|Page tools: Print Page|
There are four principle sources of error in Census data: respondent error, processing error, partial response and undercount. Quality management of the Census program aims to reduce error as much as possible, and to provide a measure of the remaining error to data users, to allow them to use the data in an informed way.
The Australian Census is self-enumerated. This means that householders are required to complete the Census form themselves, rather than having the help of a Census Collector. The Census form may be completed by one household member on behalf of others. Error can be introduced if the respondent does not understand the question, or does not know the correct information about other household members. Self-enumeration carries the risk that wrong answers could be given, either intentionally or unintentionally. The ABS has a number of ways to minimise respondent error.
Self-enumeration imposes limits on the types of topics and questions that can be included in the Census. Topics which require complex questions or question sequencing are not suitable for a Census as the responses obtained may not be reliable. There is also the need to limit the total number of questions asked in order to minimise the amount of time it takes for a respondent to complete the Census form.
Topics are selected for inclusion in the Census following extensive community consultation. Topics are selected based on the following criteria:
The Census form is designed so that questions are easily understood and simple for respondents to answer. Most questions are answered by a box being marked, although some questions require written responses.
Questions are tested on focus groups to ensure they are clear, well worded and can be answered on behalf of others. The focus groups are made up of people from diverse backgrounds who are representative of the Australian population. Following the successful completion of the focus group phase, field tests are conducted in various cities and rural locations. These assist in assessing how the questions and the Census form work in a real environment.
Raising public awareness
To achieve high quality Census data it is essential that people understand the importance of being counted and of giving the right answers in the Census. Raising public awareness through advertising and community briefings contributes to high levels of participation in the Census. It helps people understand the benefits to the community of complete and accurate Census counts and minimises intentional respondent error.
The public relations campaign also aims to make people aware of the help that is available for people who have problems filling out their Census form. Help is available from the 'Census Guide' brochure, the Census web site and from the Census Inquiry Service telephone help line. This assistance helps to reduce respondent error.
Much of Census data is recorded using automatic processes, such as scanning, Intelligent Character Recognition and other automatic processes. Quality assurance procedures are used during Census processing to ensure processing errors are kept at an acceptable level. Sample checking is undertaken during coding operations, and corrections are made where necessary.
Once forms are received, they are checked for damage and errors, such as tears, multi-mark responses and illegible handwriting. Where required, these problems are fixed manually to assist the automatic coding processes.
Most responses are coded automatically using official classifications with legal value checks built into the system. In addition, a random sample of codes is checked manually against the original response on the form. Errors are more likely to arise during automatic coding of 'write in' answers. Clerical staff resolve problems that arise if text responses cannot be automatically matched to the index of possible responses. Their work is subject to a quality management process to ensure that errors are not being made.
The completed data are put through a series of automated checks to ensure internal consistency. The data are also scrutinised for changes over time, by comparison with previous Census data and other data sources, and across categories, where expected trends can be identified, and unexpected trends investigated.
In preparing Census data for output, various derivations and recodes are applied to the data to produce the variables listed in this dictionary. Data are processed further to create the range of Census data products. A series of checks occur at each stage of the output process to ensure data consistency and accuracy.
When completing their Census form, some people do not answer all the questions which apply to them. While questions of a sensitive nature are generally excluded from the Census, all topics have a level of non-response. However, this level can be measured and is generally low. In those instances where a householder fails to answer a question, a 'not stated' code is allocated during processing, with the exception of non-response to age, sex, marital status and place of usual residence. These variables are needed for population estimates, so they are imputed using other information on the Census form, as well as information from the previous Census.
The goal of the Census is to obtain a complete measure of the number and characteristics of people in Australia on Census Night and their dwellings, but it is inevitable that a small number of people will be missed and some will be counted more than once. In Australia more people are missed from the Census than are counted more than once. The net effect when both factors are taken into account is an undercount.
During the delivery and collection of Census forms to households, quality assurance field procedures are put into practice to ensure the maximum number of households are included in the Census.
Area Supervisors are responsible for eight to ten Census Collectors. The Area Supervisors' main role is to ensure accuracy and completeness of coverage within their areas. They must take into account any changes in the number and type of dwellings in their area since the completion of collector workload design. They also review each Collector's work during and after the collection, using a defined set of checks. This ensures that all relevant details are recorded in the Collector's record book, and that a form exists where expected.
Every effort is made to ensure that all households receive a Census form and that these are collected and completed. For example, Census Collectors are required to return to a household up to a total of five times after Census Night to attempt to collect the form. This also applies where a householder states they returned their form via electronic lodgement (eCensus) or mail but the collector has not received notification of the receipt of the form.
All forms are registered to the collector workload they come from, so that Data Processing Centre staff can account for all forms received as well as those still to be returned by mail or by electronic lodgement (eCensus). Ensuring receipt of the expected number of forms for each collector workload from the collection phase is a critical measure of the completeness of the Census.
Some groups of people in the population are undercounted in the Census. These include Aboriginal and Torres Strait Islander people, ethnic groups who have trouble reading or speaking English, the homeless and people with certain disabling conditions which prevent them from filling in a Census form. In addition, some areas are more difficult to enumerate, including secure apartment buildings and remote discrete communities. Special strategies have been developed to ensure a more complete count of these groups and areas.
A measure of the undercount in the Census is obtained from a sample survey of households undertaken shortly after the Census, called the Post Enumeration Survey. It collects information about where people were on Census Night and their characteristics, which are compared to the actual Census forms. The Post Enumeration Survey for the 2006 Census indicated an undercount of 2.7% in the Census. The Post Enumeration Survey results are discussed in more detail in Information Paper: Measuring Net Undercount in the 2006 Population Census, 2007 (cat. no 2940.0.55.001).
Information from the 2006 Post Enumeration Survey was used in planning the collection procedures for the 2011 Census, with the aim of improving the distribution and collection of Census forms in the identified undercounted groups.
Quality Assurance of Census Products
Decisions about how and what is released from each Census are influenced by feedback from users of Census data. Extensive user consultation was carried out for the 2011 Census. Feedback from users has indicated they would like more information about data accuracy, consistency, comparability and accessibility, non-response rates, and undercount and overcount measures. Based on this feedback, the ABS has reviewed its methods of providing information on Census data quality and is committed to providing a clear explanation of Census operations. This includes providing general data quality information, such as the Census Dictionary, and providing at least some basic data quality information such as non-response rates with every table of data provided by the ABS.
Introduced random adjustment
Individual Census records are confidential. Before Census data are released, small random adjustments are made to allow the maximum amount of detailed Census data possible to be released without breaching confidentiality. Consequently, care should be taken when interpreting cells with small numbers, since randomisation, as well as possible respondent and processing errors, have a greater impact on small cells than on larger cells (see also 'Introduced random error' in the Glossary section).
Where to Find Data Quality Information
For the 2011 Census, data quality information will be available with the Census data as they are released, through links on Census Web pages. These pages will allow the data quality information to be printed or downloaded along with the data.
Data quality statements
When Census data are released, each variable will be linked to the corresponding entries in the 2011 Census Dictionary. Data quality statements will include the non-response rate for each Census variable and a brief outline of any known data quality issues.
Assistance in the use and interpretation of Census data will be provided through a series of information sheets. They will provide a summary of conceptual and data issues, and changes that have occurred since the last Census. These information sheets will be available after data release.
Further analytical and evaluation papers will also be made available to address other data quality issues that require investigation. They will be released at www.abs.gov.au/census.