled Indigenous population nationally increased by 1.6% (from 118,581 to 120,486), which was noted as "unfavourable" compared to the 4% decrease observed between 2017 and 2018.
- State-Specific Analysis:
- New South Wales (NSW): Experienced a slight decrease in its estimated Indigenous enrolment rate from 87.3% (2018) to 86.2% (2019). Despite this, the rate was still up from 85.9% in 2017. Critically, the estimated unenrolled NSW Indigenous population grew by 10.9% (from 20,954 to 23,239), which was identified as an "outlier."
- Northern Territory (NT) & Western Australia (WA): Both showed small increases in their enrolment rates and decreases or minor changes in their estimated unenrolled Indigenous populations, continuing upward trends.
- Queensland (Qld): Also recorded a small increase in its enrolment rate, continuing an upward trend from 70% in 2017.
- Explanation for NSW Outlier: Angelo Rojas provided a statistical explanation for the 10.9% increase in unenrolled Indigenous in NSW. He clarified that a 1 percentage point decrease in the enrolment rate (from 87% to 86%) was not statistically significant and fell within estimation error. The seemingly large percentage increase in unenrolled numbers was attributed to NSW having the largest estimated eligible Indigenous population and a high enrolment rate, meaning a small percentage drop affects a relatively smaller base of unenrolled people, thus making the percentage increase of the unenrolled appear higher compared to states with larger unenrolled populations.
Relevance to FOI Request LS7888:
This document is highly relevant to FOI request LS7888 as it directly addresses the "outcomes" and "impact" aspects of the AEC's new methodology for its Indigenous Electoral Participation Program (IEPP).
- Outcomes: It provides specific figures for estimated Indigenous enrolment rates for 2019 (national 76.6%), confirming the shift and significant increase in estimated rates compared to the old methodology's 58% and the 2018 rate of 76.4% mentioned in the FOI overview.
- Impact: The detailed national and state-level breakdowns of enrolled and unenrolled Indigenous populations demonstrate the impact of the new methodology in tracking electoral participation and identifying areas of focus (like the NSW outlier). The discussion about the growth of the unenrolled population, nationally and in specific states, directly relates to efforts to track enrolment stimulation.
- Planning/Implementation (indirect): While not detailing the development of the methodology, the email exchange itself reflects the internal AEC processes for discussing, analysing, and preparing to report on the results derived from the new methodology, indicating ongoing implementation and monitoring.
- Missing Information: Consistent with the FOI overview, this document does not provide explicit details on the number of Indigenous people involved in the methodology's development, its final costs, or dedicated AEC staffing for the program.
7888-041.pdf (pdf)
Download file--- Page 1 --- ESTIMATES BRIEF – HOT ISSUE Supplementary Budget Estimates, October 2019 CURRENT ENROLMENT FIGURES, TRENDS Issue For the 2019 federal election, more Australians were enrolled than ever before— both in terms of raw numbers and as a percentage of the eligible population. Key facts and figures 16,424,248 Australians were enrolled to vote in the 2019 federal election, representing an enrolment rate of 97.0 per cent. The youth enrolment rate (18-24 year olds) at the federal election was 88.8 per cent. Like the general population, young people are enrolled at record levels. . The Indigenous enrolment rate remained steady and for 2018-19 the national estimated Indigenous enrolment rate is at 76.6 per cent. Key points The Australian Electoral Commission works to ensure the full electoral participation of all eligible Australians. The electoral roll for the 2019 federal election was the largest and most complete ever. A record 16,424,248 Australians were determined as enrolled to vote in the 2019 federal election, from of an estimated eligible population of 16.9 million people. This represents an enrolment rate of 97.0 per cent, an increase on the previous record rate for the 2016 federal election. SB19-000044 . 1 LS7888 Released Document No. 411919191919Released under the Freedom of Information Act 1982 --- Page 2 --- Background 2019 federal election A record 16,424,248 electors were enrolled from of an estimated eligible population of 16.9 million people. This represents a national enrolment rate of 97.0 per cent. Immediately following the roll close the AEC reported a preliminary enrolment rate of 96.8 per cent. This, however, was subsequently revised to 97.0 per cent following the application of more up-to-date eligible enrolment population estimates. The enrolment rate exceeded the AEC’s target of 95 per cent in all states and territories except the Northern Territory, where it increased from 81.1 per cent at the 2016 election to 83.9 per cent. The youth enrolment rate (18-24 year olds) was 88.8 per cent. Like the general population, young people are enrolled at record levels. Table: Summary of key enrolment metrics Enrolment metric FE2016 AMLPS FE2019 Change between FE2016 and FE2019 Enrolled electors 15,676,659 16,011,891 16,424,248 747,589 Enrolment rate Estimated not enrolled 95.0% 95.7% 97.0% 2.0 percentage points 816,437 724,055 509,891 306,546 Youth enrolment rate 86.7% 88.6% 88.8% 2.1 percentage points Current state of the roll As at 30 September 2019, the national enrolment rate was 96.8 per cent, which is 0.2 percentage points lower than for the 2019 federal election. It is normal for the enrolment rate to decline following an election. The estimated number of people not enrolled has grown by around 37,000 since the election with the vast majority of these being people who turned 18 following the election. In the lead up to the 2019 election, the AEC conducted an enrolment stimulation exercise, targeting 410,000 people nationally who were not enrolled. Youth enrolment Younger electors tend to enrol in proximity to an electoral event. The youth enrolment rate (18-24 year olds) as at 30 September 2019 was 86.5 per cent with 1,643,338 enrolled youth, a reduction of 2.3 percentage points since the federal election. While the AEC works to maintain the youth enrolment rate as high as possible, it is not unusual for the youth enrolment rate to improve at the time of a federal election and then decline between elections. SB19-000044 2 LS7888 Released Document No. 41Released under the Freedom of Information Act 1982 --- Page 3 --- Indigenous enrolment The AEC commenced publication of Indigenous enrolment rate estimates by state and territory as at 30 June 2018. A previous national Indigenous enrolment estimate of 58 per cent was calculated in 2014 using a different methodology. The national Indigenous enrolment rate estimate remains steady and as at 30 June 2019 the rate was 76.6 per cent compared with 76.4 per cent in 2018. Any gap between Indigenous and non-Indigenous enrolment rates is concerning. The AEC is working to ensure the full electoral participation of all eligible Australians and is investing in partnerships and targeted activities that extend its reach into remote communities. Remote Area Mobile Polling remains an important component of election service delivery. Northern Territory enrolment Enrolment in the Northern Territory has shown consistent improvement. The enrolment rate in the Northern Territory increased by 9.8 percentage points between 30 June 2010 and 30 June 2019 while the national enrolment rate increased by 7.4 percentage points. Northern Territory Enrolment Estimated Eligible Population Enrolment rate Estimated not enrolled 30/06/2010 30/06/2011 30/06/2012 30/06/2013 30/06/2014 30/06/2015 30/06/2016 30/06/2017 30/06/2018 30/06/2019 118,401 121,919 122,715 126,934 130,857 129,133 133,553 137,773 138,581 140,064 158,894 155,559 155,537 156,552 161,868 162,780 164,012 165,876 165,118 166,136 74.5% 78.4% 78.9% 81.1% 80.8% 79.3% 81.4% 83.1% 83.9% 84.3% 40,493 33,640 32,822 29,618 31,011 33,647 30,459 28,103 26,537 26,072 Public reporting A range of enrolment statistics are published regularly on the AEC website. Date: Cleared by (FAC): Telephone: Branch: Contact officer: Telephone: Consultation: PDR number: 22 October 2019 Tim Courtney for Andrew Gately (02) 6271 4410 Roll Management and Community Engagement Robyn Legg (02) 6271 4786 Nil SB19-000044 SB19-000044 3 LS7888 Released Document No. 41Released under the Freedom of Information Act 1982 --- Page 4 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Document Summary
This document, an "Estimates Brief – Hot Issue" from October 2019, provides an overview of enrolment figures and trends for the Australian Electoral Commission (AEC), particularly in the lead-up to Supplementary Budget Estimates. It highlights record national enrolment rates of 97.0% for the 2019 federal election and a youth enrolment rate of 88.8%.
Crucially, the brief details Indigenous enrolment rates, stating that the AEC commenced publication of these estimates by state and territory as of 30 June 2018. It explicitly contrasts the national estimated Indigenous enrolment rate of 76.6% (as at 30 June 2019, and 76.4% in 2018) with a "previous national Indigenous enrolment estimate of 58 per cent calculated in 2014 using a different methodology." The document also notes the AEC's ongoing efforts to ensure full electoral participation for all eligible Australians, including targeted activities and partnerships in remote communities, and mentions an enrolment stimulation exercise targeting 410,000 unenrolled people before the 2019 election.
Relevance to FOI Request LS7888
This document is highly relevant to FOI request LS7888, which sought information on the AEC's new methodology for its Indigenous Electoral Participation Program (IEPP) from 2018-2019.
- New Methodology & Outcomes: The document directly addresses the shift in methodology by clearly stating the AEC began publishing new Indigenous enrolment rates from June 2018 and contrasting them with the 2014 estimate based on a "different methodology." It provides the exact "outcomes" requested by the FOI, confirming the new estimated Indigenous enrolment rates of 76.4% (June 2018) and 76.6% (June 2019), which directly reflect the impact of the new approach (as described in the FOI overview, likely due to Centrelink data).
- Timeframe: The document falls squarely within the 2018-2019 timeframe of the FOI request, providing data points from this period.
- Implementation (Implicit): While not detailing the planning or explicit implementation steps of the new methodology's development, it confirms its application by stating when the new estimates began to be published.
- Missing Information: Consistent with the FOI overview, this document does not provide specific details on the number of Indigenous people involved in the methodology's development, its final costs, or dedicated AEC staffing, thus aligning with the gaps identified in the released content for LS7888.
7888-05.pdf (pdf)
Download file--- Page 1 --- Released under the Freedom of Information Act 198219LS7888 Released Document No. 5 --- Page 2 --- Released under the Freedom of Information Act 198219LS7888 Released Document No. 5 --- Page 3 --- Released under the Freedom of Information Act 198219LS7888 Released Document No. 5 --- Page 4 --- Released under the Freedom of Information Act 198219LS7888 Released Document No. 5 --- Page 5 --- Released under the Freedom of Information Act 198219LS7888 Released Document No. 5 --- Page 6 --- Released under the Freedom of Information Act 198219LS7888 Released Document No. 5 --- Page 7 --- Released under the Freedom of Information Act 198219LS7888 Released Document No. 5 --- Page 8 --- Released under the Freedom of Information Act 198219LS7888 Released Document No. 5 --- Page 9 --- Released under the Freedom of Information Act 198219LS7888 Released Document No. 5 --- Page 10 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
This document, identified as "Released Document No. 5" for FOI request LS7888, consists of nine blank placeholder pages followed by one page detailing a list of "REDACTION CODES."
Relevance to FOI Request LS7888:
This document does not contain any substantive information directly related to the Australian Electoral Commission's (AEC) new methodology for its Indigenous Electoral Participation Program (IEPP). It provides no details regarding the program's planning, implementation, outcomes, Indigenous involvement in its development, or its associated costs. Its sole functional content is a list of codes explaining the types of information redacted from other documents released under the Freedom of Information Act 1982, thereby serving as an administrative key rather than a substantive response to the FOI request's core inquiries.
7888-06.pdf (pdf)
Download file--- Page 1 --- Indigenous Franchise KPIs using Centrelink data Background As Australia’s federal electoral roll does not have a personal Indigenous identifier, estimating the enrolment, rates for Indigenous persons is problematic. Having these key franchise KPIs for the Indigenous population is important for Closing the Gap purposes and the AECs focus on how the franchise could be better delivered. In 2014 the AEC’s Research Section commenced indirect estimation of these figures, using a regression model of the relationship between the Indigenous population and each of the three franchise KPIs at a small area level, such as polling place catchment or the ABS’ SA2 geography. While the models did not fit the data particularly well, the resulting Indigenous enrolment, real data, rather than anecdote. Yet the resulting enrolment and turnout rates in particular were alarmingly low, and hopefully represented a worst case picture. rates were statistically based on Centrelink Data Around the same time, the Research Section informally proposed that person-level Centrelink extracts that the AEC received occasionally for other Roll validation work, could perhaps be extended to include Centrelink’s Indigenous indicator. This came to fruition in 20171. As part of the Annual Roll Integrity Review (ARIR), the person-level matching between the Roll and Centrelink now included Centrelink’s Indigenous indicator. However as Centrelink customers represent only around half of the Roll, the matching (even if perfect) would be far from assigning all enrolments an Indigenous flag. Nevertheless, for the first time the AEC has a very large number of enrolled persons, over 6 million, with an explicit Indigenous status. Enrolment Rate 1. The enrolment rate, even for all persons regardless of indigenaity, is difficult to calculate reliably due to conceptual, collection and methodological differences between the Roll (numerator) and the ABS-based enrolment-eligible population (denominator, “EEP”). In the case of the Indigenous population these problems are magnified by having no Indigenous identifier on the Roll and less reliable estimates of Indigenous EEP. However, matching the Roll to Centrelink data allows a more direct method of estimating the Indigenous enrolment rate. It is still subject to quite courageous assumptions, but as fewer are required it likely represents an improved approach. In summary: For all Centrelink customers with an identified Indigenous or non-Indigenous status, calculate the proportion on the Roll. This is the raw enrolment rate of Centrelink clients. Calculate the ratio of the actual AEC enrolment rate versus the Centrelink-based rate, for the same point in time. This is the overall factor to adjust any Centrelink-based rates to align with AEC enrolment rates. Apply the adjustment factor to raw Centrelink-based enrolment rates by Indigenous status to give Indigenous and non-Indigenous enrolment rates. This process can be done at the State level, though results will have lower reliability. NOTE: While this method avoids indirectly modelling Indigenous enrolment rates via geographic association, it does assume that any bias in Centrelink clients being enrolled is equal for Indigenous and non-Indigenous, and that there are no substantive Indigenous-status biases inherent in the Roll-to-Centrelink person matching. 1 Acknowledge with appreciation the work of the former Electoral Integrity Unit and the IT Branch. 1919LS7888 Released Document No. 6Released under the Freedom of Information Act 1982 --- Page 2 --- Conclusion and Results The Centrelink client file with Indigenous indicator newly available to the AEC has presented an excellent opportunity, via matching to the Roll , to improve estimation of Indigenous franchise KPIs. While the results are subject to a number of non-trivial assumptions, the more direct estimation methods lead to a higher likelihood of robust results. The figures should nevertheless still be regarded as indicative only. 191919LS7888 Released Document No. 6Released under the Freedom of Information Act 1982 --- Page 3 --- INDIGENOUS-STATUS FRANCHISE KPIs Non-enrolment (2017) KPI Indigenous 29% Non- Indigenous 3% ^ enrolment INDIGENOUS-STATUS (NON-)ENROLMENT 30 June 2017 State NSW VIC QLD SA WA TAS NT ACT AUST Indigenous 17% 28% 32% 34% 39% 19% 30% 26% 29% Non- Indigenous 2% 4% 4% 3% 3% 3% 2% 1% 3% Notes: These modelled national estimates contain many assumptions and cannot be ascribed to any particular sub-national location. The absence of an Indigenous identifier on the Federal Roll has been partially supplemented with Centrelink information. These figures can be footnoted "unofficial internal AEC modelling". Business Intelligence, Research and Electoral Integrity Section. 8 May 2018. 19191919LS7888 Released Document No. 6Released under the Freedom of Information Act 1982 --- Page 4 --- REDACTION CODES Personal Information (name) redacted. Personal Information (date of birth) redacted. Personal Information (photograph) redacted Personal Information (facsimile of signature) redacted. Personal Information (facsimile of manuscript initialling) redacted. Personal Information (Individual’s address) redacted. Personal Information (individual’s telephone number) redacted. Personal Information (individual’s opinion) redacted. Personal Information (opinion about individual) redacted. Personal Information (employment history) redacted. Personal Information (qualifications) redacted. Personal Information (health) redacted. Personal Information (identifying individual) redacted. Business information (Bank Account details) redacted. Business information (Billing Account details) redacted. Business information (internal operations) redacted. Legal Professional Communication redacted. Deliberative material redacted. Irrelevant material redacted. Electoral Roll material redacted. Tests, examinations or audits material redacted. Management or assessment of personnel material redacted. Proper and efficient conduct of the operations of AEC material redacted. Lawful methods or procedures for preventing, detecting, investigating, or dealing with matters arising out of, breaches or evasions of the law. Trade Secrets. Commercial valuable information. Financial interests of the Commonwealth. Property interests of the Commonwealth. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Document Summary and Relevance to FOI Request LS7888
This document, titled "Indigenous Franchise KPIs using Centrelink data" and dated 8 May 2018, details the Australian Electoral Commission's (AEC) new methodology for estimating Indigenous electoral enrolment rates.
Document Summary:
The document outlines the AEC's challenge in accurately estimating Indigenous electoral participation due to the absence of a specific Indigenous identifier on the federal electoral roll. It describes a shift from an earlier (2014) indirect regression model, which yielded "alarmingly low" and less reliable estimates, to a more direct approach leveraging Centrelink's self-identified Indigenous status data.
Beginning in 2017, the AEC integrated Centrelink's Indigenous indicator into its Annual Roll Integrity Review (ARIR) through person-level matching, providing explicit Indigenous status for over 6 million enrolled persons. The new methodology involves:
1. Calculating the raw enrolment rate for identified Indigenous and non-Indigenous Centrelink clients.
2. Determining an adjustment factor by comparing the overall AEC enrolment rate with the Centrelink-based rate.
3. Applying this factor to derive adjusted Indigenous and non-Indigenous enrolment rates.
While acknowledging that the results are indicative and rely on certain assumptions, the document concludes that this method provides a "higher likelihood of robust results." It presents indicative national non-enrolment figures for 30 June 2017, showing a 29% non-enrolment rate for Indigenous Australians (implying a 71% enrolment rate) compared to 3% for non-Indigenous Australians. The development of this method is attributed to the AEC's Research Section, with acknowledgment to the former Electoral Integrity Unit and the IT Branch.
Relevance to FOI Request LS7888:
This document is highly relevant to FOI request LS7888, which sought information on the AEC's new methodology for its Indigenous Electoral Participation Program (IEPP) from 2018-2019.
- New Methodology: The document directly addresses the core of the FOI request by detailing the new methodology for estimating Indigenous electoral enrolment using Centrelink data, replacing the previous 2014 model. This shift from an indirect, less accurate method to a more direct, data-driven approach is a central theme of the request.
- Planning & Implementation: It outlines the genesis of this shift, from an "informal proposal" by the Research Section to its "fruition in 2017" and integration into the ARIR, indicating the planning and initial implementation phases within the requested timeframe.
- Outcomes: The document provides initial outcomes of this new methodology, presenting estimated Indigenous non-enrolment rates for 2017 that are significantly higher (indicating lower non-enrolment) than those implied by the previous methodology's "alarmingly low" figures mentioned in the FOI overview (e.g., 71% enrolment vs. ~58%). This demonstrates the immediate impact of the methodological change on reported rates.
- Indigenous Involvement: The document does not explicitly detail the number of Indigenous people involved in the methodology's development, which was a specific aspect of the FOI request that the released content did not fully address.
- Costs/Staffing: Similarly, the document does not explicitly detail the final costs or dedicated AEC staffing associated with the development or ongoing use of this methodology. It only mentions the AEC sections involved.
- Timeline: Dated May 2018 and presenting 2017 data, the document falls squarely within the 2018-2019 period of the FOI request, providing insight into the foundational work and early results of the new methodology.
7888-08.pdf (pdf)
Download file--- Page 1 --- Indigenous Franchise KPIs using Centrelink data Background As Australia’s federal electoral roll does not have a personal Indigenous identifier, estimating the enrolment, rates for Indigenous persons is problematic. Having these key franchise KPIs for the Indigenous population is important for Closing the Gap purposes and the AECs focus on how the franchise could be better delivered. In 2014 the AEC’s Research Section commenced indirect estimation of these figures, using a regression model of the relationship between the Indigenous population and each of the three franchise KPIs at a small area level, such as polling place catchment or the ABS’ SA2 geography. While the models did not fit the data particularly well, the resulting Indigenous enrolment, real data, rather than anecdote. Yet the resulting enrolment and turnout rates in particular were alarmingly low, and hopefully represented a worst case picture. rates were statistically based on Centrelink Data Around the same time, the Research Section informally proposed that person-level Centrelink extracts that the AEC received occasionally for other Roll validation work, could perhaps be extended to include Centrelink’s Indigenous indicator. This came to fruition in 20171. As part of the Annual Roll Integrity Review (ARIR), the person-level matching between the Roll and Centrelink now included Centrelink’s Indigenous indicator. However as Centrelink customers represent only around half of the Roll, the matching (even if perfect) would be far from assigning all enrolments an Indigenous flag. Nevertheless, for the first time the AEC has a very large number of enrolled persons, over 6 million, with an explicit Indigenous status. Enrolment Rate 1. The enrolment rate, even for all persons regardless of indigenaity, is difficult to calculate reliably due to conceptual, collection and methodological differences between the Roll (numerator) and the ABS-based enrolment-eligible population (denominator, “EEP”). In the case of the Indigenous population these problems are magnified by having no Indigenous identifier on the Roll and less reliable estimates of Indigenous EEP. However, matching the Roll to Centrelink data allows a more direct method of estimating the Indigenous enrolment rate. It is still subject to quite significant assumptions, but as fewer are required it likely represents an improved approach. In summary: For all Centrelink customers with an identified Indigenous or non-Indigenous status, calculate the proportion on the Roll. This is the raw enrolment rate of Centrelink clients. Apply the raw enrolment rates to Indigenous and non-Indigenous EEP benchmarks2 to yield draft Indigenous status enrolments. Calculate the factor to scale these to the overall enrolments used in the AEC’s published enrolment rates. Apply the adjustment factor to the draft Indigenous status enrolments, then recalculate the enrolment rates against the EEP benchmarks. This process can be done at the State level, though results will have lower reliability. NOTE: While this method avoids indirectly modelling Indigenous enrolment rates via geographic association, it does assume that any bias in Centrelink clients being enrolled is equal for Indigenous and non-Indigenous, and that there are no substantive Indigenous-status biases inherent in the Roll-to-Centrelink person matching. 1 Acknowledge with appreciation the work of the former Electoral Integrity Unit and the IT Branch. 2 Indigenous benchmarks use ABS projected adult populations, non-Indigenous are then the residual with published EEP. LS7888 Released Document No. 81919Released under the Freedom of Information Act 1982 --- Page 2 --- Conclusion and Results The Centrelink client file with Indigenous indicator newly available to the AEC has presented an excellent opportunity, via matching to the Roll , to improve estimation of Indigenous franchise KPIs. While the results are subject to a number of non-trivial assumptions, the more direct estimation methods lead to a higher likelihood of robust results. The figures should nevertheless still be regarded as indicative only. LS7888 Released Document No. 8191919Released under the Freedom of Information Act 1982 --- Page 3 --- Released under the Freedom of Information Act 1982 --- Page 4 --- LS7888 Released Document No. 819Released under the Freedom of Information Act 1982
This document, "Indigenous Franchise KPIs using Centrelink data," outlines the Australian Electoral Commission's (AEC) transition to a new methodology for estimating Indigenous electoral enrolment rates. It details the limitations of the previous 2014 regression model, which produced "alarmingly low" and statistically unreliable figures due to the absence of an Indigenous identifier on the electoral roll. The document describes the informal proposal and subsequent implementation (from 2017) of a more direct method leveraging self-identified Indigeneity data from Centrelink. This new process involves matching electoral roll data with Centrelink customer records (over 6 million, representing approximately half the roll) to establish Indigenous status, then applying these enrolment rates to Indigenous and non-Indigenous Enrolment-Eligible Population (EEP) benchmarks. The document concludes that despite some inherent assumptions, this Centrelink-based approach offers an "excellent opportunity" for improved estimation and yields a "higher likelihood of robust results" compared to the prior indirect modelling.
Relevance to FOI Request LS7888:
This document is central to FOI request LS7888 as it directly addresses the AEC's "new methodology for its Indigenous Electoral Participation Program (IEPP)."
- Planning and Implementation: It details the strategic planning (informal proposal by the Research Section) and the timeline of implementation (came to fruition in 2017 as part of the Annual Roll Integrity Review) of the Centrelink-based method. This directly explains the shift from the 2014 methodology.
- Outcomes: While not stating the specific enrolment percentages, the document explains the methodological change that underpins the significant increase in estimated Indigenous enrolment rates (e.g., from 58% to over 76%) as noted in the FOI request overview, framing the new approach as leading to more "robust results."
- Indigenous Involvement and Costs: Consistent with the FOI overview, this document does not explicitly detail the number of Indigenous people involved in the methodology's development or its specific final costs and dedicated AEC staffing.