Census of Population and Housing: Understanding the Increase in Aboriginal and Torres Strait Islander Counts methodology

1 This publication presents analysis that draws from a range of data sources:

• Australian Census of Population and Housing
• Census Post Enumeration Survey (cat. no. 2940.0)
• Births, Australia (cat. no. 3301.0)
• Deaths, Australia (cat. no. 3302.0)
• Estimated Residential Population (cat. no. 3238.0.55.001.0)
• Australian Census Longitudinal Dataset (cat. no. 2080.5)
   

2 This publication presents counts from the Census of Population and Housing (1971-2016) for people who identified as being of Aboriginal and/or Torres Strait Islander origin and were counted in the Census of Population and Housing.

3 The 2016 Census of Population and Housing was held on 9 August 2016. The Census is the largest statistical collection undertaken by the Australian Bureau of Statistics (ABS). By counting the number and key characteristics of people in Australia on Census night, the Census provides a reliable basis for the estimation of the population of the states, territories and local government areas, for use in:

• determining the number of seats allocated to each state in the House of Representatives.
• distributing billions of dollars of annual goods and services tax revenue to the states and territories.
• determining state grants to local government areas.

4 The Census counts all people in Australia on Census night (excluding foreign diplomats and their families). This means that visitors to Australia are counted regardless of how long they have been in the country or how long they plan to stay, but that Australian residents who are overseas on Census night are not included.

5 People in Norfolk Island on Census night were counted for the first time in the 2016 Australian Census following passage of the Norfolk Island Legislation Amendment Act 2015. The Territories of Cocos (Keeling) Islands and Christmas Island have been included in the Australian Census since 1996. Expeditioners to Australian bases in the Australian Antarctic Territory (and other locations) are also included in Census counts.

6 The Census Post Enumeration Survey provides an independent measure of Census coverage through its estimate of undercount and overcount. Coverage and response rates are important measures in understanding Census data quality. The key indicators for the 2016 Census of Population and Housing support that the Census data is of high quality, with high response rates (95.1%) and low levels of net undercoverage (1.0%).

7 The Census question on Aboriginal and Torres Strait Islander status is designed to determine the number, distribution and characteristics of people of Aboriginal and/or Torres Strait Islander origin. It is also used in estimation and projections of the Aboriginal and/or Torres Strait Islander population. The Census is the only comprehensive source of small area data on the Aboriginal and Torres Strait Islander population.

8 For all Censuses prior to 1971 respondents were asked to state their race and, where race was mixed, to specify the proportions of each. In the 1971 and 1976 Censuses, a question with response categories of European, Aboriginal, Torres Strait Islander and Other was included. Since 1981, a specific question has been asked to identify Aboriginal and Torres Strait Islander peoples.

9 The responses to this question are captured automatically from mark box responses on the form, so the risk of processing error is minimal. Where respondents identify with both Aboriginal and Torres Strait Islander origins, they were instructed to mark both the Aboriginal and Torres Strait Islander boxes. In cases where a respondent marked both No and Yes boxes, then the Indigenous Status was set to Not Stated. This treatment differs to that used in the 2011 Census where if both No and Yes mark boxes were selected, the responses were accepted in the order they appeared on the form and the extra response was rejected. This change in treatment aligns with the Indigenous Status Standard, 2014, Version 1.5 (cat. no. 1200.0.55.008).

For more information on definitional changes and Census questions, refer to the Census Dictionary (cat. no. 2901.0).

10 For the 2011 Census and earlier Censuses, the ABS relied on a large field workforce to list every household across Australia. Field officers knocked on every door in an attempt to make contact with householders and to deliver paper Census forms or online Census login codes. The field officers not only delivered and collected Census material, but also recorded some characteristics about the houses in which the people lived.

11 For the 2016 Census, the ABS developed a new method that addressed the difficulties in recruiting field staff, as well as investing in the advantages and efficiencies of new technologies. Changes were designed taking into account international best practices in Census taking and building on the Australian public's increasing access to and use of the internet, and their willing support of the Census. The new method made three main changes from that used in earlier Censuses.

1. The development of a national Address Register to support mailing out of materials to households across Australia. The Address Register was formed using the Geocoded National Address File (GNAF) as its base and then built on using information gained through visiting every address through a large canvassing exercise, in addition to analysing other available data. This register formed the basis of addresses to which information was mailed out.
2. The postal delivery of an instruction letter detailing how to complete the Census online or how to request a paper form. Paper forms were then despatched and mailed back via Australia Post. This ensured that Census field officers only visited a minority of homes, where the use of the mail service was not considered viable, or where a Census form had not been returned.
3. The use of a smarter online form. Many enhancements were made to the online form to improve quality and make it easier for respondents to complete. These enhancements are detailed in the Online form section.
    

12 As in previous Censuses, in 2016 there were a range of approaches used for collecting information from specific population groups. This included strategies to increase participation of remote Aboriginal and Torres Strait Islander peoples and urban Aboriginal and Torres Strait Islander peoples. These approaches were designed in consultation and collaboration with relevant communities and/or service providers to ensure the coverage of all people in Australia (including these specific populations) was as complete as possible.

13 Since 1976, measures have been adopted to maximise coverage of the Aboriginal and Torres Strait Islander population. These include specific Census awareness activities, greater use of Aboriginal and Torres Strait Islander staff, greater involvement of Aboriginal and Torres Strait Islander organisations, and better coordination of the related field operations. In remote areas, interview forms have been used to facilitate the enumeration of this population.

14 The 2016 Discrete Communities and Remote Areas Strategy incorporated improvements to enumeration of Aboriginal and Torres Strait Islander people living in remote Aboriginal and Torres Strait Islander communities. At a broad level, the improvements for the 2016 Census included:

• earlier and ongoing engagement;
• a reduction in the overall enumeration period for Aboriginal and Torres Strait Islander communities;
• an increase in the number of field staff recruited; and
• a greater level of support to Aboriginal and Torres Strait Islander people requiring assistance in completing their form (in both urban and remote areas).
   

15 For more information on the Discrete Community and Remote Areas Strategy in the Census Dictionary, 2016 (cat no. 2901.0).

16 For the 2016 Census the ABS also incorporated improvements to assist Aboriginal and Torres Strait Islanders living in urban communities to participate in the Census. As part of this strategy in pre-identified urban areas increased support and assistance was provided to Aboriginal and Torres Strait Islander people, including the option of collecting Census information through an interview.

16 For extensive explanatory information on the PES please see Census of Population and Housing: Details of Overcount and Undercount, Australia (cat.no. 2940.0).

17 For Census and PES purposes, scope refers to the group of people about which information is required. A set of rules is applied to determine whether a selected person is within this population of interest.

18 The scope of the Census is every person present in Australia on Census night with the exception of foreign diplomats and their families. Ideally the PES would sample from all people who were, or should have been, counted in the Census. For practical reasons, however, there were a number of areas, dwellings and people outside the scope of the 2016 PES.

19 Of the people present in Australia at the time of the PES, the following are not included in the scope of the PES:

• overseas visitors who were not in Australia on 9 August 2016 (Census night)
• foreign diplomats and their families
• people in non-private dwellings such as hotels, motels, hospitals and other institutions
• homeless people (as the sample selected in the PES is based on a selection of dwellings)
• babies born after 9 August 2016; and
• people in Other territories (e.g. Christmas Island, Norfolk Island and Jervis Bay Territory) or Australian external territories (e.g. Australian Antarctic Territory, Coral Sea Islands Territory).
   

20 The PES also does not obtain information about people who died between Census and PES.

21 The 2016 PES sample included remote areas and Discrete Communities, which ensured a more complete geographic coverage of the PES. In 2016, the risk to statistical independence in these areas and communities was effectively managed through careful monitoring of Census field operations, interviewer training and well established field procedures. No PES interviewers worked on Census collection activities in these same areas.

22 The 2016 PES sample excluded people living in NPDs, as has been the case in previous PES cycles. NPDs are establishments which provide predominantly short-term accommodation for communal or group living, and often provide common eating facilities. They include hotels, motels, hostels, hospitals, religious institutions providing accommodation, educational institutions providing accommodation, prisons, boarding houses and short-stay caravan parks.

23 Coverage refers to a set of rules designed to give each in scope person in the PES sample a single chance of selection in the survey. These rules are implemented by associating each person with a single dwelling through a series of questions in the PES interview, such as where each person usually lives and whether they (or anyone else) are staying at their usual residence during the PES enumeration period.

24 An example of the need for coverage rules is for a visitor staying at a dwelling selected in the PES. If the visitor reports that someone else is staying at their usual residence during PES enumeration, or they return to their residence at a later point during the enumeration period, this visitor could have two chances of being included in the PES (i.e. once at their usual residence and again at their current location). In this example, the PES coverage rules would associate the visitor with their place of usual residence and not the dwelling they were visiting, thereby giving them only one chance of being included in the PES enumeration.

25 Specially trained PES interviewers collected data through interviews that started seven weeks after Census night. Interviews were conducted with any responsible adult of the household, who was asked to respond on behalf of all household members.

26 Most interviews were conducted face-to-face; however, respondents were given the option to register for a telephone interview, which had a take-up rate of 32%. Respondents were provided registration instructions on a Primary Approach Letter delivered by mail to the selected dwelling approximately one week prior to the commencement of interviewing.

27 All dwellings were enumerated using Computer Assisted Interviewing (CAI).

28 The PES collection methodology differed from the way Census collected its information, where most forms were self-completed. A major advantage of interviewer-administered questionnaires is that people can be provided with assistance if they are uncertain about the meaning of questions, and help is also given to ensure no questions are left unanswered.

29 For extensive explanatory information on the quality of Aboriginal and Torres Strait Islander births data please see Births, Australia (cat. no. 3301.0).

30 The ABS Birth Registrations collection includes all births that occurred and were registered in Australia, including births to mothers whose place of usual residence was overseas.

31 The scope of the statistics includes:

• all births that were live born and were not previously registered
• births that occurred within Australian Territorial waters
• births to temporary visitors to Australia
• births that occurred in Australian Antarctic Territories and other external territories
• births occurred on Norfolk Island from 1 July 2016. This is due to the introduction of the Norfolk Island Legislation Amendment Act 2015. Norfolk Island births are included in statistics for 'Other Territories' as well as totals for all of Australia. Births occurring on Norfolk Island prior to 1 July 2016 were not in scope.
• births that occurred in transit (i.e. on ships or planes) if registered in the Australian state or territory of 'next port of call'
• births to Australian nationals employed overseas at Australian legations and consular offices (i.e. children born overseas to Australian diplomats or their families); and
• births that occurred in earlier years that have not been previously registered (late registrations)
   

The scope of the statistics excludes:

• still births/fetal deaths (these are accounted for in perinatal death statistics published in Causes of Death, Australia (cat. no. 3303.0), and previously, Perinatal Deaths, Australia (cat. no. 3304.0)
• adoptions, sex changes, legitimations and corrections; and
• births to foreign diplomatic staff in Australia
   

32 The ABS Birth Registrations collection records a birth as being an Aboriginal and Torres Strait Islander birth where at least one parent reported themselves as being an Aboriginal person, Torres Strait Islander, or both on the birth registration form. Therefore, Aboriginal and Torres Strait Islander births may be attributed to either:

• Aboriginal and/or Torres Strait Islander mothers, including births where both the mother and father are Aboriginal and/or Torres Strait Islander Australians; or
• Aboriginal and/or Torres Strait Islander fathers and non-Indigenous mothers.
   

33 From 2006 to 2013, the Indigenous status of the mother and father for births registered in New South Wales was inconsistent with other jurisdictions. Specifically, where one parent was an Aboriginal and/or Torres Strait Islander Australian, the other parent was processed as either 'Aboriginal and Torres Strait Islander' or 'Not stated'. Furthermore, where one parent was 'Non-Indigenous', the other parent was processed as 'Non-Indigenous' or 'Not stated'. The Indigenous status of the child, where the birth was registered in New South Wales, was derived from the Indigenous status of either of the parents. The Indigenous status of births registered in New South Wales from 2006 to 2013 should therefore be interpreted with caution. The New South Wales Registry has since changed their processing rules and registration data since 2014 is consistent with the rest of Australia.

34 The ABS has investigated recent annual variability in the number of births of Aboriginal and Torres Strait Islander peoples. In 2016, 108 births were to mothers who reported themselves as being an Aboriginal and/or Torres Strait Islander Australian (0.8% of children to Aboriginal and/or Torres Strait Islander mothers), and who were born overseas. Since 2009, these instances have accounted for less than 1.3% of annual births to Aboriginal and/or Torres Strait Islander mothers. Of these births in 2016, 58% were to mothers born in neighbouring countries in the Pacific, such as New Zealand, Samoa and Papua New Guinea, and of these the majority (63 births) were to mothers who reported themselves as being Torres Strait Islander, or both Aboriginal and Torres Strait Islander on the birth registration form. This may indicate that these mothers are seeking to report their Pacific Islander ancestry (for which provision is not made in the Birth Registration Statement forms in Australia) rather than reporting to be Torres Strait Islanders. However, the six births to mothers born in Papua New Guinea in 2016 who reported being Torres Strait Islanders may be correctly recorded.

35 For extensive explanatory information on the quality of Aboriginal and Torres Strait Islander deaths data please see Deaths, Australia, 2017 (cat no. 3302.0)

36 The ABS Death Registrations collection includes all deaths that occurred and were registered in Australia, including deaths of persons whose place of usual residence was overseas. Deaths of Australian residents that occurred outside Australia may be registered by individual Registrars, but are not included in ABS death statistics. However, deaths of identified Australian diplomats while overseas are included.

37 The scope of the statistics includes:

• all deaths being registered for the first time
• deaths of temporary visitors to Australia (including visitors from Norfolk Island)
• deaths that occurred within Australian Territorial waters
• deaths that occurred in Australian Antarctic Territories or other external territories
• deaths occurred on Norfolk Island from 1 July 2016 are included in this publication for the first time. This is due to the introduction of the Norfolk Island Legislation Amendment Act 2015. Norfolk Island deaths are included in statistics for "Other Territories" as well as totals for all of Australia. Deaths occurring on Norfolk Island prior to 1 July 2016 were not in scope.
• deaths that occurred in transit (i.e. on ships or planes) if registered in the Australian state or territory of 'next port of call';
• deaths of Australian nationals employed overseas at Australian legations and consular offices (i.e. deaths of Australian diplomats while overseas) where they are able to be identified; and
• deaths that occurred in earlier years that have not previously been registered (late registrations).

The scope of the statistics excludes:

• still births/fetal deaths (these are included in perinatal death statistics published in Causes of Death, Australia, (cat. no. 3303.0), and previously, Perinatal Deaths, Australia, (cat. no. 3304.0)
• repatriation of human remains of those whose death occurred overseas; and
• deaths of foreign diplomatic staff in Australia (where able to be identified).
   

38 The ABS Death Registrations collection identifies a death as Aboriginal and Torres Strait Islander where the deceased is recorded as Aboriginal, Torres Strait Islander, or both on the Death Registration Form (DRF). The Indigenous status is also derived from the Medical Certificate of Cause of Death (MCCD) for South Australia, Western Australia, Tasmania, the Northern Territory and the Australian Capital Territory from 2007. For 2015 data, the Queensland Registry of Births, Deaths and Marriages also used MCCD information for the first time to derive Indigenous status. This resulted in a noticeable decrease in the number of deaths for which the Indigenous status was 'not stated' and an increase in the number of deaths identified as Aboriginal and Torres Strait Islander in Queensland. For New South Wales and Victoria, the Indigenous status of the deceased is derived from the DRF only. If the Indigenous status reported in the DRF does not agree with that in the MCCD, an identification from either source that the deceased was an Aboriginal and/or Torres Strait Islander person is given preference over non-Indigenous.

39 While it is considered likely that most deaths of Aboriginal and Torres Strait Islander Australians are registered, a proportion of these deaths are not reported as such by the family, health worker or funeral director during the death registration process. That is, whilst data is provided to the ABS for the Indigenous status question (99.4% of all deaths registered in 2016), there are concerns regarding the accuracy of the data. The funeral director may not always directly ask the Indigenous status question of the deceased's relatives and friends.

40 In addition to those deaths recorded as either Aboriginal and Torres Strait Islander Australian or non-Indigenous, a number of deaths occur each year where Indigenous status is not stated on the death registration form. In 2016, there were 901 deaths registered in Australia for whom Indigenous status was not stated, representing 0.6% of all deaths registered.

41 Data may therefore underestimate the level of Aboriginal and Torres Strait Islander deaths and mortality in Australia. Lags in registrations may also affect the reliability of measures of Aboriginal and Torres Strait Islander mortality. Caution should be exercised when interpreting data for Aboriginal and Torres Strait Islander Australians presented in this release, especially with regard to year-to-year changes.

42 Estimated Resident Population (ERP) is the official measure of the population of Australia. For extensive information on ERP, please see Estimates of Aboriginal and Torres Strait Islander Australians (cat. no. 3238.0.55.001.0).

43 The ERP is based on the concept of usual residence. It refers to all people, regardless of nationality, citizenship or legal status, who usually live in Australia, with the exception of foreign diplomatic personnel and their families. It includes usual residents who are overseas for less than 12 months. It excludes overseas visitors who are in Australia for less than 12 months. In the compilation of the ERP, important adjustments are made to the Census count based on place of usual residence.

44 Estimated resident populations by Indigenous status are compiled using Census, Post Enumeration Survey (PES) and other demographic information. Starting with Census counts by place of usual residence, a number of steps are involved. These include:

• Imputation of Indigenous status for Census records with unknown Indigenous status (as a result of either non-response to the Aboriginal and Torres Strait Islander origin question in the Census, or unknown Indigenous status on Census records imputed by the Australian Bureau of Statistics (ABS) when a form could not be obtained from occupied dwellings identified in the field);
• An allowance for net Census undercount: in a complex exercise such as the Census, it is inevitable that some people will be missed and some will be included more than once. The PES, conducted shortly after Census night, collects information about where people were on Census night and their characteristics to estimate net Census undercount;
• An estimate of the number of Australian residents temporarily overseas at the time of the Census;
• Backdating from the Census date of 9 August 2011 to the ERP reference date of 30 June 2011 using data on births, deaths, and interstate and overseas migration for the intervening period; and
• Minor demographic adjustments designed to address any anomalies in age and sex composition.
   

45 For further information, see Technical Note: Estimated Resident Aboriginal and Torres Strait Islander Australian Population - Method of Calculation.

46 The Australian Census Longitudinal Dataset (ACLD) uses data from the Census of Population and Housing to build a rich longitudinal picture of Australian society. The ACLD can uncover new insights into the dynamics and transitions that drive social and economic change over time, and how these vary for diverse population groups and geographies.

47 For extensive explanatory information on the methodology and quality of the ACLD please see Information Paper: Australian Census Longitudinal Dataset, Methodology and Quality Assessment (cat. no. 2080.5).

49 The proportions of the population shown in the tables in this publication include 'not stated' responses in the denominator unless otherwise annotated. For example, the proportion of people in Australia who are of Aboriginal and/or Torres Strait Islander origin is calculated by dividing the number of persons identified as Aboriginal and/or Torres Strait Islander origin by the total population in the Census, and expressing the result as a percentage. The 'total population' includes records coded to 'not stated'.

50 The Census counts presented in this publication are based on usual residence. For information on usual residence and place of enumeration refer to the Census Dictionary (cat. no. 2901.0).

51 Various analyses in this publication attempt to track age cohorts over time to assess whether the changes that occurred in each cohort were in line with expectations. This is generally done by following selected age groups of Aboriginal and Torres Strait Islander people forward in time (for example, from 2011 forward to 2016) using age cohorts. An age cohort is a group of people with the same birth year within a defined period (for example, people aged 30-34 years in 2011 were aged 25-29 years in 2006).

52 Natural changes in geographical classifications can impact on the increase or decrease in the Aboriginal and Torres Strait Islander population and should be considered when interpreting results in this publication.

53 Census data for geographic area in which a person usually lived five years ago is coded to Statistical Area 2 (SA2) level based on the response to the question of where the person lived five years ago. An image of this question is provided below. If Place of Usual Residence Five Years Ago cannot be coded to the SA2 level, it is assigned to a 'capital city undefined' category, or a 'state undefined' category. Where place of usual residence five years ago data has been used to understand migration between Remoteness regions or Indigenous Regions data has been allocated at the SA2 level based on the population distribution of the Aboriginal and Torres Strait Islander population in the Census.

54 The base geography used to construct the five Remoteness Areas in the 2016 Census was different to that used in the 2006 and 2011 Census; however, the impact of the change in classification is considered to be minimal. In most cases it is possible to make a comparison of the same Remoteness Area across the two Censuses. For more information about the classification refer to Statistical Geography Fact Sheet: Remoteness Structure. For further information about the changes in Remoteness Areas between 2011 and 2016 refer to Australian Statistical Geography Standard (ASGS): Volume 5 - Remoteness Structure, July 2016 (cat. no. 1270.0.55.005).

55 Under the Census and Statistics Act (1905) it is an offence to release any information collected under the Act that is likely to enable identification of any particular individual or organisation. The Australian Bureau of Statistics (ABS) is committed to protecting the personal information it collects. Not only does the ABS have strong legislative protections that safeguard the privacy of an individual's information, we have a proud 100 year history of maintaining community trust in the way it collects, uses, discloses and stores your personal information collected in the Census.

56 The ABS has developed a technique to adjust counts to maintain confidentiality of information. This technique, known as perturbation, is applied to all counts, including totals, to prevent any identifiable data about individuals being released. These adjustments result in small introduced random errors and can mean that the rows and columns of a table do not sum to the displayed totals. However, the confidentiality technique is applied in a controlled manner that ensures the information value of the table as a whole is not significantly affected. Further information on the methodology of perturbation can be found in Confidentialising Tabular Output to Protect Against Differencing paper.

57 Perturbation can be a source of frustration to users because rows and columns do not add to totals, but this technique is implemented to protect personal information. Most tables reporting basic statistics will not show significant discrepancies due to perturbation. However, as the degree of complexity of a table increases, the need for perturbation remains and it will continue to be used in the release of 2016 Census data.

58 For 2006 and 2011 Census data, an additional 'additivity step' was applied that made further small adjustments to each table to ensure rows and columns added to totals. This extra adjustment meant that comparisons between tables which contained similar data items had minor discrepancies. In addition, as the tables at different geographic levels are adjusted independently, tables at the higher geographic level may not be equal to the sum of the tables for the component geographic units. Because of these inconsistencies, for 2016 Census data this additivity step has been removed.

60 If it were possible to achieve complete coverage in Census enumeration, and consistency in the reporting and recording of each person's Indigenous status, change in Census counts between the 2011 and 2016 Censuses would be entirely attributable to births, deaths and migration combined across each age cohort.

61 In reality there will always be an element of the total change in counts between Censuses that cannot be explained by births, deaths and migration. This includes change that occurs as a result of individuals identifying as an Aboriginal and/or Torres Strait Islander person in one Census but not in another, errors in the measurement of births, deaths and migration and changes in Census coverage, response and methodology.

62 In order to measure the change in Census counts between 2011 and 2016 that occurs as a result of individuals identifying as an Aboriginal and/or Torres Strait Islander person in one Census but not in another, the framework applies the net effects of births, deaths and overseas migration to each age cohort to the 2011 Census count of Aboriginal and Torres Strait Islander people to produce an expected 2016 count. The expected 2016 count is then compared with the actual 2016 Census count.

63 The methodology can be represented as:

64 Analysis in this publication also presents an age cohort analysis of change not explained by demographic factors. This analysis uses the same framework described above, but the calculation is completed for each age cohort.

Aboriginal and Torres Strait Islander status is derived from the ABS standard question for Indigenous status (SIQ). The SIQ was first introduced in 1996 and has remained unchanged since this time. Between 1971 and 1981, Indigenous status was reported as 'Racial origin'. In 1986, Indigenous status was reported as 'Aboriginal origin'. When the SIQ was introduced in 1996, people of both Aboriginal and Torres Strait Islander origin could identify as such for the very first time.

Where a person’s Indigenous status is not provided or cannot be determined, they are considered as having an unknown or not stated Indigenous status. Changing numbers of persons with an unknown Indigenous status may impact the extent of the recorded movement in counts of Aboriginal and Torres Strait Islander people between Census years.

The proportion of people with an unknown Indigenous status rose from 1,058,586 persons (4.9%) in 2011 to 1,411,491 persons (6.0%) in 2016 – an increase of 352,905 people (33.3%). This coincides with a small fall in the overall Census response rate from 96.5% in 2011 to 95.1% in 2016.

The distribution of persons with an unknown Indigenous status in 2016 was fairly similar to 2011. The Northern Territory (10.2%) had the highest proportion of persons with an unknown Indigenous status followed by Western Australia and Queensland (both 6.5%).

The biggest percentage point increase in the proportion of persons with an unknown Indigenous status between 2011 and 2016 was in Tasmania (2.2 percentage points).

In Outer Regional Australia, there was a 3 percentage point increase in the proportion of persons with an unknown Indigenous status between 2011 and 2016 (from 5.6% to 8.5%). Meanwhile, Remote Australia had the highest proportion of persons with an unknown Indigenous status in 2011 (8.5% or 25,520 persons) and again in 2016 (11.1% or 31,413 persons). This was more than double the proportion in Major Cities (5.4% or 895,790 persons).

Indigenous Regions (IREGs) that fall either fully or partly in Regional and Remote Australia recorded some of the highest proportions of persons with an unknown Indigenous status. Excluding Other Territories, the IREGs with the highest proportion of persons with an unknown Indigenous status in 2016 were:

• South Hedland (14.5%)
• Katherine (13.5%)
• Alice Springs (11.1%)
   

a. Usual residence Census counts. Excludes overseas visitors.
b. Includes No usual address, Migratory-Offshore-Shipping and Other Territories (not represented on the map).
c. Please note in 2016, Other Territories includes Norfolk Island.

Note: Totals and components may not be consistent within and between tables due to introduced random error to protect confidentiality of Census respondents - see Census Dictionary, 2016 (cat. no. 2901.0).
Source: Australian Census of Population and Housing, 2016

In the Census, there are two scenarios that result in a person’s Aboriginal and/or Torres Strait Islander status being unknown:

1. The respondent chose not to answer the SIQ. These persons are classed as ‘responding’ or ‘not imputed’ persons.
2. No Census form is received from a dwelling (private or non-private). These persons are classed as ‘not responding’ or ‘imputed’. For these persons, Indigenous status is coded to 'not stated’ and data for key demographic variables - age, sex, marital status and place of usual residence - is imputed.

Scenario 1 represents 1.0% of all persons responding to the 2016 Census, down from 1.4% in 2011. The proportion of responding persons with no answer to the Indigenous status question fell across all states and territories.

Persons who chose not to answer the SIQ on all types of Census forms accounted for 16.1% of all persons with an unknown Indigenous status in 2016. Tasmania had the biggest proportion (21.9%), closely followed by South Australia (21.0%) and Victoria (17.6%).

Scenario 2 accounted for the majority (83.9%) of all persons with an unknown Indigenous status in 2016. The Northern Territory (89.3%), the Australian Capital Territory (87.7%) and Western Australia (87.4%) had the highest proportion of persons with Indigenous status unknown as a result of no Census form being received.

There were 227,589 persons who returned a Census form, but chose not to answer the Aboriginal and Torres Strait Islander status question. As it is not possible for a person to return a blank form and still be counted as a responding person in the Census, it is possible to investigate responses to other questions on the Census form to provide further information on persons in this group.

Analysis of these records reveals limited information about the Indigenous status of these persons or any other detail that could inform future Census enumeration strategies. It is also complicated by the higher likelihood of persons in this group to return only partially completed forms. In the 2016 Census, persons who did not respond to the SIQ were far less likely to provide their sex, age and/or marital status than those who did. Most people with an unknown Indigenous status did however provide information about their state or territory of usual residence (94.6%).

In the 2016 Census, the proportion of responding persons with an unknown Indigenous status was highest for people aged 65 years and over (3.9%) and those aged 0-14 years (3.3%). This pattern is similar to the 2011 Census and is consistent across remoteness areas.

Major Cities had the lowest proportion of responding persons with an unknown Indigenous status (0.9%) followed by Inner Regional and Very Remote areas (1.3% respectively). Outer Regional and Remote areas had the highest rate of responding persons with an unknown Indigenous status (1.5% respectively).

a. Usual residence Census counts. Excludes overseas visitors.
b. Includes No usual address, Migratory-Offshore-Shipping and Other Territories (not represented on the map).

Note: Totals and components may not be consistent within and between tables due to introduced random error to protect confidentiality of Census respondents - see Census Dictionary, 2016 (cat. no. 2901.0).
Source: Australian Census of Population and Housing, 2016

In discrete Aboriginal and Torres Strait communities, the Census is primarily collected by an interviewer using a tailored Interviewer Household Form (see Census of Population and Housing: Understanding the Census and Census Data, Australia (cat. no. 2900.0)). Very few people counted this way in the 2016 Census had an unknown Indigenous status (0.4%) – on par with 2011 (0.5%).

Nationally, non-response to the SIQ for responding persons in 2016 was slightly higher for females (51.3%) than males (48.7%). This was true for all jurisdictions except the Northern Territory (52.7% for males and 47.6% for females). The distribution of persons who did not respond to the SIQ followed a similar pattern to the overall sex distribution for each jurisdiction.

If a person does not identify as an Aboriginal and/or Torres Strait Islander in the SIQ, they will be counted as a non-Indigenous person or coded as not stated (depending on their response). This will be the case even if they report they have Aboriginal or Torres Strait Islander ancestry in the ancestry questions. Of all persons with an unknown Indigenous status in 2016, 754 (0.1%) persons reported Australian Aboriginal or Torres Strait Islander ancestry. There were however 3,224 persons who identified as non-Indigenous but recorded their ancestry as Australian Aboriginal or Torres Strait Islander – more persons than in 2006 (2,716) and 2011 (2,046).

There were 40,735 children aged 0-14 with an unknown Indigenous status who were living in couple and lone parent/guardian families in 2016. Over three-quarters of these children were in families where one or both parents/guardians were non-Indigenous (75.8% or 30,877 children).

Only a small proportion of children with an unknown Indigenous status were in families where one or both parents/guardians were Aboriginal and/or Torres Strait Islander (4.0% or 1,648 children).

Children with an unknown Indigenous status living in couple families where one parent/guardian also had an unknown Indigenous status were much more likely to have a parent/guardian who identified as non-Indigenous (5.3%) than a parent/guardian who identified as Aboriginal and/or Torres Strait Islander (0.2%). These observations are consistent with other findings in this analysis.

There were a number of children aged 0-14 years old with an unknown Indigenous status with parents/guardians who also had an unknown Indigenous status (8,214 children or 20.2%). In these instances, there is very little to no information available from the Census that could indicate what the Indigenous status of these children may be. For detailed analysis of identification patterns within families, see Changing propensity to identify.

Analysis of responding persons with an unknown Indigenous status does not provide any further clarity around what the Indigenous status of these persons might be. At the National level, the characteristics of persons with an unknown Indigenous status appear different to those who identified as either Aboriginal and/or Torres Strait Islander or non-Indigenous.

Overall response to the 2016 Census varied across Australia. As expected, IREGs with a low dwelling response rate had a higher proportion of records with an unknown Indigenous status than IREGs with a high dwelling response rate.

The Census Post Enumeration Survey (PES) is run shortly after each Census as a way to independently measure how well the Census has counted the Australian population. The PES estimates how many people should have been counted in the Census, how many were missed (undercount) and how many were counted more than once or in error (overcount).

The Census count includes:

• All responding persons who were counted on at least one Census form, with multiple counts for persons who were included on more than one form
• Persons who have been imputed into non-responding Census private dwellings during data processing
• Imputed records for persons who spent Census night at a non-private dwelling (e.g. a hotel or caravan park) but did not complete a Census form there.

All people present in Australia on Census night, with the exception of foreign diplomats and their families, should have been included on a Census form at the place where they stayed. However, due to the size and complexity of the Census, it is inevitable that some people will be missed and some people will be counted more.

Some of the reasons people are missed in the Census include:

• They were travelling and difficult to contact.
• They mistakenly thought they had been counted elsewhere.
• The person completing the form mistakenly thought that certain people (e.g. young babies, the elderly or visitors) should not be included.
• A Census form or online Census code was not delivered to the address or the dwelling was mistakenly recorded as being empty (unoccupied) on Census night.

Some of the reasons why people may have been counted more than once include:

• They were included on the Census form at the dwelling where they usually live, even though they stayed and completed a form elsewhere on Census night.
• They moved during the Census period and completed forms at both their new and old address.
• They intentionally or unintentionally completed more than one form, e.g. an electronic form and a paper form.

The PES collects information on the characteristics of respondents including Aboriginal and Torres Strait Islander status, which allows the ABS to understand more about people who did not respond to the Census and those who were overcounted.

Using the information provided by the PES, the ABS produces the net undercount for each Census. Net undercount for any category of person is the difference between the actual Census count and the PES population estimate (how many people should have been counted in the Census).

Net undercount is typically presented as a rate. The rate is the net undercount as a percentage of the PES estimate for a given population (i.e. as a percentage of the number of people who should have been counted in the Census). The Census is typically associated with a net undercount. Rates of net undercount or overcount can vary significantly for different population groups depending on factors such as sex, age, Indigenous status, and geographic location, and on whether these characteristics have changed or been misclassified between the Census and PES.

The total net undercount rate for Australia in 2016 was 1.0%, which was lower than in 2011 (1.7%).

The 2016 PES estimated that the net undercount rate for Aboriginal and Torres Strait Islander persons was 17.5%, equivalent to 137,750 persons. This is only slightly higher than the 2011 rate (17.2% or 114,188 persons).

Western Australia had the highest net undercount rate in 2016 (23.0%) whilst in New South Wales, the net undercount rate was similar to the national net undercount rate (17.3% compared to 17.5%). In the Australian Capital Territory (ACT), the net undercount rate was the smallest – 12.1% however this is offset by a comparatively large standard error (7.3%).

As discussed in Technical Note 1, respondents can choose not to answer the Indigenous Status question in the Census. Where no answer is provided, the Census does not impute for the missing response (which is also the case for imputed persons). While the person (real or imputed) will continue to be counted in broad-level Census counts, they will not be included in the Census counts for Indigenous status.

When calculating net undercount for Aboriginal and Torres Strait Islander persons, these ‘not stated’ responses are treated as people missed from the Census for this population and therefore form part of the undercount estimate.

Occasionally, the answers obtained for a person in the PES are not consistent with how they completed their Census form. This is expressed as the net difference in classification. There are a number of reasons why responses can differ:

• A person may have difficulty answering a question for themselves or another household member, either in the Census or in the PES.
• A person may interpret the question differently in the Census, where the majority of forms are self-enumerated (completed by the respondent unassisted), compared with the PES, which is administered by a trained interviewer.
• Different people in the household may provide the Census and PES responses.
• The Census may contain a not stated or imputed response while PES has a valid response.

The following analysis is based on unweighted PES counts. For further information, see Technical Note 1: Differences in Classification Between the PES and Census (ABS cat. no. 2940.0).

When the PES sample was matched to the 2016 Census, 94.6% of records that identified as Aboriginal and/or Torres Strait Islander in the 2016 Census also did so in the PES. A small number of PES records (282 or 4.6% of the PES sample) identified as non-Indigenous in the 2016 Census and as Aboriginal and/or Torres Strait Islander in the PES. There were a small number (381) of records that had identified as Aboriginal and/or Torres Strait Islander in the 2016 Census but non-Indigenous in the PES.

There were 848 PES records that had an unknown Indigenous status in the 2016 Census. In the PES, the majority of these records (803 records) identified as non-Indigenous – not unexpected given that non-Indigenous persons made up the majority of the Australian population in 2016. This suggests that Aboriginal and Torres Strait Islander people are not disproportionately represented in records with an unknown Indigenous status.

The Census and the net undercount (together with other administrative data) are used by the ABS in the compilation of Australia’s official population estimates known as the Estimated Resident Population (ERP) (cat. no. 3238.0.55.001).The ERP is also used as a benchmark for ABS population surveys that take place between Censuses such as the National Aboriginal and Torres Strait Islander Health Survey. The net undercount is therefore an essential component of all statistics collected by the ABS for the Aboriginal and Torres Strait Islander population.

Further detail on the 2016 undercount and overcount is available in Census of Population and Housing: Details of Overcount and Undercount, Australia (ABS cat. no. 2940.0).

The Australian Census Longitudinal Dataset (ACLD) was first released following the 2011 Census. The ACLD is a longitudinal probabilistically and deterministically-linked sample of Census records. At the time of this publication’s development two ACLD datasets were available, one linking a sample from the 2006 and 2011 Censuses and the other linking a sample from the 2011 and 2016 Censuses. A file linking a 5% sample across all three of these Censuses is planned for the future.

The ACLD provides a longitudinal picture of Australian society as captured by the Census. This includes allowing for analysis of persons who changed how they answered the Aboriginal and/or Torres Strait Islander Origin question between Census years. In total, the 2011-2016 ACLD contains 927,520 records that were linked between 2011 and 2016. Of these, 18,515 individuals identified as being Aboriginal, 1,174 as Torres Strait Islander only, and a further 802 as both Aboriginal and Torres Strait Islander. Of the rest of the sample, 9,948 did not have their Indigenous status stated, with the remaining 897,076 identified as being non-Indigenous.

Despite being the largest longitudinal dataset available, the sample of Aboriginal and Torres Strait Islander people in the ACLD is relatively small. This is due to a combination of the small proportion of the total population that is Aboriginal and Torres Strait Islander (2.9% in the ACLD) and of the lower linkage rates for Aboriginal and Torres Strait Islander persons. This is also an effect of the ACLD sample being for 5% of all Australians. Consequentially, the analysis presented below uses weighted counts instead of sample counts to ensure the data is as representative as possible. For more information on Aboriginal and Torres Strait Islander persons in the ACLD please refer to Information Paper: Australian Census Longitudinal Dataset, Methodology and Quality Assessment (cat. no. 2080.5).

Further detail on the ACLD is available in Information Paper: Australian Census Longitudinal Dataset, Methodology and Quality Assessment 2006-2011 (ABS cat. no. 2080.5) and Information Paper: Australian Census Longitudinal Dataset, Methodology and Quality Assessment 2011-2016 (cat. no. 2080.5).