Delivering the 2021 Census

The ABS conducts the Australian Census of Population and Housing every five years. It provides a snapshot of who we are on Census day and shows us how we are changing over time. It counts every person and home in Australia and is the only source of information about small geographic areas and small population groups across the whole country.

The information we collect in the Census helps estimate Australia's population, which is then used to:

• set electoral boundaries
• determine the number of seats in the Australian House of Representatives allocated to each state and territory
• calculate how government funds are distributed
• inform planning for services and infrastructure. 

Governments, local councils, academic institutions, and community groups use Census data to help inform important decisions about planning for:

• schools
• health care facilities
• transport
• local services. 

The Census is compulsory for everyone in Australia, except diplomats and their families. This includes people in Norfolk Island, the Territories of Cocos (Keeling) Islands and Christmas Island. It also includes international visitors and students.

For the 2016 Census, we changed the Census collection model to work on an address list basis. Before 2016, field staff were required to find all the dwellings in an area. However, as our national residential address lists are accurate and detailed, we knew we could trust them for coverage of most of the country. The new model was an adaption of similar models used for the United Kingdom and Canadian censuses.

To maximise the public’s ability to participate in the Census, we divided the country into three collection areas, each with their own strategy for public engagement and collection of forms:

1. Mail-out areas
2. Drop-off areas
3. Remote areas.

The mail-out areas had known addresses to which we mailed a Census Instruction Letter (CIL). Each CIL had a unique login code for respondents to use as well as information on how to request a paper form. In 2021, this included 84% of all dwellings, which were predominantly in urban areas and can be seen on the purple areas of the above map.

Drop-off areas were mostly rural areas and unmailable urban areas, particularly those where there had been significant new growth. Census field staff delivered paper forms to dwellings in these areas which, in 2021, comprised 14–15% of all dwellings.

Finally, we had remote areas, which included most of the Northern Territory and some large parts of Western Australia, South Australia and Queensland. In remote areas, our field staff often needed to travel long distances and, to avoid having to return, delivered materials and collected responses at the same time.

While the majority of Australians were staying in private dwellings on Census day, we also needed to find and count people in a range of other types of dwellings and more transient locations. There is a large variety of these dwellings and each type needs a tailored strategy. The strategy makes sure we have engaged with the dwellings before the Census and arrangements put in place so every person in them can fill out forms on Census day.

A number of these establishments are called non-private dwellings and include:

• retirement villages
• aged care facilities
• boarding houses
• hospitals
• hotels
• mining sites
• military facilities
• corrective facilities and prisons.

Other types include groups of private dwellings that are in one location. These are called private dwelling establishments, and include the dwellings (sometimes temporary such as tents) in:

• caravan parks
• camping grounds
• marinas.

People who are experiencing homelessness on Census night generally need additional support to participate and include those who are:

• sleeping rough
• in supported or crisis accommodation
• in a boarding house
• staying in temporary accommodation (like a motel)
• staying temporarily with friends or family.

Other areas that need extra attention for collection are:

• Aboriginal and Torres Strait Islander communities
• holiday areas.

The methods for counting people in these living arrangements vary but often involve training people who work or live at these establishments, or in the local community, to hand out and collect materials. In some cases, they may also help respondents to complete their Census forms.

Protecting respondents’ privacy and security is one of our highest priorities. Under the Census and Statistics Act 1905, the ABS is not allowed to release personal information in a way that can identify a person or business. No individual or organisation can access any personal information that respondents include on their Census form. This includes government departments and direct marketing companies.

We use the strongest encryption technology available to make sure information from the online Census form is secure. If a respondent uses a paper form, we supply them with a reply-paid envelope. Any member of a household (including a visitor) could ask for their own individual login number for the online form, or a separate paper form and envelope, if they wanted to fill in the form privately.

The 2021 Census was the first time we had a dedicated privacy team to work with Census areas to ensure privacy was embedded into the design and operation of the Census. Security and Privacy by design has more information on how privacy and security were central to the 2021 Census.

To reach every person who is in Australia on Census day, and ensure we count every dwelling, the ABS needs at least six years for planning and coordination. The diagram below shows the stages involved in moving towards operations and preparedness for the 2021 Census.

Preparations for the 2021 Census start before the last one has been conducted and in the year after the 2016 Census, initiation for the 2021 Census program was well underway.

In order to ensure systems and processes are robust, we undertake testing, and in the lead up to the 2021 Census we ran two major tests. In 2019, we engaged 40,000 dwellings in Queensland and New South Wales to test our new:

• field management structure
• recruitment process
• Census questions.

In 2020, we engaged 100,000 dwellings in parts of New South Wales, the Australian Capital Territory, South Australia, Western Australia and the Northern Territory to test our:

• COVID strategy
• digital service
• processing and coding systems
• Remote Area Strategy.

The Census collection period starts at the beginning of a Census year with the recruitment of engagement staff, who contact communities and larger establishments such as hospitals and hotels. This preparation is important for raising awareness of, and sharing procedures for, the upcoming Census. In July a communications campaign begins, a contact centre opens and a website is launched. In the lead up to Census day, typically one to two weeks before, materials are sent to households through the post or, where mail is less reliable, delivered by field staff. These materials explain how to participate and encourage people to respond either online or using a paper form.

After Census day, field staff visit dwellings that have not responded to remind them to complete the Census. Field staff also visit establishments (like hotels) to collect any forms left there.

After the Collection period, we spend 10 months processing the responses before delivering the data. The 2021 Census data is being delivered in a phased approach with three releases: the first on 28 June 2022, the second on 12 October 2022, and the third in March 2023.

The ABS employed 35,000 temporary and ongoing staff to work on the 2021 Census. This included:

• 33,000 temporary field staff across the country
• 400 temporary field support staff centrally managing the field operations
• 800 temporary staff within our Data Capture Centre and data operations centre to process all the paper forms and code all the responses
• 800 ABS office staff including those in the Census Program, and in technology, human resources and office support.

This number does not include some large workforces involved in the Census through our external vendors including:

• Australia Post to deliver and return letters and paper forms
• establishment officers (up to 10,000) to deliver and collect forms at large establishments like hospitals and hotels
• logistics providers such as printing, materials handling and transport companies
• human resources suppliers working to recruit and pay our temporary staff
• the Services Australia Contact Centre (peaking at 3,500 operators) to receive calls from the Australian public
• our major IT partners – PwC Australia and Amazon Web Services
• another 40 suppliers and their staff.

The scale of everything in the Australian Census is large, including the workforce, the materials and the engagement with everyone living in Australia on Census day. Below are a few examples.

The 2016 Census provided many important insights and lessons. A number of reviews were conducted into the 2016 Census, primarily because of the high-profile 40-hour website outage, but there were also other concerns about privacy and communication. These reviews included a report the Australian Government commissioned by Alastair MacGibbon, Review of the events surrounding the 2016 eCensus, and a review through the Australian Senate, 2016 Census: Issues of trust. We also commissioned our own review into the data quality of the 2016 Census by the Independent Assurance Panel, Report on the quality of 2016 Census data.

We combined the recommendations from these reviews with the findings of our own internal evaluation and a review of other international census experiences. This resulted in the following key lessons on how we should prepare for, and conduct the 2021 Census:

• Improve risk management and risk mitigation.
• Seek more external expertise and assurance.
• Be cautious of past successes.
• Plan for things to go wrong.
• Be attuned to changing contexts and be respondent focused.
• Improve communication and engagement.
• Have trust in the public to do the right thing.
• Understand that perceptions are as important as reality.
• Plan for the possible, not just the expected.

You can also read our actions from the MacGibbon recommendations on the ABS website.

We were keen to meet public expectations for a modern experience in conducting the 2021 Census. We wanted to provide respondents with a simple, streamlined service and remove any barriers they had to completing a Census form.

The stories in this report describe some of our major initiatives and how we succeeded in implementing what we learned from 2016. The initial Census data was released on 28 June 2022 and we can proudly claim we:

• achieved a very high response rate
• achieved a high-quality statistical outcome
• provided an improved experience for the public and our staff compared to 2016.

Public expectations of government services are changing rapidly. The public expects interactions to be simple, clear, fast and available 24 hours a day.

The approach to the 2016 Census was transformative and despite the 40 hour outage of the online form, delivered a sustainable digital-first model for future Censuses. Filling in the online form was an easy way for most respondents to participate in the Census and reduced the overall time needed to do so. However, in a number of ways, it did not fully meet all user expectations.

For the 2021 Census, the ABS took a user-centred design approach to building a digital service, not just a website. As part of this, we followed the Digital Transformation Agency Digital Service Standard. This user focus was central to the success of the Census and the delivery of an easy, simple and secure modern experience to the Australian public.

The ABS undertook research with Meld Studios to get an insight into the motivators and barriers to Census completion. This research included looking back on the 2016 experience and mapping the user journey to understand the ideal end-to-end experience for the public.

Barriers to completion were also explored. Previously, these barriers often meant people needed to engage with our call centre or wait for our visiting field staff or, in the worst case for us, did not respond to the Census. The barriers we looked at included when people:

• did not receive or lost their Census number or form
• had a disability that made it more difficult to complete a form
• had poor English literacy
• had poor digital literacy or digital coverage
• lacked understanding of the Census
• perceived they were not included or not required to participate (for example, international students and travellers)
• were not at home, on holidays or in hospital.

The journey mapping identified the principles for designing the 2021 experience so that it delivered a service that met user needs and expectations:

• Focus on value – Identify, translate and celebrate the value the Census brings to individuals and Australia.
• Set up for trust and deliver on that promise – Inspire and reinforce public confidence in the Census.
• Design for motivations, capabilities and needs – Build an evolving human-centred understanding of our respondents.
• Engage through all channels – Proactively connect with our public, when and where they need us.
• Design a contemporary government experience – Reframe the way we work, focusing on the Census as a government service, rather than as a set of distinct product teams.

When designing a ‘usable service’, it is important to not assume anything and to conduct extensive testing. Some of our usability and accessibility testing to improve the public’s experience of the Census included:

• cognitive testing – observing participants undertaking tasks within the digital service and then interviewing them after their experience
• a usability test involving 500 professional testers and 500 volunteers from the ABS
• accessibility testing by Intopia, experts in digital accessibility
• two large-scale tests with the public in 40,000 households in 2019 and 100,000 households in 2020.

Several innovations were implemented to improve the user experience and support respondents to complete their Census as effortlessly as possible, and through their channel of choice.

The online form

• Designed for a contemporary experience, the form had simple login features, navigation and help. All aspects of the service were responsive to device type and screen size, from large desktops to mobile phone screens, and smoothly moved between display types. The form had built-in smart features to minimise the time needed to complete it. This included sequencing to only show relevant questions and autofill responses to subsequent related questions based on previous responses. The form also enabled respondents to save their form and complete it later.

A dedicated website

This was launched earlier than previous Censuses, in April 2021. It provided:

• ‘how to’ information and access to self-service forms
• information about how to answer each question and why it was asked
• advertising campaign material
• information for Census supporters and a map showing where to get help in person
• translated information, Auslan videos and audio explanations.

The website had 19 million unique visitors and 195 million page views.

Online self-service options

These improved beyond the basic ‘contact us’ forms provided in 2016, and were more tailored to respondents’ needs. They allowed people to easily tell us that their dwelling was unoccupied on Census day, or that they needed a paper form or Census number. The self-service options also included a chatbot called Claire with 1,800 questions and answers to help respondents find the information they needed and keep them in their preferred online channel. More than 250,000 respondents used the self-service forms and more than 376,000 conversations were completed through Claire.

A ‘no Census number’ pathway

This enabled respondents to complete their online Census if they had not received a login code or they had lost their Census letter. They could request a code after providing their address and mobile phone number. This self-service pathway received a very high uptake, with 1.75 million requests for a code, and reduced the need for respondents to interact with the Contact Centre or a field staff member. The ‘no Census number’ pathway proved invaluable and popular, accounting for 20% of all online responses.

An extended response window

This allowed people to complete their Census early, at a time convenient to them. The 2021 Census communication campaign encouraged this approach. The approach contributed to very high response rates early in the Collection period and a relatively lower peak on the Census Digital Service and Census Contact Centre on Census day.

The Census Contact Centre

Run by Services Australia call centre team on behalf of the ABS, this was open seven days a week, 8am – 8pm, and focused on resolving queries on the first call. A great deal of work was done on scripting so the information given to respondents was clear and anticipated any further questions they might have. On-hold messages were developed so that respondents could hear relevant information such as how they could self-serve online while waiting for their call to be answered. The Contact Centre answered 645,833 calls between 5 July and 1 October, significantly fewer than expected. The Contact Centre did not need to implement any congestion management strategies during the entire period. The 2021 Census night was the first Australian Census that did not see the Contact Centre overwhelmed by high volumes of callers.

An escalations team

This was established to respond more quickly to difficult queries or complaints. Fifty-three staff managed and responded to 40,000 escalations and 6,000 contact us forms. They also managed correspondence that came in via:

• the ABS website
• the ABS privacy team
• the Census Data Capture Centre
• other channels.

This team made 30,000 calls to respondents, including 650 assistance calls where they completed the respondents’ Census form with them over the phone. Part of the escalation team looked after the Parliamentary Hotline, which responded to more than 400 phone and email enquiries made by Members of Parliament to support their constituents.

A 24-hour phone automated Paper Form Request Service

This was established for respondents who had their Census number and wished to order a paper Census form. In 2021, for the first time, this service could also be used to order large print or braille forms. This service processed 400,000 requests for paper forms.

Social media to answer respondent enquiries

The 2021 Census social media team responded to 85,000 social media comments from the public.

The increased focus on user-centred design was highly successful in removing barriers respondents faced in completing the form or informing us of their circumstances. The statistics below displays some of the evidence of respondents' improved experience:

• There was an increase in online uptake – from 63% in 2016 to almost 80% in 2021.
• 34% of respondents completed their form before Census day, compared to 14% in 2016.
• 63% of respondents completed their form on or before Census day, compared to 45% in 2016.
• Calls to the Contact Centre requesting Census numbers dropped from 500,000 calls in 2016 to 56,000 in 2021.
• There were fewer general contact us online requests, from 402,000 in 2016 to 31,500.
• 1.5 million online forms were completed through the no Census number pathway, 97% of which were done through the digital self-service.
• The how to fill in a Census form videos were played more than 13,000 times and the videos were produced in 35 different languages.

There was also a reduction in total calls to the Contact Centre. In 2016, there were 3 million calls, only 1.3 million of which were answered. In 2021, there were 645,000 calls and we answered 88% of these within 300 seconds. This exceeded the service level agreement we had with Services Australia.

All online form respondents were asked for feedback after submitting their form. 69% of those that responded said their experience was “good”, 27% were “neutral” and only 4% thought it was “bad”.

Most feedback reflected positive sentiments similar to these:

• "The interface was really good and easy to navigate. It was really easy to complete. Well done. It didn’t take as long as I thought."
• "Form was very easy to navigate, even from a phone."

We also engaged Meld Studios to conduct a respondent experience evaluation after the Collection phase of the 2021 Census. Meld interviewed people who recently completed their 2021 Census online or by paper form to understand their overall experience. Participants described a positive experience that was straightforward and quicker than expected. They particularly appreciated the time-saving features of the digital service and being able to complete the form before Census day.

The 2021 Census was the most accessible Australian Census ever. Our work in this area was recognised at the 2021 Australian Access Awards, where the ABS and the 2021 Census won ‘Government Website of the Year’ and was a finalist in the ‘Accessibility Initiative of the Year’ category. The Census Digital Service was heralded as ‘world class’ for its support of accessibility for people with disabilities through its thorough application of the Web Content Accessibility Guidelines (WCAG) 2.0 AA.

Stewart Hay, Intopia co-founder and managing director, said, "The 2021 Census is a good example of a government putting citizens first and making sure as many people as possible are included in an important service. Nothing is necessarily always perfect, but the ABS definitely prioritised accessibility".

The website was designed and tested to work across desktop and mobile platforms, with a variety of common assistive technologies such as screen readers, keyboard navigation, screen magnification and voice recognition.

A range of information and support was available, including for people in the deaf or hard of hearing and blind or low vision communities:

• 33 downloadable audio files and 66 Auslan videos providing information on every Census question, including why we ask it and frequently asked questions
• closed captioning and transcripts for all video content on the Census website
• braille forms type 1 and 2, and large print forms available to order through multiple channels.
• the Census Contact Centre was available through the National Relay Service for support and assistance 
• a simple English (Easy Read) guide to the Census.

"The 2021 Census was an overall fantastic experience. It had 66 Australian Sign Language (Auslan) videos accompanying the online form, including one for each question. As a deaf person, English is my second language and having the online videos on the Census was a huge help enabling me to answer the questions appropriately." – Olivia Beasley, Community Engagement Manager, Expression Australia.

A national mass media campaign is an essential component of every Census, and this was more important than ever given the global pandemic environment in 2021. The objective of the campaign was to increase awareness of the Census and highlight its value to encourage informed and willing participation, therefore supporting the target 95% response rate.

The campaign was built on a long history of trust and participation by the Australian public in the Census. Further to this developmental research informed the campaign which aimed to educate people about how Census data is used to inform services for individuals, families and communities.

The advertising tagline, ‘Every stat tells a story’, was based on the idea that the Census is not just numbers, but what those numbers tell us to inform important decisions. Advertising agency BMF was engaged to develop the advertising creative.

Phases of our advertising and communication activities were closely aligned with the operational activity. The campaign focused on positively influencing people to complete the Census and contributed to a higher than forecast response by Census night and a high online take up at nearly 80%.

The ABS worked closely with the Service Delivery and Coordination Committee of Cabinet, with support from the Cabinet Advice Branch at the Department of Finance, to build flexibility into the Census campaign and receive the necessary government endorsement.

A phased advertising campaign ran between 4 July and 19 September 2021, and was coordinated with public relations, stakeholder engagement, and operational communications which ran between June 2020 and October 2021.

The 2021 Census communication campaign was rolled out in five phases:

• Phase one, establishing value (June 2020 – June 2021)
• Phase two, persuasion (4 – 27 July 2021)
• Phase three, completion (28 July – 12 August)
• Phase four, reminder (13 August – 19 September)
• Phase five, thank you (20 – 26 September)

Communication agencies 33 Creative and Etcom were engaged to adapt the mainstream campaign for Aboriginal and Torres Strait Islander peoples and CALD audiences respectively. Advertising was translated into 48 languages, including 19 Aboriginal and Torres Strait Islander languages.

The ABS engaged Universal McCann to manage the advertising placement across channels, including television, cinema, newspapers, radio, outdoor locations (buses and billboards), digital and social channels.

Our Story. Our Future.

The materials and resources for the Aboriginal and Torres Strait Islander audience featured the artwork ‘Our Story. Our Future.’

‘Our Story. Our Future.’ Was created by proud Wiradjuri, Wotjobaluk, Yuin and Gumbaynggirr artist Luke Penrith and Maluililgal people, Badu Island artist Naseli Tamwoy. It tells the story of how the ABS works with communities to see, hear and acknowledge Aboriginal and Torres Strait Islander peoples, cultures and experiences in our national story. 

The artwork celebrates our stories being seen and heard for the benefit of future generations. It embraces the cultural importance of storytelling and information sharing, entwined with a focus on working together for a strong future for children, family and community. 

The use of colours and inclusion of country, land and sea, rivers and desert, reflect the diversity of our peoples and culture across the country.

The 2021 Census campaign increased awareness of the Census.

• In 2021, 88% overall had heard something about the Census, awareness peaked at 98% between 9-15 August.
• In 2016, 76% overall had heard something about the Census (compared to 75% in 2011), awareness peaked at 86% around Census night.

The ABS commissioned Hall and Partners to undertake benchmark, tracking and evaluation research of the campaign. The phased advertising approach was highly successful, with strong and accurate message cut through. Those who had seen the campaign were generally better informed about the Census, held more positive attitudes and perceptions, and were more likely to have taken key actions such as completing the Census.

In other awareness raising activities:

• the media strategy included successfully delivering a Census press conference, hosted by the Hon Michael Sukkar MP, then ABS Minister, that increased awareness and coverage of the Census.
• a national spokesperson conducted over 200 media interviews during Census operations.
• supporting ABS spokespeople included 21 Aboriginal and/or Torres Strait Islander people, and 15 multilingual spokespeople speaking 17 different languages.

The proactive and strategic media engagement resulted in 41,038 media clips from 1 February to 19 September 2021 with a cumulative audience reach of 1.6 billion. Media sentiment across the Census coverage was 93% positive, trending positive or neutral.

More than 60 case studies were created and shared through media channels to highlight how Census data is used to inform services for communities.

Social media channels were a key component of communication and customer service support for the 2021 Census and for the first time we answered queries by a dedicated Social Media team. We resolved 85,000 engagements this way.

The social media strategy, including posts sponsored by the ABS, facilitated growth of the 2021 Census Facebook page from zero to 25,842 followers in six months. Misinformation, disinformation and malinformation were monitored through ’social listening’ to quickly address incorrect information about the Census.

After Census night we used several social media posts to prompt action in slow to respond areas. We observed there was a low response in some holiday areas, so we developed this post to ask people who had empty second homes to tell us their dwellings were unoccupied. This paid social media activity led to improvements in response, with spikes in submissions of ‘report not at home’ occurring after each post.

The ABS conducted research to inform the 2021 campaign and understand how communication and engagement activities could increase participation. It included audience segmentation research by Meld Studios and developmental research about attitudes, motivations and barriers to completion by Kantar Public.

The Census audience segments Meld Studios identified were:

• Experienced Ed – Motivated to complete but needs to know the Census is easy and secure to participate.
• Driven Dahna – Needs to know how to participate and that their information is secure.
• Obligated Olive – Understands the benefits of the Census to Australia and feels a sense of obligation to complete.
• Prisoner Peta – The fact that the Census is compulsory is their main motivation to complete it.
• Recalcitrant Robin – They see the Census as a waste of time and need to understand its importance.
• Non-completer Nell – Does not know much about the Census. They are motivated by being told it is easy and compulsory to participate.

The research informed messaging and strategy in developing the advertising campaign and reaching diverse audiences. The research informed three guiding principles to encourage people to participate in the Census including:

• educate people on the importance of the Census and how data is used
• highlight the benefit of completing the Census
• humanise and personalise the benefit, showing both individual and collective outcomes.

Contact materials such as Census Instruction Letters and envelopes were designed and tested with the Behavioural Economics Team of the Australian Government. Materials were designed and tested to encourage immediate action. Important elements of the materials included:

• an envelope that people would identify as important and requiring action
• messaging that the Census is compulsory
• wording that encouraged immediate action, including completing the form before Census night
• simple instructions on how to participate.

Testing included variations of materials provided to 40,000 respondents in the 2019 Census Test. This showed a greatly improved response from relatively minor changes to the instruction wording on both letters and envelopes. As an example, the messaging that the Census is compulsory drove action from people who were hard to engage or non-completers. It also made already willing participants feel more positive about completing the form. The wording on the envelope prompted more people to open it when it arrived.

A critical part of any Census is to make sure that everyone who is in Australia on Census day is counted. For the 2021 Census, we had a team to design inclusive strategies for groups who faced barriers to taking part in the Census. The team developed these strategies through a combination of:

• lessons learned from past Census experiences
• information gathered from engaging with stakeholders
• user-centred design activities.

The strategies focused on supporting the following population groups:

• Aboriginal and Torres Strait Islander peoples across urban, regional and remote areas of Australia
• people from CALD communities, including international students
• people experiencing homelessness
• people who prefer to get Census assistance in person.

Each strategy provided different types of support for each community, depending on their needs. The support could include tailored communication, extra engagement within the community and direct support, including providing multilingual staff where needed.

We employed nearly 60 Census Engagement Managers in early 2021, who were assigned one of these population groups for a specific region. Their role was to enact the strategy for that population through extensive engagement with the respective communities. They included establishing contacts for the Census Collection period, promoting inclusion in the Census, emphasising its importance and providing guidance on where to find help for these communities.

We had a strong focus on increasing opportunities for Aboriginal and Torres Strait Islander people to influence the way in which the Census was conducted and communicated.

This was a major recommendation from the Report on the quality of 2016 Census Data:

‘The ABS should consider methods to improve the enumeration of Aboriginal and Torres Strait Islander people, explored in consultation with Aboriginal and Torres Strait Islander communities.’

In response to the recommendation, all development and testing activities to support Aboriginal and Torres Strait Islanders included Aboriginal and Torres Strait Islander voices:

• at the ABS Round Table on Aboriginal and Torres Strait Islander statistics
• in user-centred design research
• in testing the Census Digital Service and form content
• in testing the communications campaign.

The ABS has a permanent team focused on engagement with Aboriginal and Torres Strait Islander peoples called the Centre for Aboriginal and Torres Strait Islander Statistics. This team employs a number of Aboriginal and Torres Strait Islander people and played a pivotal role in the 2021 Census by building on established relationships and expanding the team to conduct extensive outreach with community and organisations across the country.

Field engagement staff, many who identified as Aboriginal and Torres Strait Islander, were recruited to work with Aboriginal and Torres Strait Islander communities. They moved beyond a typical transactional engagement to promote the shared benefits of the Census and encourage community-led action. A number of these people also volunteered to be trained as official Census spokespeople for media interviews.

Specific recruitment strategies were established to help us increase the proportion of Aboriginal and Torres Strait Islander people employed in the Census. This resulted in 6.4% of our Census field workforce identifying as Aboriginal and Torres Strait Islander.

The activities to support participation by Aboriginal and Torres Strait Islander communities in the Census included:

• engagement with 1,826 organisations
• training of 24 Aboriginal and Torres Strait Islander staff members as media spokespeople
• recruitment of Census engagement staff directly from Aboriginal and Torres Strait Islander communities
• recruitment of 1,530 community field staff to conduct interviews to fill in especially designed forms (the Interviewer Household Form).

The national Culturally and Linguistically Diverse (CALD) Strategy was designed to ensure all CALD population groups were counted in the Census and that the ABS tackled barriers to their participation. We identified which population groups needed support through extensive engagement with CALD stakeholders. Targeted communications and support, including face-to-face help events, were then developed to reach culturally diverse population groups.

The ABS formed a close relationship with the Federation of Ethnic Community Councils of Australia to ensure the Census was as inclusive as possible for CALD communities.

To help encourage CALD communities to participate in the Census there were:

• 900 CALD organisations engaged
• 15 ABS media spokespeople speaking 17 different languages
• 340 CALD fill in the form sessions
• 400 Census engagement staff recruited from diverse communities
• 25% of 33,000 field staff self-identifying as speaking a language other than English
• social media tiles in 29 languages shared directly with communities
• 17,000 calls made to the Translating and Interpreting Service
• 4,000 views of an Easy Read Guide to the Census.

Homelessness takes many forms including:

• staying in temporary or supported accommodation such as boarding houses, hostels and shelters
• sleeping rough
• staying temporarily with family and friends (couch surfing)
• living in severely crowded dwellings.

We designed our Homelessness Strategy to ensure all people experiencing homelessness could participate in the Census. Engaging and collaborating effectively with the homelessness sector was critical to improving participation and the quality of responses.

We created strong relationships with a range of stakeholders through the Homelessness Strategy, which involved:

• all levels of government
• peak homelessness bodies
• a wide range of service providers.

One crucial element of this engagement was ensuring all organisations across the sector understood how the Census produces homelessness estimates and the strong link between estimates and funding for services and programs. It was also important to provide clear direction on how people experiencing homelessness could participate in the Census and respond to key questions, like the ‘usual address’ question. Organisations were then able to use their networks and relationships with their clients to share this information and help people to take part.

Through engagement with state and territory governments, we were also able to get administrative data to help us plan and ensure we correctly classified the different types of homelessness.

Our tactics to reach people experiencing homelessness varied depending on where they were staying. For example, we:

• employed people who work in homelessness shelters to help their residents to participate
• had roving teams that offered interviews to rough sleepers on a special short form
• worked with targeted accommodation providers such as hotels, motels and caravan parks
• had a youth ambassador program that targeted schools to explain to students experiencing homelessness why they should participate in the Census and how they could do so.

We employed almost 600 engagement and field staff for the Homelessness Strategy. Many of these were from the homelessness sector or had previous experience working in the homelessness sector.

We also produced a range of homelessness communications materials, such as posters, fact sheets, videos and social media tiles. An important aspect of these was to educate people experiencing homelessness on how to accurately fill in the ‘usual address’ question so that their situation was accurately recorded.

We actively adapted our homelessness approach to suit local circumstances including lockdowns. This included provision of alternative online and telephone support, mail-out of additional forms and extending the period during which we provided support for communities in some areas.

In some areas, the COVID-19 pandemic changed where and how people were sleeping rough. The ABS worked with state and local governments and the homelessness sector in areas impacted by COVID-19 to develop the safest and most appropriate response.

A new service for the public this Census was pop-up hubs. More than 300 hubs across Australia were introduced to provide accessible, face-to-face support options to members of the public. They were set up in both urban and regional locations across Australia in high traffic areas such as shopping centres and supermarkets. The hubs provided a friendly and supportive environment to encourage people to participate in the Census. Members of the public could have their Census questions answered and some were helped to complete their form on the spot.

While a wide demographic visited the hubs, they were particularly popular with people on holiday, older people and people with disabilities.

The ABS also offered more than 100 fill in the form sessions at community centres like libraries. At these sessions members of the public could receive one-on-one support to complete the Census. The sessions attracted a range of people including older Australians, people with disabilities, people with lower levels of literacy and members of the CALD community.

Ensuring the security and privacy of Census data was vital in running the Census and maintaining the trust of the Australian public. For the 2021 Census, the ABS took a ‘security and privacy by design’ approach which permeated every level of the Census Program. All decisions on system architecture and design were made with security in mind. Security experts were involved in all major technology decisions as well as in testing and monitoring processes and systems. 

Cyber security challenges have changed significantly in the last 10 years. All government agencies have experienced considerable increases in the frequency and sophistication of attempted cyber-attacks on their IT environments. Some of these realised threats have resulted in security incidents that affected the delivery of government services.

Australia’s Cyber Security Strategy 2020 outlines:

'Australians are increasingly reliant on the internet and the internet-connected devices we use daily. The digital economy is the future of Australia’s economy. Cyber security needs to be a fundamental and integrated part of everyday life, enabling Australians to reap the benefits of the internet safely and with confidence.'

The 2016 Census experience

In 2016, the online Census form was targeted on Census day by a series of cyber-attacks known as Distributed Denial of Service (DDoS) attacks. This resulted in the ABS taking down the service for 40 hours. This outage disrupted many Australians attempting to complete their Census, affected the ABS’s reputation and undermined the Australian government’s digital agenda. Three separate reviews were conducted after this outage. Many of the recommendations that came out of the reviews were about security and privacy.

‘Australia now knows that cyber security is not just about national security. Cyber security is about availability of services and confidence in government in a digital age. And the public’s confidence in the ability of government to deliver took a serious blow, more so than any previous IT failure.’ – extract from the Review of the events surrounding the 2016 Census.

The Census Digital Service and external partnerships

To ensure the security and success of the 2021 Census, the ABS worked with public and private sector experts to design and implement a completely new online platform. The Census Digital Service (CDS) was built by PwC Australia in partnership with Amazon Web Services. Our aim was to deliver a completely new, innovative, high performance and secure solution on the Amazon Cloud.

We engaged early in development with the Australian Cyber Security Centre and the Digital Transformation Agency.

The Australian Cyber Security Centre played a major role in ensuring the security of the CDS by:

• reviewing the architecture and design
• reviewing system configurations
• conducting IT source code reviews
• performing penetration testing – simulating cyber-attacks to check for weaknesses.

It also supplied extra cyber security protection and detection capabilities for other ABS IT systems. The Australian Cyber Security Centre provided a highly skilled and dedicated extension to the ABS IT Security and Security Operations Centre teams.

The CDS was developed with end-to-end data encryption. This means that no public provided data was unencrypted at any time when stored in files or in transit, keeping that data secure at all times until it reached the ABS. All services used were accredited to ‘Protected’ status by an independent security assessor accredited by the Australian Signals Directorate.

During the Collection period when the Census Digital Service was live, our processes repelled almost 1 billion attempted cyber-attacks, and we also blocked 130,000 malicious IP (network) addresses.

Security testing 

Although the CDS was one of the keys to the success of the 2021 Census, there were many other IT systems that supported collection, processing and dissemination for the Census.

We ran security risk assessments during the collection and processing period on more than 48 individual IT systems. We assessed many twice, once for the operational readiness exercise in October 2020 and again for the Census itself. We conducted more than 111 security risk assessments for the 2021 Census. Cyber security personnel worked closely with each of the technical teams to make sure they kept security and privacy in focus throughout the design, delivery and operation of all IT systems.

We did extensive independent testing and assurance to ensure we delivered a safe and secure Census. Specialist cyber security firms Cyconsol, North, Cyber CX and Red Wolf helped us in this work. We conducted these tests at regular intervals and in line with major system releases.

The tests included:

• nine major source code security reviews
• more than 20 large scale penetration testing exercises against Census systems to make sure the systems and data were secure
• four DDoS tests that finished in the ABS performing the largest DDoS test ever conducted in Australia
• many performance and load tests, up to more than double the expected peaks and for extended time periods
• 20 independent security risk assessments to cover all our Census systems. These were conducted by certified third party assessors through the Australian Cyber Security Centre Information Security Registered Assessors Program.

Operations security monitoring

Throughout the Collection period of the Census we ran a Security Operations Centre (24 hours, seven days a week) to monitor all Census systems with real-time information and alerts of potential security issues. During Census week, we had Australian Cyber Security Centre personnel working with us onsite for security operations and potential incident response. PWC monitored the Census Digital Service alongside their 3rd party providers throughout the entire collection period.

2022 IPAA Spirit of Service Award for Collaboration

The ABS and Australian Signals Directorate were awarded the 2022 IPAA Spirit of Service Award for Collaboration. Our submission for the 2021 Census Digital Service – Building trust and partnerships to achieve excellence in cyber security was shortlisted out of 46 nominations from across the Australian and ACT public sector.

The award recognised our achievements and significant collaboration with the Australian Cyber Security Centre and over 50 other public and private sector organisations.

In awarding us, the judges commented they were impressed by the way we had incorporated recommendations from 2016, and the robust planning and testing undertaken ahead of the 2021 Census.

Protecting peoples’ privacy and keeping their information secure is a key principle of the Census and the ABS. The Census and Statistics Act 1905 has provisions for protection of personal information. Under this Act, the ABS cannot release personal information in a way that can identify a person or business. This includes not releasing any personally identified data to other government agencies.

It also means that all our staff who collect, process, analyse and share Census data are legally bound to protect Census information. You can find more information about how we protect private information in the 2021 Census Privacy Statement.

Separation principle and data retention

An important aspect of our privacy position is how we collect, store and process Census data, particularly the separation of names and addresses from other personal and household information. This ‘separation principle’ makes sure no one can view a name or address with other Census data nor view names with their addresses. For this Census we will keep names for 18 months and addresses for up to three years.

Any member of a household (including a visitor) could also choose to fill in the Census form privately by asking for their own individual login number for the online form, or a separate paper form and reply-paid envelope.

Privacy impact assessments

The Australian Senate Inquiry, 2016 Census: Issues of trust, recommended the following:

The committee recommends that all future privacy impact assessments relating to the census, are conducted externally with the final report published on the ABS website 12 months in advance of the census to which it relates.

We selected independent privacy experts Galexia to conduct a privacy impact assessment for the 2021 Census. They identified and evaluated matters that could impact privacy at every stage of the Census and recommended ways to manage, minimise or remove these impacts.

IIS Partners also produced a second privacy impact assessment – the 2021 Census Administrative Data Privacy Impact Assessment. This assessment looked at privacy implications for how the ABS used information collected by other government agencies, businesses or organisations. This information, called administrative data, was used to reduce the Census costs and improve Census operations and data quality.

Both Census privacy impact assessments are published in the ABS Privacy PIA register.

Engagement

The ABS engaged the privacy community throughout the lead up to the 2021 Census. We held numerous meetings with the Australian Privacy Foundation and regular discussions and correspondence with the Office of the Australian Information Commissioner. We also met with state and territory Privacy Information Commissions. Their support reduced the concerns of the public and led to less negative press on privacy this Census. For instance, in 2016 there was a large-scale privacy campaign, centred around the ABS decision to keep names and addresses.

Privacy strategy and roles

We contracted privacy and security consultants IIS Partners in June 2018 to support our privacy work. This included guidance to help build trust and social licence for the Census. Social licence is the level of acceptance or approval that the community has for a project or organisation.

We formed a Census Privacy team in June 2018 and employed a dedicated Census privacy officer. The privacy team liaised with all Census teams to promote privacy discussions and provide privacy advice.

A separate Census Privacy Working Group was also created, including members external to the Census and the ABS. The group helped identify privacy risks, promote communication on privacy and review recommendations from the privacy impact assessments.

All staff employed on the Census attended Privacy by Design training by Salinger Privacy and completed the annual ABS privacy training. This ensured staff were aware of privacy issues as well as their privacy related responsibilities.

Census archive

Census respondents can choose to have their Census information archived. For those who take this option, we will transfer Census information to the National Archives of Australia as part of the Census Time Capsule, where it will be preserved for 99 years. For the 2021 Census, this information will not be made available for any purpose until 2120 and cannot be accessed, altered or retrieved before that time. In 2021, 61% of people gave us permission to archive their response. This compares to 50% in the 2016 Census.

For the 2021 Census, the ABS took a large step forward in its journey to move more of its services and systems into the cloud. This started with the Census Digital Service and also included:

• the Operations Insights platform
• the Call Centre Agent tool for use by Services Australia and ABS staff
• the Census Knowledge Base
• the Paper Form Request Service
• the MyWork mobile app for field staff
• Claire, the chatbot
• online training courses
• the field staff payroll solution
• Dynatrace (system monitoring software).

From August to November 2017, the ABS successfully made its first real foray into using cloud services with the Australian Marriage Law Postal Survey. During this process we got to experience the scale and security available in the cloud. We built an online form in a cloud platform primarily for Australians on the electoral roll living overseas. While this only accounted for 34,447 responses, it was tested on millions of transactions in an hour, easily proving the benefits of cloud services for large volume IT solutions. They are far more scalable and when well designed, can be more cost effective than traditional local hosted environments.

Extensive security testing and validation by IT security experts, including the Australian Cyber Security Centre, demonstrated that cloud-based solutions are both viable and preferable, especially from a security perspective. They are now ABS’ preferred solution for all future ICT developments.

When we created the requirements for the 2021 Census Digital Service (CDS), we chose a cloud based solution despite there being very few Australian government cloud-based services at that time.

ABS contracted PwC Australia after an open tender in May 2019 to deliver the CDS in partnership with Amazon Web Services and hosted on the Amazon cloud platform in Australia. They were able to show how a system could be delivered that was highly secure, reliable and could perform at the scale needed. We had a target of 350 form submissions per second at peak or 1.26 million responses per hour on Census day.

The system was built primarily using Amazon’s “native technologies” that are specifically designed for their cloud services, including Lambda, Dynamo DB and S3 buckets. Amazon Web Services provided architectural design input, conducted ‘well architected reviews’ and a reliability review.

The system integrated with third party services that we used for the Census including Google’s ReCaptcha, Geoscape’s Address Look-Up system and Clevertar’s Census guided chatbot, Claire.

The CDS included:

• the online form
• the Census website
• a content management system
• self-service facilities such as ‘not-at-home’ reporting, request a new Census number and paper form requests.

"A key focus of the 2021 Census was to ensure the digital service had redundancy, performance and protection against cyber threats." -  Gwil Davies, Partner, Digital Innovation and Cloud Engineering, PwC Australia.

Our next large cloud development was the Operations Insights platform, also hosted on the Amazon cloud. ABS staff, Amazon Web Services, and their partners Shine and the ARQ Group, worked together to create this first cloud-based data lake for the ABS to store data on how the Census was progressing.

This system ingested operations data from multiple data sources and produced approximately 100 reports showing progress of Census operations and enabling us to make numerous critical, real-time decisions.

The volume of data, complexity of the data transformations, and volume of reports made it an ideal candidate to be cloud hosted. See Story 8 - Monitoring operations during the Census for more detail on the use of this service.

While cloud systems can be scalable and secure, they need to be designed and built with that in mind. They have complexities that need to be clearly understood. To maximise success, partnering with experts across a range of disciplines was essential. It is important to begin with the end in mind in terms of performance, scale and security.

Independent testing by third party vendors and other government agencies (the Australian Cyber Security Centre in particular) was critical for reassuring us we had the right solution.

Our use of other fully established cloud services such as Geoscape’s Address Lookup, Google’s Captcha and Amazon Web Services Connect, provided confidence that the system could handle the load and performance required.

Delivering high-quality data and managing risks to data quality was an important consideration throughout the Census. While most of the teams in the Census are established to ensure a smooth running Census operation, there were complementary teams that focused on ensuring the quality of Census data. This story summarises the aspects of their work.

Producing high-quality data was one of the three 2021 interdependent Census objectives.

The 2021 Census Data Quality Plan helped us make sure the Census produced high-quality data with three main goals:

1. Innovate Census processes to improve data quality.
2. Manage risk to data by implementing strong risk controls and quality gates.
3. Understand and manage the impact on Census data from changes in the Census.

The Census shows long-term trends in important aspects of the lives of Australians. We repeat topics (a grouping of questions) in each Census to collect this valuable trend data. However it is also important to make sure Census data stays relevant by adding, removing and changing topics when needed.

Following public consultation, the ABS added two new topics to the 2021 Census, the first additions since the 2006 Census:

• long-term health conditions
• service with the Australian Defence Force.

Changes to existing questions were also made to keep the Census relevant and improve the quality of the data. These included:

• removing the question on internet access in people’s homes
• adding a non-binary option to the sex question
• enhancing response options for the language and ancestry questions for Aboriginal and Torres Strait Islander peoples

The ABS ran public consultation on topics in the 2021 Census between 3 April and 30 June 2018 with results published in November 2018. More than 450 submissions were received through the ABS Consultation Hub from a range of sectors including all levels of government, academia, community and advocacy groups, industry bodies, businesses and individuals.

The ABS received 36 suggestions for new topics along with suggestions for changes or additions to the 37 existing topics. From this list we shortlisted eight topics to consider further:

• Aboriginal and Torres Strait Islander cultural identity
• current or previous service in the Australian Defence Force
• journey to place of education (including mode of transport and name and address of educational institution)
• long-term health conditions
• more contemporary measures of household and family relationships, including shared care of children where parents live apart
• non-binary sex and/or gender identity
• sexual orientation
• smoking status.

We then assessed the shortlist according to:

• strength of the need for the data
• the cost and feasibility of collecting that data
• respondents’ ability to answer
• statistical impact the topic has on the quality of Census data.

Recommendations on the topics to be included in the 2021 Census were put to the Australian Government in mid-2019. After the Government decided on the 2021 Census topics, an amendment to the Census and Statistics Regulation 2016 was tabled in both houses of Parliament in February 2020.

We used qualitative testing to assess the ability of respondents to answer the questions. Testing included:

• focus groups with key populations to look at respondents’ understanding of central concepts
• cognitive testing by observing people undertaking tasks and then interviewing them after their experience. This is to assess whether respondents understood and were accepting of the new questions
• cognitive testing in Aboriginal and Torres Strait Islander communities with a focus on the specially designed Interviewer Household Form for Indigenous communities. This was tested in Alice Springs, Yarrabah, Torres Strait and Tiwi Islands
• testing of the Special Short Form for people experiencing homelessness including online interviews with the homelessness sector
• a substantial field test in October 2019 with 40,000 dwellings to provide quantitative assessment of the possible new topics.

Census content has to be finalised more than a year in advance of the Census to allow for the paper forms to be printed and for testing of the online form. This meant we had limited time to make changes in response to the pandemic. Some small changes were made to the workplace address question just before we finalised the form to make it clearer for respondents working from home. For the online form, additional instructions were added to help people living in lockdown in Sydney and Melbourne.

The 2021 Census quality gates have been a critical control for managing Census risks. They are a set of measures that each aspect of the Census needed to pass before they could “go live”. They involved rigorous documentation of all processes, systems, operational procedures and the relevant testing results. The documentation was then peer and senior staff reviewed before official sign off.

Quality gates have been a crucial means for assuring Census processes and services are well understood, tested and are ready in time. It was particularly critical that we minimised risks to an acceptable level.

The quality gates for the 2021 Census started in July 2020 with the sign-off of printing the Census paper forms. Quality gates will continue to be processed well into 2022 before the release of all Census products. This will involve the clearance of 107 gates for the entire Census.

The Data Quality Specialist Working Group guided the development of the 2021 Census. The Working Group offered expert statistical advice to the 2021 Census teams on how decisions might affect the accuracy and relevance of the Census data. The group was made up of Senior Executives from across ABS statistical areas and has met quarterly since August 2018. They provide support and data quality advice on a wide range of issues including:

• Census content and forms
• data coding processes
• occupancy determination and imputation of the number of people in non-responding dwellings
• response window (the time between when responses open and when they close).

Story 7 – Excellence in planning and governance also details activities by the 2021 Census Statistical Independent Assurance Panel.

Administrative data is information that government departments, businesses and other organisations collect. The ABS only uses this data for statistical purposes and it is not shared, or released, in a way that could identify someone. For the 2021 Census we used it to improve the quality of the data collected and repaired in the Census. Administrative data was used to:

• help decide areas that need specialised field staff who speak common languages or have similar cultural backgrounds
• help determine whether a non-responding dwelling was unoccupied on Census day
• update our register of addresses
• give us counts of people in larger institutions like prisons and some hospitals
• quality assurance of data processing
• imputation of people in non-responding dwellings.

We commissioned a Privacy Impact Assessment into the potential use of administrative data contained here in the Privacy PIA register (released 21 July 2020). In some cases, it was decided not to use the data on the basis of the assessment’s finding (in particular, see KEY RISK AREA 3 – Electricity usage data).

Success of an endeavour as large and complex as the Census requires a structured and disciplined approach to governance and planning. Milestones, risks, issues and dependencies all need to be managed across the many streams of work. Program and project reporting must drive necessary focus and remedial actions.

With the lessons from 2016 in mind, the 2021 Census team transformed the Census Program and staff mindset in terms of preparation, readiness and adaptability. An important element included changes to governance, risk, issues and crisis management and how independent experts were engaged to test our thinking and systems.

The Census is challenging given its scale and complexity. Decisions or changes in one aspect of the program often affect other parts of the program. As a result, it is vital to have an effective governance structure with strong communication across all levels of the program.

The program adopted the Managing Successful Programmes (MSP) methodology with adaptations to manage the huge scale of the 2021 Census. The program implemented a structured framework in alignment with MSP methodology to manage the complexity and risks inherent to the Census. This included planning and control, risk and issue management, quality and assurance management and program organisation. The program approach was complemented by effective project management arrangements.

For the 2021 Census the ABS implemented a governance model with clear accountability arrangements. The governance model was supported by three layers of assurance to ensure the program met its objectives. Monitoring and control activities increased as the program moved closer to the operations phase. This enabled the ABS to identify areas of concern earlier and apply the necessary attention to them.

Primary oversight for the program was provided by the 2021 Census Executive Board, which had responsibility for overseeing the strategic direction of the Program and making decisions about policy, priorities, and direction. Sub-committees to the 2021 Census Executive Board provided advisory, monitoring and review functions to support the implementation of the program, both within the program structure and across the broader ABS.

The Australian National Audit Office (ANAO) audited the planning of the 2021 Census in 2020 to assess whether we had learned and implemented lessons from the 2016 Census. The audit investigated whether we:

• had established appropriate oversight frameworks for the Census
• were taking appropriate steps in developing IT systems for the Census
• were addressing key Census risks and implementing Census recommendations from previous reviews. These included the 2016 Issues of Trust Senate Inquiry and the MacGibbon Review into the events surrounding the outage of the online Census form.

The audit found our planning was partly effective and we had “established largely appropriate planning and governance arrangements for the Census”. It made seven recommendations related to program oversight and assurance, data quality, privacy, risk and security. We accepted all the recommendations and implemented them before Census day.

Risk management formed an integral part of Census program design and delivery. A comprehensive, but pragmatic approach to risk management ensured we were able to quickly identify, evaluate and respond to risks and issues as they arose.

We continually monitored the external environment to identify any factors that may cause risk. With an evolving COVID-19 environment, the risk management approach served us well.

The main areas of focus to ensure had low or no risk were:

• work, health and safety of staff and members of the public
• compliance with acts and government policies
• statistical data quality
• security and confidentiality
• the public’s trust and our social licence.

In memory of Thérèse Lalor, Director of the 2021 Census Program Management Office, and a highly valued member of the Census Leadership Team.

Thérèse Lalor joined the 2021 Census Program in August 2019 having worked at the ABS for 16 years. 

The success of the 2021 Census was, in no small part, due to Thérèse’s work as Director of the Program Management Office. Thérèse brought colour and movement to the Census Program Management Office where she lifted the capability of the section and the Census leadership team more broadly with her professionalism, talent and expertise.

Thérèse brought her team and the Census Leadership Group along the journey through her cheerful, positive and professional demeanour. She was approachable and had a knack of explaining project management simply. 

Thérèse made an enormous contribution to guiding the 2021 Census with her approach to project management.

Thérèse passed away on 23 February 2022 after helping to deliver a successful 2021 Census. Thérèse is sorely missed and fondly remembered by the Census Leadership team, her colleagues and friends across the Australian Bureau of Statistics.

Throughout the Census cycle, we ensured expert external perspectives were brought to governance and assurance functions.

The ABS engaged professional services firm, KPMG as the whole of program assurer to monitor and report on all Census activities. KPMG’s role was to challenge, broker and review the quality of the Census program against deliverables. This included considering the integrity of reporting and mitigation strategies for emerging risks and issues. In undertaking this role, KPMG used a broad range of specialists from across their company. Taking a program wide view meant they could support dependency management as well as provide assurance in respect of individual streams of work. The independent program assurer reported directly to the Census Senior Responsible Officer and Australian Statistician.

The Census program governance committees included both private and public external board members who had extensive experience delivering large projects. They included senior executives from Telstra, the Australian Taxation Office, Queensland Treasury, the Australian Electoral Commission, IIS Partners and the Digital Transformation Agency.

The ABS has significant expertise in testing systems and processes. However, there are areas where support from experts across a range of testing areas, was required to provide additional capacity and skills. In some cases, we sourced a separate layer of independent testing to that already provided by our partners delivering whole solutions.

We engaged external experts to test areas such as:

• accessibility – Intopia and Vision Australia
• system performance – Ampion (formerly known as Revolution IT)
• device and browser compatibility – Ampion
• security – Cyconsol, North, Cyber CX and Redwolf
• behavioural economics – the Behavioural Economics Team of the Australian Government
• public sentiment – Meld Studios, Kantar Public.

In 2016, we established a Statistical Independent Assurance Panel to review the quality of Census data. The Panel concluded the data was fit-for-purpose and provided recommendations on ways to improve future censuses particularly in how we:

• determine occupancy
• impute for non-response
• determine dwelling structure
• collect Aboriginal and Torres Strait Islander data
• collect in non-private dwellings
• address privacy.

The ABS implemented all their recommendations each of which successfully improved outcomes in the 2021 Census. This panel was reconstituted, with some overlap in membership, for 2021 and has undertaken a similar role to 2016.

The 2021 Census Statistical Independent Assurance Panel findings were released on 28 June 2022 the same day as the 2021 Census data - first release. The Panel concluded that the 2021 Census data is fit-for-purpose, is of comparable quality to the 2011 and 2016 Censuses and can be used with confidence.

An excerpt from the Panel report: “The high response rate for private dwellings (96.1%) was an outstanding achievement given the challenges provided by the pandemic and is comparable to response rates seen in other countries, such as Canada and the United Kingdom, which undertook censuses during the pandemic, and higher than that of New Zealand, where the census was conducted pre-pandemic.”

The 2021 panel identified eight opportunities for enhancing future Censuses, all of which are being actively considered by the 2026 Census team.

Every Census we rely on private partnerships to help us deliver some of our large-scale logistics and specialist operations. Many of these partners have been involved in the Census before and have developed a detailed understanding of critical aspects of our business. This Census we worked with more than 50 private sector providers, including 30 IT vendors.

Helping these businesses to understand our processes and outcomes is critical to delivering a successful Census.

There are some fundamental reasons these partnerships are successful:

1. Carefully chosen – we conduct strictly controlled tenders and do extensive research to ensure the product or service offering will meet Census needs. This ensures there are no surprises during delivery.
2. Both sides have ‘skin in the game’ – vendors need to be committed to the success of the Census. They need to understand our needs and the benefit to their reputation and their products in working with us.
3. We have collaborative relationships – we work closely with our partners and they provide personal support and understand our needs.
4. There is strong governance – this ensures good project management and focus on decisions to ensure delivery of outcomes with well-managed risks.
5. There is coordination and integration – the solution the vendor provides fits well with other Census services.
6. We have regular reviews – this often involves independent reviews of important parts of some contracts, particularly how the contract is working for the ABS and the vendor.

Below are some of the key private partners for the 2021 Census:

• PwC Australia in partnership with Amazon Web Services built the Census Digital Service and Census website.
• Recruitment agency Adecco recruited most of our short-term field and processing staff.
• BMF designed our 'Every stat tells a story' advertising campaign.
• Fuji Xerox were responsible for printing and mail house services.
• Australia Post delivered 8.3 million approach letters, 2.7 million reminder letters and returned nearly two million paper forms to our processing centre.

The Census is a whole of government delivery project. Also critical in our success were the many government partners, notably:

• Services Australia
• Digital Transformation Agency
• Australian Cyber Security Centre
• all state and territory governments.

The issues and crisis management framework and planning for the 2021 Census Program was shaped by reviews into the 2016 Census, as well as by lessons from the 2017 Australian Marriage Law Postal Survey. The Framework was aligned with the ABS Incident Management Framework and based on the Australasian Inter-Service Incident Management System – the nationally recognised system of incident management for Australia’s fire and emergency service agencies.

'Plan for things to go wrong and ensure sufficient investment in issues management and disaster recovery plans and strategies.' – Duncan Young, 2016 Census Program Manager, in the evaluation of the 2016 Census.

‘Crisis plans existed but appear not to have been fully tested, nor were they consulted when most needed.’ - extract from the ‘Review of the events surrounding the 2016 Census’.

The Issues and Crisis Management team assessed, escalated, coordinated, and supported the decision-making process for major incidents and critical events that were raised. For each issue, clearly defined roles and responsibilities were established, including a chairperson, decision maker, issue controller, action owners and note takers.

We worked closely with experts in issues and crisis management, SOCOM, who provided us with important, professional guidance.

In the planning for the Census, operational areas were provided training and prepared their action plans to respond to predicted operational issues. More than 130 action plans were developed for the 2021 Census.

An important aspect of the planning was to routinely conduct simulated scenarios that covered the most likely and most critical things that could go wrong. This was about building ‘muscle memory’ so that staff were prepared, and had time to plan for process improvement, and build team coherence and confidence. These simulation exercises started about three months before the operational readiness exercise in 2020 and were conducted regularly until final Census operations commenced.

‘A significant outage of the Census form’ was simulated in April 2021. The issue was treated as a ‘true event’ by key staff required for that incident, including several major vendors.

In another room the exercise was observed by senior staff from the Minister’s office, the ABS, Department of the Treasury, Prime Minister and Cabinet, Australian Cyber Security Centre, Digital Transformation Agency, Services Australia and KPMG. At the end of the session the observers were asked to suggest improvements to the process.

All were impressed with the thoroughness of the process. Many commented they had not participated in such an exercise before and were keen to conduct similar exercises in their own organisations. All offered their support in dealing with any major incident that might occur during the Census operations.

All left confident that the ABS was far more prepared for many risks than in 2016.

The 2021 Census collection operations period was notable for the absence of any major incident or issue and the strong support received from stakeholders both during and after it.

Altogether:

• 15 major simulation exercises were conducted
• 134 Response Action Plans were prepared across the Census collection operations period
• 17 Level 2 incidents occurred during the Census operations period. All were successfully managed and resolved in a timely manner
• No Level 3 incidents occurred.

Following a Level 2 event, one Census Executive said, ‘The approach was a really good way of making swift decisions and making sure everyone was aware of the changes and decisions’.

‘When an incident or crisis presents itself, it is hard to stay focused. The adrenalin starts pumping and the emotions start to escalate – especially when this experience is new to you and your team.

Operating in this heightened state is not the best way to quickly gain clarity of the situation, determine and implement the best response and try to return to business-as-usual operations having minimised the impact on people and the organisation.

It is important to have a process the team can trust, follow and then practice implementing under stress.

After several training sessions and a few simulation exercises, the Census team began to trust the Assess, Decide, Act process. The results were transformational.

The day before Census 2021, there was a meeting to discuss an actual level 2 incident. Everyone in the room was calm and focused on understanding the situation and determining the outcomes that the Census needed to achieve to address the situation.

Within 20 minutes, the meeting was over and people started to work on the action plans to achieve the outcomes. Well done Census team, it was validation that they had listened, learnt and achieved.’

- David Hawkins, Managing Director of SOCOM

In 2016, the ABS adapted to a new Census collection model that helped pave the way for further operational innovations in 2021. Adapting similar methods used for the United Kingdom and Canadian censuses, the ABS changed to an address list basis in the 2016 Census. Previously we gave field staff a map with their area marked and they had to visit every dwelling they found. This was a far less efficient method before 2016.

The new model allowed us, for the first time, to see real-time response rates across all areas and population types. It allowed us to see potential issues and make decisions to deal with them, while we still had a large field workforce and time to engage the public. Before 2016, some data issues were only found after the forms came back a few months after Census day and were thus very difficult to fix.

It was clear we could do more and better in 2021. A key recommendation from the 2016 Census was to consolidate the operational and management data into a single, unified channel. This initiated the Operations Insights project, in which data from all sources was ingested and transformed into a single, cloud-based data lake.

The aim of the Operations Insights project was that all Census staff could monitor 2021 Census operations with near real-time data and make informed decisions quickly. Some specific aims for the Operations Insights tool were:

• having a single source of truth
• ensuring data transformations (converting the data from one format to another) are traceable – so it was easy to understand where the final values in the reports came from
• having a secure platform with centralised access controls – so that access could be easily granted and removed.

We achieved these aims developing the tool in partnership with Amazon Web Services, Shine Solutions, and ARQ Group.

Our tool had three processes that are typical of a data lake:

1. The ingestion of the raw operational data (most of which was not designed for reporting).
2. Transform (or derive) the data it into a format that can be easily understood for reporting.
3. Visualise and report the data. We used the Microsoft Power BI tool for end users to view these reports themselves.

Throughout the 2021 Census we had more than 100 Power BI reports servicing a wide range of users. The reports were in the form of tables, graphs and maps, and used interactive features such as filters and views. Reports were generally designed to meet one of three reporting needs:

1. Strategic reporting – providing a high-level view of Census operations. These reports informed decisions at a macro level for the Census Program.
2. Tactical reporting – providing short-term analysis that allowed a deep-dive into data for areas and population groups. This provided a detailed understanding of operational metrics and it informed decision making.
3. Operational reporting – enabling operational areas to monitor day-to-day progress at a granular level. This informed decisions about the allocation of resources and helped determine how effective processes were.

The data in most reports refreshed at least once an hour with some refreshing every five minutes. This streamlined approach was a large improvement from the spreadsheets used for monitoring in the 2016 Census, which were circulated once a day and included only a fraction of the breadth and depth of metrics in the 2021 reports.

The new Operations Insights tool for 2021 significantly improved the capability for operations staff to make decisions. It provided them with near real-time data, tailored reports and visualisations.

As explained in About the Census, there are standard processes for the mail-out, drop-off and remote areas strategies. For some areas, though, we need to vary these processes to mitigate potential risks to response rates.

The ABS first introduced these new strategies in 2016 and built on them in 2021. Without them, it is likely some parts of the country would have had a poorer response. We classify these strategies as ‘adaptive variations’ and ‘responsive treatments’.

Adaptive variations are pre-emptive strategies used for areas where it is historically difficult to obtain a high response or that have characteristics such as low internet uptake. For the 2021 Census, these areas were chosen using 2016 response data.

The adaptive variations for 2021 included:

• Paper form on approach – this involved mailing a paper form to respondents instead of the Census Introduction Letter. It was used in areas likely to have lower online uptake. Almost one million dwellings received a paper form on approach.
• Early start to visit – in areas with typically lower compliance and self-response, field staff were employed to commence door knocking visits earlier than in other areas. This typically involves commencing visits a few days after Census day instead of the standard ten days. It meant more visits to dwellings in these areas. In areas with a predicted high volume of empty dwellings, such as holiday areas, it also helped identify occupancy where there was no response. We selected 1.05 million dwellings for early start to visit.

Responsive treatments are interventions that can be applied during the Collection period in response to local issues and when there are risks to response rates and other data quality outcomes.

The responsive treatments for 2021 included:

• Targeted communication – this involved quick and targeted communication to specific areas using a variety of channels, particularly radio and social media adverts. Targeted communication was also used for raising property owners’ awareness that they could self-report vacant dwellings.
• Extended follow-up – some field visits were extended beyond the planned timeframes. This method was used in Sydney and Melbourne due to lockdowns. Regional areas, particularly parts of the Northern Territory also had extended follow up due to the time needed to cover large distances and to counter lower response rates.
• Deploy flying squads – groups of ABS office staff, also known as ‘flying squads’, were despatched to door knock when other options were exhausted. They were deployed throughout the collection period, particularly in areas with field staff recruitment shortages. COVID-19 travel restrictions impacted our ability to readily use flying squads, particularly crossing state borders. This treatment was effective in improving response and other quality outcomes.
• Unaddressed mail service – in a selection of postcodes with staff shortages, and an unreliable address list, we used Australia Post’s unaddressed mail service. Dwellings in these areas received a letter with details on how to complete the Census online well before their paper form was delivered by a Census field staff member.

In the 2021 Census, we made several changes to the way we operated in the field. This led to more efficient processes for managers in the office and an improved experience for field staff and the public.

Field staff operations were previously a hierarchical structure across four levels. Each level recruited and trained teams that reported to them. While this model prior to the 2021 Census was effective for enabling local decision making, managing workloads and follow-up in standard ways across the country was very difficult.

With the new national address list approach described earlier, local decision making was no longer as necessary. In contrast, national decisions and communications became more important, such as recruiting or redirecting staff into lower responding areas or areas with recruitment issues.

The centralised Field Operations Management model was supported by a suite of technology applications.

After Census day, we sent all non-responding dwellings a reminder letter and shortly after that, we visited any that had not responded. This meant we needed to divide two million non-responding addresses between the 33,000 field staff. We developed the Automatic Workload Allocation Tool, which divided these addresses into a group of addresses each field officer needed to visit (called a workload). The tool is a complex mathematical problem solver that calculated distances needed to be travelled by field staff from their homes and between dwellings. As planned, it proved to be far more efficient and timely than undertaking this process manually.

Central to the field staff role was a mobile app, MyWork, developed by IT vendor Tigerspike, that staff installed on their devices. MyWork provided staff with their workloads, including a list and map of addresses they needed to visit, and led them through the tasks they needed to undertake for each dwelling. The app worked both online or offline allowing field officers full functionality when working in remote locations across Australia with little or no connectivity.

The offline capability enabled the ABS to digitise field operations in remote Aboriginal and Torres Strait Islander communities. In the past, collection was done on paper with the data later transcribed into our systems. Digitisation enabled office staff to identify missing dwellings while field staff were still in the area.

Census field staff also had access to the online Knowledge Base, which contained all Census procedures and supporting information. The Knowledge Base was also a communication tool, with notices providing important updates for field staff, particularly because of the pandemic.

For the first time, in the 2021 Census, the ABS used an external payroll provider, EPI-USE, to pay field staff. The EPI-USE mobile app allowed staff to enter claims for time worked and travel. This service provided value for money and it was easy for staff to use. In 2021, we also changed the payment model from expected hours of work to a simpler pay by the hour model that proved very popular with the staff.

The centralised Field Operations Management team were based in Canberra, with smaller teams in Melbourne and Darwin. This team peaked at 369 staff in August 2021 and included sections for:

• operations management
• field management and training
• special collection areas
• workload management for the field officers
• frame (address list) management
• IT systems support
• the field staff support centre

Each team had specific roles, importantly allowing teams to focus on areas of expertise.

The field staff support centre team had a single help line that serviced more than 76,000 calls. Of these calls, 80% were solved on the first call and 20% were escalated to specialist teams. This included help on procedures, workloads, technical issues, and work health and safety support.

Separate phone lines were provided for recruitment queries to the recruitment agency, ADECCO, and pay queries to payroll specialists, EPI-USE. Both these agencies had dedicated teams of 50 to 60 people.

The delivery of field staff training was also centralised. Field managers were trained by a central training unit and then directly trained their field officers. This led to consistent interpretation of procedures and efficient training.

Due to the COVID-19 related restrictions on travel, we delivered field manager training through Zoom. This required us to rapidly adapt content to enable online delivery and upskill the trainers. The majority of field staff were trained online and most field managers when asked for feedback said they felt confident in their roles following the training.

The ABS sourced a large temporary workforce of around 34,000 ABS officers for the 2021 Census. Recruitment of field staff was conducted jointly by ADECCO and the ABS. In urban areas across the country ADECCO advertised, received online applications, and vetted candidates according to targets set by the ABS. ADECCO recruited approximately 20,000 temporary Census staff this way. The remaining staff were recruited directly by the ABS. In outer regional and remote areas of Australia the ABS had multiple strategies, sometimes using ADECCO candidate lists but also more direct approaches through local community engagement, field staff contacts (often friends and families), grey nomad social sites, and newsletters to former Census employees. In areas of high recruitment difficulty, the ABS redeployed existing Census staff from surrounding areas.

The shifts outlined above led to efficiencies and improved outcomes, notably increased response rates in many areas. The centralised model enabled us to take advantage of many well-designed processes and technologies. Decision making, resourcing and communication were streamlined and efficient.

Materials logistics is a massive part of the Census, requiring the assistance of numerous companies. Our vendors needed to be particularly agile and have strong communication processes this Census due to the pandemic. Materials production and transport were challenged in many ways due to shortages of supplies, staffing and at times, access issues. The success and apparently seamless nature of the operation showed the maturity of our vendors and their willingness to work closely with us to adapt to changing situations.

Census forms were all printed and packed in Australia. The logistics included:

• printing and posting more than 2.2 million paper forms to the public
• printing and posting 8.3 million Census Instruction Letters and 2.7 million reminder letters
• producing 170,000 cartons of materials in 6,000 consignments to field managers.

We provided materials such as forms, satchels, ID cards, personal safety alarms and masks to our 33,000 field staff across the country. These processes also involved manufacturing, packing and distribution.

The figure below shows the production and flow of Census forms and other materials around the country. Each arrow shows a high-volume movement of materials, each of which has unique complexities and logistical elements. Forms were also delivered to Norfolk and the Cocos and Keeling Islands (not shown on the map).

Printing forms for the 2021 Census commenced in August 2020 and was completed in May 2021. ABS awarded Print Media Group (PMG) the forms printing contract, which included producing all four form types (Household, Personal, Interviewer Household and Special Short Form) for the 2021 Census. PMG also printed the forms for the 2019 and 2020 Census Tests and have been the paper forms print provider for the last four Censuses. Bulk pallets of forms were distributed to our logistics provider, National Mailing and Marketing and to IVE Group (the 2021 mail house provider for forms distribution by mail).

Australia Post played a critical role in the 2021 Census, delivering Census letters and paper forms to dwellings across Australia and returning paper forms to our Data Capture Centre in Dandenong, Victoria. An inter-agency strategic partnership agreement between senior staff of Australia Post and the ABS was signed in March 2021. This ensured there was a good understanding of Census requirements by Australia Post and strong communication between the two agencies.

FujiFilm and IVE Group provided mail house services, which included envelope production, printing letters, printing of addresses onto pre-printed paper forms and lodging mail with Australia Post. FujiFilm used warehouses in five sites: Sydney, Melbourne, Brisbane, Adelaide and Perth. This ensured delivery around the country within a few days of our national advertising schedule.

Throughout the COVID-19 pandemic, we worked closely with Australia Post and both mail houses. For example, IVE Group and Australia Post adjusted to enable a large number of paper forms to be mailed to areas where ABS faced staff recruitment and lock-down challenges.

The effective collaboration between the ABS, IVE Group, FujiFilm and Australia Post resulted in the successful and on-time delivery of more 13.6 million mail items to dwellings during a seven-week period.

National Mailing and Marketing were awarded the fulfilment and distribution vendor contract to prepare our field staff material despatches. They established a dedicated centralised logistics warehouse in Canberra where a 40-person team picked and packed more than 170,000 cartons. Many of these needed to contain material specific to their destination, meaning the logistics needed to be carefully managed. The cartons were despatched in a staged program over about eight months.

With their sub-contractors TNT/FedEx, National Mailing and Marketing coordinated the distribution of field material to nearly 3,000 field managers who then delivered the materials the other 30,000 field staff in all corners of Australia. They also helped monitor the return of materials from non-private dwellings (hospitals, nursing homes, hotels, and so on), the remote area strategies staff (who supported remote indigenous communities) and the special strategies staff (who supported homeless, mining, defence and island communities, as well as providing multilingual support in Census hubs).

The 114 remote areas required considerable additional planning, and coordination among our logistics team, TNT/FedEx, National Mailing and Marketing and the Census field teams. Direct drive, charter flights and barge deliveries were all used to make these deliveries.

Collections from remote areas

Our longest remote collection was 9,550 kilometres. It departed from Perth, then went to Kununurra, to Broome, back to Perth then across the continent to the Data Capture Centre in Dandenong. Another involved five pickup locations throughout remote Queensland, taking five days and covering about 5,500 kilometres.

Other field support material was produced by a print vendor, Neilson Communications. They produced more than 90 different items varying from single sheets to entire booklets and folders to support the field teams.

There were other vendors that played important roles in supplying, manufacturing and dealing with field material:

• Intandem supplied 30,000 satchels, more than 50,000 ID cards and lanyards, 35,000 security stickers, 40,000 personal alarms and 120,000 face masks.
• EPIRBHire supplied remote support devices including satellite phones, vehicle trackers and remote messaging devices.
• First Aid Kit Australia supplied 300 field first aid kits.
• WINC supplied all stationery required by field teams through a whole of government contract.
• Shred X provided all secure disposal services at the completion of field work.
• National Storage supplied storage facilities in capital cities where office capacity was not available.

Completed Census forms provide raw data from the public. In order to turn the Census form responses into statistics to inform critical decision making across Australia, the data needs to be processed – cleared, edited and coded. This work takes several months, significant resources and technical expertise by ABS staff. We received just over 10 million forms, just under 80% being via the online system and the remainder mailed to us through Australia Post.

The first step in processing is Data Capture. The Census Data Capture Centre was a temporary facility established in Dandenong, Victoria to receive and process the paper forms. We ‘registered’ mail as soon as possible after it was received so that field staff ceased following up households that had already mailed back a form.

The paper forms were then prepared, guillotined and fed into industrial scanners to turn them into electronic images. These images were then converted into digital data using Intelligent Character Recognition (ICR) software, and any issues were addressed through human intervention (such as where the ICR software couldn’t interpret a handwritten response).

During this process, names and addresses were separated from the responses to Census questions and other personal or household information.

With the responses to all questions in our databases we need to turn the answers to 64 Census questions into the 500 variables that are released as statistics.

An important aspect of this is to classify or code all answers to a standard list so that counts can be published in a recognisable and consistent way. As an example, one person might report their country of birth as ‘Stockholm’, while another writes ‘Sweden’. Both of these responses would be coded to ‘Sweden’ and eventually output as part of the count for Sweden in the Country of Birth topic. This coding activity occurs for all topics against all returned forms.

We have examined previous Census responses and other ABS surveys to develop an index of the most common ways people describe their job, the industry they work in, the qualification they have received, the address they live/work at and other specific demographic type questions. The Statistical Standards and Infrastructure section of the ABS then assign each of these entries in the index to the corresponding classification number in the relevant classification.

With over 1 billion Census questions responded to, we rely heavily on automation to undertake this coding. With every Census we find more efficient ways of doing this, while maintaining the expected data quality.

We have also added some 'smarts' to the online form to assist us in automation and to help respondents answer our questions. An example is targeted supplementary questions for Occupation and Industry. If you entered that you were a 'Nurse' into the online Census form, this would trigger a list of further options to choose from, for example, Nurse Practitioner, Midwife, Registered Nurse etc. These are assigned to the correct classification automatically without further clerical intervention.

Our systems also correct spelling mistakes, find relevant suburbs within given geographic areas and accurately match most responses to our indexes.

As good as these automated processes are, we still need to manually code responses that are too vague or unclear for our systems to recognise. This could be due to poor handwriting on paper forms or where the situation requires human intervention to resolve. Our coding staff are trained to apply their judgement in these situations. Other examples include where we receive duplicate forms for a dwelling, we may delete one of the forms or need to merge them if there are different people. We may also need to take some action if the form has a different address from where we sent information about Census participation.

While most responses are resolved through automation, even a small percentage of responses on the Census leads to a large volume of manual work. For the 2021 Census, this manual activity in processing required up to 550 staff over nearly seven months.

The 2021 Census achieved a very high response rate of 96.1%. However, this means we did not receive forms for a small number of occupied dwellings on Census Day. For these dwellings we apply a statistical process called imputation that scientifically 'fills in the blanks' with people counts along with some of their basic demographics such as age and sex.

The ‘hot-deck’ imputation method is used when a Census form is not received from the household in an occupied dwelling. This method replaces the missing data with values from a similar responding household, called a ‘donor’. In the 2021 Census, administrative data was used in hot-decking imputation to improve the selection of donors for non-responding households.

Also, where respondents have started but not fully completed their form, we also impute the important variables of age, sex and marital status to all people in the dwelling with incomplete data. This process utilises other responses on their form and other carefully considered statistical methods reflecting these data items in the responding population, to best guess their situation.

We have a strong commitment to maintaining quality throughout the entire process in order to disseminate a rich and accurate data set for the country. While a 96% is a very strong response rate, we need to do our best to ensure that no small community or population grouping is insufficiently counted. This is particularly important for disadvantaged parts of the country.

It is also very important to not introduce any error during our processing such as coding an address to a wrong area.

One method of checking for error is to compare counts between this and the previous Census for all topics for all geographic areas and key population subgroups. Any variations detected are carefully checked to reassure us that they are valid, such as being the result of known population movements. We also use other data sources from government and non-government entities to assess the accuracy of all Census topics and special populations. This includes data for such things as homelessness, defence bases and mining areas.

We use data visualisation tools during this quality assurance to highlight areas, population groups and/or topics with results outside of expectations.

Looking after the health, safety and wellbeing of the 35,000 Census staff across Australia was a high priority for the ABS. The nature of office and field roles are quite different, requiring unique approaches for each group.

Preparing for and running a Census comes with considerable effort and can result in pressure on our staff to consistently meet numerous deadlines over a long period of time. Census staff were given all the employee support and safety mechanisms that ABS office staff are given. However, we identified the need to expand on this support to address staff wellbeing in the Census more explicitly.

Underpinning this was the development of a Census Wellbeing Framework and Wellbeing Strategy.

Meld Studios built the framework using research into best practice models, such as those used by Beyond Blue, the Blackdog Institute, the Australian Electoral Commission and Safe Work Australia. The model was adapted for the special aspects of the ABS and Census culture.

The Framework vision was: ‘The 2021 Census is committed to supporting staff, promoting staff wellbeing, safety and positive relationships to enable staff to reach their full potential’. It identified six areas of wellbeing to consider: mental health, work, relationships, environment, physical health and workplace culture. The Wellbeing Strategy outlined specific and measurable actions to be taken to address each of the Framework elements.

A key action of the Strategy was to set up a Wellbeing and Engagement Project Team. The team was established in March 2020, with staff from across the Census self-nominating to join the team in addition to their regular role.

Some examples of the activities coordinated by the team include:

• monthly social activities, many run online – such as 10,000 steps, program breakfasts, a bird spotting competition and milestone celebrations
• a survey of Census staff to identify risk factors and generate ideas for actions that the leadership and wellbeing teams could take to further support staff
• guidelines for conducting meetings to improve their effectiveness and minimise their impact on staff time with particular consideration to people working from home or in a different office.

‘Put our staff first’ was ingrained throughout the Census program, including in the issues and crisis management activities, where the safety of our staff was paramount.

When COVID-19 restrictions required staff to work from home, the ABS was well placed, having enabled staff to do this for several years. We were ready both from a technical and cultural perspective to easily switch to this model. A subsequent staff survey indicated that the majority of staff (93%) were working at home at least some of the time, and 73% of staff were working from home all of the time. We recognised that with an extended work from home period, it was even more important to focus on wellbeing and actively build social connections across the Census Program. Due to security and privacy protections, staff processing the Census responses could not work from home, so active management of a small number of COVID cases in the office was necessary.

The Census team organised induction programs for all staff, including Census 101 training and a whole of Census Program workshop in early 2021, called the Census Survival Guide.

Census Survival Guide sessions were recorded and teams were encouraged to participate together. This included watching presentations and then working through team and personal reflection exercises.

Some of the topics for the Census Survival Guide included:

• understanding peaks and troughs of the Census so staff could mentally prepare for when extra effort was going to be required
• listening to reflections from veteran Census staff on what working on the Census is like
• recording their team definition of success
• developing a personal Census wellbeing plan that aligned with the Census Wellbeing Framework.

Census field staff were provided with a phone application, called SafeZone, that allowed us to monitor their safety when working outside their homes. They were required to log in when they started and stopped their work and if a staff member failed to check out as expected, alerts would be automatically triggered. Staff could also trigger incidents in the case of an emergency or the ABS could send safety alerts to staff in the field. This was the first Census where field staff had automated and centrally monitored check-in and check-out services.

The ABS put in place a dedicated Census Workplace Health and Safety Team and Safety Advisor Line that provided timely and consistent advice and support to all field staff on work health and safety procedures, processes and training. The team:

• managed 22,000 active SafeZone users, with 284,843 check-ins
• responded to 14,000 alerts, with 700 escalated to the field staff support centre and 12 escalated for emergency contact or police action.

The ABS supports a reporting culture of incident and hazard management. Field staff were able to lodge these in real time and when appropriate, link them to addresses, so that any action could be taken immediately.

The ABS classed an incident as an unplanned event that caused, or had the potential to cause, injury, illness, exposure, damage or loss. This included near-misses. For example: 

• getting an electric shock
• falling over and spraining an ankle
• being bitten or chased by a dog
• being threatened by another person
• being involved in a car accident
• experiencing dangerous levels of fatigue while driving.

We classed a hazard as a situation or thing with the potential to harm a person. For example:

• a road in poor condition
• an unlit stairway
• an upset or agitated person
• fatigue
• extreme weather
• an unrestrained dog.

For the duration of field activities there were more than 1,331 incidents reported and 2,188 hazards where action was required. In addition, more than 44,000 beware hazards were recorded for the attention of field staff.

Measuring success of a cultural program of this type is not simple or obvious. Anecdotally it has been a big success, with staff surveys showing positive reflections and feedback. The majority of staff reported feeling:

• good or very good about their day
• empowered and supported in their job
• positive about their supervisor and management
• safe.

These results have been consistently strong across diverse groups, locations, job classifications and Census teams.

The ABS was supported by the FBG group throughout Census preparations and operations to help build a resilient workforce. They facilitated Census workshops, provided coaching and support to the Census Leadership team, and helped establish the Census Wellbeing and Engagement Project team. Their support assisted us in establishing a strong culture of open communication and inclusiveness.

"The Census has been one of the most interesting (and significant) projects in my career. Many organisations pay lip service to wellbeing – particularly when the challenges are a bit curly. Census has taken a long-term and inclusive approach to ensuring every effort is made to support its people. A staff driven wellbeing working group did amazing work – and kept the wellbeing discussion “real”. I was delighted to find wellbeing embedded in the strategic planning phase – and personally participated in workshops with multiple groups. The willingness for senior leaders to initiate and participate in these activities was constant and always positive.

One highlight was the week I spent in Canberra over Census week. My brief – “Simon – just make sure we are all ok – you’ll work it out.”

I am grateful for the openness to a bit of creativity – some of our workshop content or titles were not exactly mainstream, but so “Census”. Overall – a privilege to work with such fantastic people and play a small part in a successful Census. Bring on 2026!’

Simon Brown-Greaves, CEO and organisational psychologist, FBG Group

When Australia has taken censuses during and immediately following significant world events, such as the Spanish Flu, the Great Depression and the World Wars, the data has been instrumental in understanding the effects these events have had on the economic and social fabric of Australia. This Census will be no different, and the data we are releasing has immediate value to users and enduring value to researchers, because it will reveal how the pandemic has changed life in Australia.

When the COVID-19 pandemic emerged in Australia in early 2020 and its impacts became known, the ABS began planning for the prospect of it affecting Census collection in 2021. Being able to observe the potential implications a year out from most of our collection and processing activities, provided some opportunities.

We planned extensively on how to engage with the community, and how to protect staff in the event of a COVID-19 outbreak. We developed a three-step COVID-19 safe plan that described how we would conduct Census fieldwork based on the relevant step we were using at the time.

We conducted an operational readiness exercise in October 2020 involving 100,000 dwellings in three states. As part of the exercise, we tested the effectiveness of our COVID-19 safe plan:

• In Victoria, we used a mail-only approach with no field staff, relying on letters and advertising because Victoria was locked down at that time.
• In Sydney, we tested a contactless approach where field staff dropped material at letterboxes instead of knocking on doors.
• In South Australia, we tested how we would withdraw our field staff in response to a sudden change in restrictions in that state and how we would organise the safe return of field staff once restrictions were eased. 

We took advantage of similar international census experiences affected by COVID-19, particularly:

• the United States of America - Census day was 1 April 2020
• the United Kingdom - Census day was 21 March 2021
• Canada - Census day was 11 May 2021.

The key goals of our COVID-19 approach were to maximise response while keeping the community and our staff safe. This involved continuously monitoring the changing COVID-19 situation to ensure that our safety measures were up-to-date and met the guidelines and advice from the Commonwealth, state and territory governments. It also meant actively adapting our approach to suit local circumstances, including lockdowns.

We developed specific reports to monitor potential impacts on collection such as observing areas with low field staff recruitment, lower than expected response or low number of dwellings. From these reports we could take predefined actions such as:

• adjusting recruitment targets
• moving field staff to different areas
• adjusting the timing of Census reminder letters or the messaging in operational communications.

A critical aspect of our planning was how we used technologies, particularly those for our field staff. We established ways to engage our field staff online including through virtual onboarding and training. Important functions such as payroll and workload distribution were also managed online. This helped trainers or field managers who were not able to have face-to-face interactions with staff. This also involved developing more online training.

The ABS COVID-19 safe plan included strategies to count people in quarantine facilities. This ensured both incoming travellers from a declared COVID-19 hotspot and international arrivals staying in a government quarantine facility could participate in the Census. To ensure the safety of field staff and the community, there was no direct contact with people in quarantine facilities.

COVID-19 quarantine facilities were removed from field staff workloads and were managed remotely by ABS office staff. A point of contact was established for each facility to coordinate instructions and to identify the estimated number of residents staying at the facility on Census day. In line with health advice, we did not provide paper forms and all quarantined travellers completed the Census online.

Given the ongoing labour shortage in Australia during the pandemic, the ABS faced challenges in recruiting people with the skills and experience to fill some Census field roles. This situation occurred across the country, from the inner cities to remote regional areas in Northern Queensland, Central and Northern Western Australia that traditionally rely on transient workers.

Half of the eastern coast of Australia was in lockdown on and around Census day, forcing us to use casual contact and contactless procedures for much of our engagement. The lockdowns also directly resulted in an increased number of unoccupied dwellings in the cities and surrounding holiday areas.

The logistics of our operations were at times quite challenging. Moving staff across state boundaries and even some Local Government Areas was restricted. When we could move staff, we also had trouble finding temporary accommodation and rental cars for them. Our plans for establishing pop-up information hubs in shopping centres and libraries to help people from non-English speaking backgrounds or older Australians was also restricted.

Some Aboriginal and Torres Strait Islander communities were closed for long periods during the pandemic, forcing us to use different engagement strategies. This included accepting help from other departments, such as Health and Services Australia, to access these areas.

In the large urban centres of New South Wales and Victoria, many people experiencing homelessness were moved to supported temporary accommodation. We engaged extensively with services supporting these people, including state and local governments, to make sure they were counted.

Despite the challenges, we quickly adapted in many ways, as described in detail in previous stories. Our preparations and agility served us well and most areas met or exceeded their target response rates.

People under COVID-19 lockdown restrictions were still required to complete the Census. The ABS advised them on how to respond to questions related to study and work impacted by the change of circumstances. Most people were able to complete the Census online. They either received instructions in the mail or had a paper form dropped off outside their dwelling or they used the no Census number pathway to log into the online form. The work completed in user-centred design also allowed respondents to readily access the services required.

Our engagement with many vulnerable communities was carried out with care and sensitivity without compromising response rates. Where we applied contactless procedures, we have seen high response rates.

Communication with our large workforce on the changing environment and any subsequent changes to our procedures worked well. The online Census Knowledge Base proved to be an extremely effective communication channel, with daily and even hourly updates of field notices keeping staff informed.

Our new centralised management model helped us in making timely decisions and quickly adapting procedures.

Australia Post’s agility in responding to our changing requirements was greatly appreciated. With little notice, they dispatched an extra one million extra paper forms and letters just before Census day, 10 August 2021.

While not initially planned in this way, 99% of all our engagement with field staff was done online.

Importantly, we kept our staff and the public safe following the three-step COVID-19 plan with effective workplace health and safety practices.

2021 Census data was published on 28 June 2022 as part of a three-phased approach to releasing the data. The first release included key demographic data about the population – age, sex, location of where people were in the country on Census day and where people usually live.

The first release of 2021 Census data also included information about cultural diversity, religious affiliation, Australian families and households.

For the first time, data was released from two new questions about long-term health conditions and service in the Australian Defence Force.

The phased release will continue with two further releases scheduled:

• 12 October 2022 – topics including employment, educational qualifications, and internal migration
• early to mid-2023 – more complex data such as distance to work, socio-economic indexes for areas and counts for people experiencing homelessness.

The first release of the Census data on 28 June 2022 showed we are a fast changing, growing and culturally diverse nation. The data provided a picture of the economic, social and cultural make-up of Australia. Australia’s population had grown to 25.4 million people, excluding overseas visitors. This was an increase of over two million people or 8.6% since the 2016 Census. Australia’s Census count had more than doubled in the last 50 years, with the 1971 Census counting over 12.5 million people.

The data showed almost 80% of Australian residents lived in eastern Australia in New South Wales, Victoria, Queensland and the Australian Capital Territory. Two thirds of people counted, 66.9%, were in Greater Capital Cities and 33.1% were in the rest of Australia. As with previous censuses, greater Sydney continued to be the largest city in Australia with 5.2 million people, followed by Melbourne with 4.9 million people and Brisbane with 2.5 million people.

On Census day, Tuesday 10 August 2021, many parts of Australia were in lockdown or subject to border restrictions. The data has been reviewed with consideration of COVID-19 impacts. International border restrictions were in place and the 2021 Census counted 61,860 overseas visitors, significantly lower than 315,531 in 2016. We have seen that lockdown and travel restrictions influenced where people completed the Census:

• the Census household response rate increased from 95.1% in 2016 to 96.1% in 2021
• the percentage of households completing online was almost 80% in 2021, an increase from 63% in 2016
• two million more people were at home on Census night than in 2016.

The median age of all Australians remained at 38 years in 2021. Males made up 49.3% of the population with a median age of 37 years and females made up 50.7% with a median age of 39 years.

Australia has undergone a significant generational shift. Baby Boomers and Millennials each have over 5.4 million people, with only 5,662 more Baby Boomers than Millennials counted on 10 August 2021.

The Census found that 812,728 people identified as Aboriginal and/or Torres Strait Islander. This has been an increase of over 25% (25.2%) since 2016, with Aboriginal and Torres Strait Islander people representing 3.2% of the Australian population. Of the Aboriginal and Torres Strait Islander people counted, 91.4% identified as Aboriginal, 4.2% as Torres Strait Islander, and 4.4% identified as both.

Australia has a rich mix of cultural backgrounds and heritage, with the number of people living in Australia who were born overseas continuing to increase.

The proportion of Australian residents born overseas (first generation) or have a parent born overseas (second generation) has moved above 50% (51.5%).

The top five most commonly reported ancestries in the 2021 Census followed previous trends and included English at 33.0%, Australian at 29.9%, Irish at 9.5%, Scottish at 8.6% and Chinese at 5.5%.

Australia has continued to be a culturally and linguistically diverse country with the growth of communities from Nepal, India, Pakistan, Iraq and the Philippines.

The diversity of religious affiliation has increased across the Australian population.

The percentage of Australians reporting no religious affiliation has continued to grow and was recorded at 38.9% of the population, compared to 30.1% in the 2016 Census.

Christianity has remained the most common religion with 43.9% of the population identifying as Christian, a decrease from 52.1% in the 2016 Census. 

The 2021 Census counted more than 10.8 million private dwellings across Australia. These dwellings were comprised of separate houses (70%), apartments (16%) and town houses (13%). The proportion of apartments has continued to increase, with apartments accounting for nearly one third (30.9%) of the increase in private dwellings since 2016.

The average number of people per household decreased from 2.6 in 2016 to 2.5 in 2021. The Census reflected that Australian families come in many shapes and forms. The 2021 Census counted 5.5 million (5,552,973) couple families, of which 53% had children living with them. Over one million one parent families were counted, of which four out of every five of those parents were female.

More than half a million Australians (581,139) have served, or are currently serving, in the Australian Defence Force. One in twenty (5.3%) Australian households (dwellings) had at least one person who had served or is serving in the Australian Defence Force.

For the first time, the Census collected information on ten common long-term health conditions in Australia. Over eight million people reported they had been diagnosed with a long-term health condition in the 2021 Census.

• 4.8 million people reported having one of the selected long-term health conditions
• 1.5 million had two of the selected long-term health conditions
• 750,000 people reported having been diagnosed with three or more of the selected long-term health conditions.

As part of the phased approach to releasing 2021 Census data, the ABS is publishing a suite of communication products and data tools to help people access the data. These will continue to be published into 2023 to support the second and third releases of the data. The communication products include national media releases, topic‑based articles delving deeper into the data insights and media backgrounders providing contextual information for journalists.

Census data tools have been published on the ABS website to help people access the data in easy to use formats. Census tools include:

• Search for Census data providing geographic based data from QuickStats (75,000 areas) and Community Profiles (15,000 regions).
• Census topics providing access to stories and visualisations that relate to Census data on topics of interest e.g., population, health, families.
• DataPacks containing comprehensive data about people, families and dwellings for all Census geographies ranging from Australia down to Statistical Area Level 1.
• GeoPackages containing bulk data for multiple areas in geospatially enabled files.

2021 Census data is also available through additional tools that incorporate other ABS data. These include ABS Data APIs and DataLab.

The ABS plans to release further products including 2021 Census data in TableBuilder and a range of microdata products.

The 2021 Census data release launch took place on 28 June at 10:00am AEST in the National Portrait Gallery, Canberra. 2021 Census data was published at the same time on the ABS website. The event was live streamed and available to view from the ABS website. More than 9,000 people tuned into the live stream link from the ABS website, including the ABC News 24 channel which also streamed the event live on national television.

The data release launch officially began with a Welcome to Country delivered by Ngunnawal man, Wally Bell. The Hon Dr Andrew Leigh MP, Assistant Minister for Competition, Charities and Treasury, delivered opening remarks at the launch. This was followed by an address from Professor Sandra Harding AO, Emeritus Professor, James Cook University, Chair of the 2021 Census Statistical Independent Assurance Panel who acknowledged the 2021 Census data was fit-for-purpose, of comparable quality to previous censuses and could be used with confidence.

Dr David Gruen AO, Australian Statistician, and Ms Teresa Dickinson PSM, Deputy Statistician and Senior Responsible Officer for the 2021 Census, provided an overview of 2021 Census data insights across the topics released.

The presentations were followed by a 30-minute question and answer session with the audience, both live and virtually. The launch was attended by media from the Australian Parliament House Press Gallery, including SBS News, Sydney Morning Herald, Australian Financial Review, ABC News, The Australian, Canberra Times, Australian Associated Press and the Ten Network.

A live audience of 50 people included Commonwealth, state and territory government stakeholders and ABS staff. Following the formal proceedings, Census spokespeople participated in media interviews to facilitate television coverage of the data release.

A recording of the launch is available on the Events page of the ABS website.

The ABS delivered a strategic approach to media engagement for Census data release by publishing a suite of 10 media releases with Australia level 2021 Census data insights on the ABS website on Monday 27 June at midnight prior to the official data release launch on Tuesday 28 June. This resulted in significant national media coverage trending positive in sentiment across various media platforms.

Almost 14,000 news clips about 2021 Census data were published across radio, television, print and online with a 78.2 million potential total audience reach between Tuesday 28 June to Monday 4 July. Census spokespeople conducted 18 media interviews during this period, highlighting the stories emerging from the data at the national and state level.

On social media, ‘Census’ was the top trending topic on Twitter on 28 June at 6:00pm AEST. Audience reach on the 2021 Census Facebook page peaked at over 3.6 million during this period. We would like to thank the journalists and media outlets that help us tell the stories from the 2021 Census.

The 2021 Census data seminar series commenced in July 2022 to share key data insights with stakeholders. The national data seminar took place on 7 July at Hotel Realm in Canberra and was live streamed for virtual attendees. The seminar included information on what the data tells us about how Australia has changed across the first release topics, uses of Census data and what Census data tools are available.

Data experts from the ABS presented an overview and this was followed by an audience question and answer session. The 2021 Census national data seminar is now available to watch on the ABS YouTube channel. This event was followed by an Aboriginal and Torres Strait Islander seminar on Monday 11 July, which is also available to watch online.

To facilitate understanding and engagement with the data, the ABS is hosting a Census data seminar for each state and territory to share insights at the regional level. Recordings of each seminar are available on the ABS website. The seminar series will continue before and after the second release of 2021 Census data. Later this year, the ABS is planning to host a series of topic-based seminars, including on multiculturalism in Australia, long-term health conditions and service in the Australian Defence Force.

We have already turned our mind to the 2026 Census and looking at ways we can build on the success of 2021. We know we must continually evolve to meet public expectation about what a Census should deliver.

As highlighted throughout this report, the Census is an enormous exercise in terms of scale and such an endeavour has its challenges. We must continue to evolve and address challenges if we are to achieve the highest quality data.

In regional Australia and remote areas, achieving high response rates remains an ongoing challenge despite consistently improving our engagement effort. The undercount for Aboriginal and Torres Strait Islander people in these areas will continue to be a high priority and an area of focus in preparation for the 2026 Census.

The Cyber environment continues to evolve, and we need to work closely with public and private cyber security experts to ensure our defences are robust and secure.

Recruiting a large workforce across Australia poses many challenges but this was particularly difficult in 2021 during the COVID-19 pandemic. In pockets across the country, we had gaps in our field workforce. This brought about difficulties in delivering Census materials to respondents and providing enough support to help people participate. We will need to continually innovate our recruitment approaches to address potential shortcomings in the 2026 Census cycle.

The technology used by the field workforce is important but a small minority of the workforce have challenges in using our technical solutions. We will continually work on ensuring we take a user-centred design approach to these systems so they are intuitive, easy to install, simple to find support and generally make it easier for our staff to complete their Census field work.

The Australian public are very supportive of the Census and appreciative of the engagement and support we provided for the 2021 Census. Understanding our customers is critical to informing what support we provide. We must continue to improve our customer support services to cater for those who need help participating, e.g. making the Census accessible, providing a braille form and improving digital and non-digital support options.

The stories outlined in this report have explained in some detail what we have done to overcome these challenges in 2021 but there is always more to do.

Given the success of the 2021 Census, we are not planning to fundamentally change the model for running the 2026 Census.

We will invest our resources to continue to:

• achieve high quality Census data at a national, state and small area level to inform decisions so that the right services can be provided across Australia
• maintain and grow the value of the Census for all Australians
• deliver a safe Census with a good public experience. 

We will be considering factors such as:

• the rising costs of running a Census
• public expectations of what constitutes a good government experience
• the opportunities arising from new technologies to give respondents an even better experience
• opportunities for expanding the use of administrative data to increase data quality in the processing phase of the Census
• key risks such as supply chain disruptions and availability of labour
• the need for continual vigilance and awareness of changes in the cyber security environment
• approaches and experiences from other international Census colleagues who have similar Census models, including USA, Canada, UK, NZ and Ireland.

While paper materials will still be available, the 2026 Census will again take a digital-first approach. Technology changes rapidly and we will continue to evolve to ensure a safe, secure and easy online Census experience.

We will continue working closely with Census stakeholders, including data users and Aboriginal and Torres Strait Islander and culturally and linguistically diverse communities and organisations representing people with disability. Connecting and seeking guidance from these communities helps us understand their needs so we can provide the right support; where, when and how it’s needed and ensure all Australians get value from the data produced from the Census.

Choosing the topics that are on the Census is one of the first activities we kick off in the Census. Many Census topics are critical to support key decisions and provide an understanding of social changes over time.

A number of Census topics, including age and location, support key government decisions, feed into official population estimates for Australia and therefore will not change. However, other topics could be enhanced or expanded in the 2026 Census.

Consultation with the Australian public plays a key part in determining the topics collected on the Census. We will be commencing the public topic consultation process towards the end of 2022. Following the consultation we will work with stakeholders, including major data users and undertake extensive testing to refine the topics. Recommendations from the ABS will then be submitted to the Australian Government who makes the final decision on Census topics.

We are looking forward to starting this very important process and working with the Australian community to understand what our Census topics will look like in 2026. This will be done in line with consideration of the broader social statistics program run by the ABS.

The terms below are in addition or complementary to the 2021 Census Dictionary Glossary.

Accessibility testing

Evaluating a product or service by testing how it can be easily used by people with disabilities. We aim to comply with WCAG standards with all our products.

Behavioural economics

A study of why people behave the way they do based on economics, psychology and emotions.

Culturally and linguistically diverse (CALD) populations

Australia’s population includes many people who were born overseas or have a parent born overseas and speak a variety of languages. Together, these groups of people are known as culturally and linguistically diverse (CALD) populations. 

Census day

Everyone in Australia is asked to fill in a Census form in relation to where they are residing on this day. For the 2021 Census, it was 10 August.  

Census instruction letter (CIL)

This letter is mailed to households with instructions and a code to complete the Census online and explains how to request a paper form if preferred.

Census reminder letter

This letter is mailed to households who have not yet responded to the Census. The letter contains instructions and a code to complete the Census online.

Census supporters

An organisation or person who promotes the Census within their community. Census supporters help by hosting information sessions or sharing Census information with their community.

Coding

The classifying of answers in each topic (group of Census questions) to a known list of answers so that counts can be published against them.

Data lake

A centralised repository designed to store, process and secure large amounts of structure, semi-structures and unstructured data.

Distributed Denial of Service (DDoS)

A cyber attack designed to load a network with so much malicious traffic that it cannot operate or communicate as it normally would. This causes the site's normal traffic, also known as legitimate packets, to come to a halt.

Dwelling

A structure which is intended to have people live in it, and which is habitable on Census day. Some examples of dwellings are houses, motels, flats, caravans, prisons, tents, humpies and houseboats. 

Field work

Work Census staff do outside of their homes such as knocking on doors, letter box drop or visiting establishments.

Household form

The Census household form (online or paper) is the primary means for collecting Census data and is used in all private dwellings. 

Imputation (persons)

Creating people counts and some basic demographics such as age and sex where the dwelling is determined to be occupied but didn’t respond to the Census.

Inclusive strategies

Strategies designed to ensure everyone is included in the Census regardless of their circumstances and where they are residing, their literacy or ability to complete a form. The strategies generally focused on improved engagement activities for awareness and supporting completion of the Census form.

Intelligent / Optical character recognition (in Census glossary)

Specialised computer software is used to interpret the handwriting on images taken of each page of the Census form. Once recognised, answers to Census questions are coded to the appropriate category of the relevant classification, for example Religion, Occupation, etc.

Interviewer household form (in Census glossary)

The Interviewer Household Form is used in Aboriginal and Torres Strait Islander communities (and areas where language differences or other factors make use of the standard self-enumeration forms impractical). It is an interview based Census form. 

Information Security Registered Assessment Program (IRAP)

The security service standard where independent assessors follow a multi-stage process to review all aspects of security for a product. Passing an IRAP assessment is a very strong sign of the security.

National Relay Service

Australian Government phone service for people who are deaf, have a hearing impairment and/or speech impairment and people wanting to communicate with them. 

Non-responding dwellings

Dwellings where a response has not yet been received, typically after Census day.

Occupancy determination

Determining if a dwelling will be/was occupied on Census day.

Penetration testing / ethical hacking

A security exercise where a cyber-security expert attempts to find vulnerabilities in a computer system. The purpose of this simulated attack is to identify any weak spots in a system’s defences which attackers could exploit.

Performance testing

Also called load testing, typically involves a simulated volume of traffic on a system to establish any break points and the systems total capacity.  

Personal form (in Census glossary)

The Census personal form (online or paper) records details for one person only. It contains the same questions as the Census household form but excludes the questions related to the dwelling. It is used for people staying in a non-private dwelling such as a hotel, motel, hostel, or nursing home or if a person in a private dwelling requests one for privacy or other purposes.

Social listening

Proactive listening to non-Census and non-ABS owned social media channels including public access forums and websites. The purpose is to:

1. search for and identify possible areas of dis/mis/mal-information, and where appropriate intervene
2. ensure communication is responsive to changing needs, interests and understanding of audiences and dispel misinformation.

Social media misinformation

Information that is false, but not created with the intention of causing harm. For example: A tweet saying that this year the Census can only be completed online.

Social media disinformation

Information that is false and deliberately created to harm a person, social group, organisation or country. For example: A post on a Facebook group saying that the Census sells your personal information and you should put in false data to protect yourself.

Social media malinformation

Information that is based on reality, used to inflict harm on a person, organisation, or country. For example: A post on Facebook saying that someone didn’t do the Census in 2016, didn’t get fined and encouraging others not to do the Census.

Special short form

Single page Census form mainly used for interviewing people experiencing homelessness.

Statistical impact

Something that has caused the data quality to be less reliable or accurate. Typically, it means the data hasn’t met an acceptable standard for release.

Statistical data quality

A measure of the data based on factors such as accuracy, completeness, consistency, reliability and whether it is up to date.

Usability testing

Evaluating a product or service by testing it with users. The goal is to identify any usability problems such as observing where the user becomes confused or follows an incorrect path.

Web Content Accessibility Guidelines (WCAG)

The Web Content Accessibility Guidelines (WCAG) is an internationally recognised standard created by the World Wide Web Consortium (W3C). The WCAG standard defines how to make web content more accessible to people with disabilities.