See Birthplace, Language (LANP), Proficiency in English (ENGP), Year of Arrival (YARP).
See Dwelling, Dwelling Structure (STRD).
See Derivations and Imputations.
See Residual Categories and Supplementary Codes.
This variable records the income level of people aged 15 years and over. People are asked to state their usual gross weekly income, which is the income before tax, superannuation, health insurance, or other deductions are made.
Gross income includes family allowance, family allowance supplement, pensions, unemployment benefits, student allowances, maintenance (child support), superannuation, wages, salary, overtime, dividends, rents received, interest received, business or farm income (less operation expenses) and worker's compensation received.
People are not asked to state their exact income, only to indicate the range into which their income falls.
Income from some sources may be negative. As income from most sources is reported before deduction of expenses incurred in the earning of the income, these incomes are always a positive figure. However, business income from own unincorporated enterprise and income from rental property is collected net of expenses incurred in the raising of this income. Therefore, income derived from business or from rental properties may be a negative value. The total income may therefore be a negative value. To record this information the 'negative income' category has been included for the 1996 Census.
For the 1996 Census, the classification ranges from 'negative income' to 'no income' to '$1,500 or more ($78,000 or more per year)'. This classification differs from 1991 where the classification did not allow for the separate identification of negative income or no income. The lowest income range was 'Less than $58 (Less than $3,001 per year)' and the highest income range was over '$70,000 per year'.
Information on income distribution is important in planning public and private sector services such as social welfare and, particularly at the regional level, retail distribution and other commercial services.
A question on income was first asked in the 1933 Census in an attempt to assess the effects of the depression. It has subsequently been included in the 1976, 1981, 1986, 1991, and 1996 Censuses.
Family Income (FINF) and Household Income (HIND) are calculated from the individual incomes. These derived incomes have a higher upper limit for the 1996 Census, in response to user demand. The income of other groups of people in a household can also be calculated on request.
See also Family Income (FINF), Household Income (HIND), Median Income.
Indigenous Areas are aggregates of Collection Districts (CDs) which together form areas with a population of at least around 300 Indigenous persons. Indigenous Areas aggregate to ATSIC Regions. Both Summary and Detailed Indigenous Profiles are produced for Indigenous Areas.
See also ATSIC Regions, Indigenous Location.
The ABS has implemented procedures tailored to the enumeration of Indigenous people living in discrete communities since the 1976 Census. The 1996 Census procedures build on those procedures and also include some additional procedures for enumerating Indigenous people living in urban areas. The Indigenous enumeration strategy is a nationwide undertaking. Central to this strategy is the appointment of the State Indigenous Manager (SIM). The role of the SIM is to lay the groundwork for a successful enumeration including gaining community acceptance for the Census and the recruitment of field staff. As in the past, Census Field Officers (CFOs) are employed to assist in these activities.
In certain Indigenous communities, an interview form designed to be appropriate to Indigenous culture is used. This part of the strategy is used in discrete communities where cultural or language concerns indicate the need. In these cases Census Field Officers recruit and train Community Coordinators to ensure accurate enumeration, as well as Indigenous interviewers to carry it out. As much as possible, Indigenous interviewers are used in collecting information from their own communities.
In other areas, Indigenous people are enumerated using standard procedures and forms. Indigenous assistants are available to assist if required.
An Indigenous Family is one where either the reference person or spouse is of Aboriginal/Torres Strait Islander origin. Clients may request tables on other definitions of Indigenous Family.
See also Family.
An Indigenous Household is a family household where any family in the household is defined as an Indigenous family or a lone person household where the lone person is of Aboriginal/Torres Strait Islander origin. Group households are not included. Clients may request tables on other definitions of Indigenous Household.
See also Household.
Indigenous Locations are single CDs or aggregates of CDs which have a population of around 80 Indigenous persons. They are the lowest level in the Indigenous geographic hierarchy (apart from individual CDs). Indigenous Locations aggregate to Indigenous Areas. Only Summary Indigenous Profiles are produced for Indigenous Locations.
See also ATSIC Regions, Indigenous Area.
This variable describes the industries in which employed people aged 15 years and over work. Questions 3135 on the 1996 Census form are used to obtain information for industry coding. They record the trading name of the respondent's employer, the workplace address of the employer, the type of industry, business or service carried out by the employer at that address as well as occupation details of the respondent .
The main index used in determining industry of employment in census processing is the Business Directory. This Directory is a listing of all known establishments in Australia involved in various economic activities carried out by companies, partnerships, sole proprietors, government departments, etc. For each establishment it contains the relevant Australian and New Zealand Standard Industrial Classification (ANZSIC) code and public/private sector code. The listing is obtained from the ABS Central Register of Businesses. This register is compiled and updated by the ABS from various statistical collections and other sources. Where the name of a business cannot be matched against the Business Directory, industry is coded according to the description of the type of industry supplied by the respondent in Question 36.
Each establishment is classified to a particular industry class according to the main activity of the establishment. An establishment can be a farm, a factory, a shop, a mine or any other place where some kind of economic activity is performed (i.e. goods produced or a service rendered).
There is not necessarily any relationship between the occupation of an individual and the industry in which he/she works; for example, a van driver for an establishment designated as in the insurance industry is employed in that industry, not in the transport industry. One establishment may employ many people in different occupations but they are all coded to the industry of the establishment.
Information on the type of industry carried out by the employer at the workplace has been gathered in each census since 1911. The name of employer and address of workplace has been collected since the 1954 Census. The inclusion of the industry topic in the Census provides a source of useful information on the regional distribution and structure of Australian industry, and allows investigation of structural change in industry over time. It also provides information on the characteristics of workers by industry (such as age, qualification and occupation), which is important for workforce studies, and the identification of migrant and other groups by industry.
The 1993 edition of the ANZSIC is used in classifying the responses given to the industry questions for the 1996 Census.
See also Australian and New Zealand Standard Industrial Classification (ANZSIC), Journey to Work, Working Population.
Industry Sector (GNGP)
This variable classifies employed people aged 15 years and over according to whether they are employed in the government or non-government (private) sector.
For the 1996 Census, industry sector has separate categories for Commonwealth, State/Territory and Local Government, Community Development Employment Program, and the private sector .
See also Australian and New Zealand Industrial Classification (ANZSIC), Community Development Employment Program (CDEP), Industry (INDP), Journey to Work, Working Population.
The completed census forms are delivered to the Data Processing Centre (DPC) in Sydney for processing as quickly as possible after census night. The first processing stage in the 1996 Census input processing strategy is the Precapture Process which undertakes to:
- ensure forms are sorted into correct sequence;
- ensure that key fields have been marked by respondents and/or collectors;
- check field procedures, including a check to ensure that dummy forms have been raised correctly for prescribed non-contact or unoccupied situations;
- insert census forms that have been mailed back into the correct Collection District (CD) packs; and
The second processing stage is the Data Capture Process. The census forms are then separated by guillotine and fed through Optical Mark Reader (OMR) machines which capture the majority of data. On the 1996 Census form, data from 35 of the 48 questions are captured directly by OMRs. The raw data are subsequently reformatted into unit record files. At the completion of OMR reading, forms are glued into 4 cm pads to enable ease of handling in subsequent processes and to ensure forms remain in correct sequence. The forms are kept in packs, according to their Collection District number.
After Data Capture, the forms are used in Computer Assisted Coding (CAC), which codes responses for those fields which are not captured by OMR, or where the respondent has failed to correctly mark a question, or has answered 'Other' and given further information in writing.
Responses that the CAC coder cannot code are raised as 'queries' which in turn are resolved by Query Resolution staff who use secondary reference material. Once queries are resolved the file is automatically updated.
Quality control checks are constantly made for coding accuracy. A large amount of the editing process is automatic, with fields being reset according to other responses on the individual census forms. Again, edits are invoked twice, once for first release variables and then again for second release variables.
All census data are extensively validated before data are released.
All census forms are destroyed by pulping after validation of the data.
See also Derivations and Imputations, First Release Data, Optical Mark Recognition (OMR), Second Release Data.
- to reconcile forms with the collector's record book count.
See Dwelling, Type of Educational Institution Attending (TYPP).
Internal migration is the movement of people from one defined area to another. Information on internal migration within Australia is available from the Census.
The Census asks a series of questions relating to each person's usual address. The indicative data from these questions are recorded as the Usual Address Indicator, Usual Address 1 Year Ago Indicator and Usual Address 5 Years Ago Indicator (UAICP, UAI1P, UAI5P).
Using the following variables, it is possible to identify the pattern of movement of people for the year prior to the census date, and for five years prior to the census date:
- SLA of Usual Residence Census Night (SLAUCP);
- SLA of Usual Residence One Year Ago (SLAU1P);
- SLA of Usual Residence Five Years Ago (SLAU5P);
- State of Usual Residence Census Night (STEUCP);
- State of Usual Residence One Year Ago (STEU1P); and
State is needed because SLA codes are unique only within a State.
Data collected in the Census only reflect the latest movement in the intercensal period, even though there may have been multiple movements during this period.
Household mobility indicators are also derived using this information. Note that visitors and households containing only visitors are excluded from these variables. The following two indicators are available for 1996 Census data:
- State of Usual Residence Five Years Ago (STEU5P).
- Household One Year Mobility Indicator (MV1D); and
The data for place of usual residence are used mainly in conjunction with household mobility indicators for detailed internal migration studies.
Such studies must be undertaken carefully; the points illustrated in the following cases should be noted.
Since the indicators are derived from usual residence at certain dates, only the net effects of any multiple movements between these dates can be derived. For example, in the case of a person who at 30 June 1991 had a usual place of residence in a South Australian rural area, and moved to Melbourne in 1993, but by 30 June 1995 was a usual resident of Adelaide, only the net South Australian country to city movement would be revealed.
In the case of 'out and back' movements, in which people moved away from a place of usual residence to live elsewhere, but then returned before the end of the reference period to the earlier address as a usual resident, no movement would be shown in the internal migration data.
The ABS produces quarterly information on internal migration in the publication Australian Demographic Statistics (3101.0) which is derived from a combination of information that is acquired from the census, and from administrative records from the
Health Insurance Commission.
See also Household Mobility, Usual Residence, Usual Address Indicator Census Night, One Year Ago and Five Years Ago (UAICP, UAI1P, UAI5P).
- Household Five Year Mobility Indicator (MV5D).
See Internal Migration.
Introduced Random Error
Many classifications used in ABS statistics have an uneven distribution of data throughout the categories of the classifications used. For example, the number of people who report themselves as being Anglican or born in Italy is quite large (4,018,800 and 254,800, respectively, in 1991), while the number reporting a religion of Buddhism or birthplace of Chile (139,800 and 24,200, respectively, in 1991), is relatively small. When religious denomination is cross-classified with country of birth, the number in the table cell who report religion as Anglican and country of birth as Italy could be small, and the number of Buddhist Chileans even smaller. These small numbers increase the risk of identifying individuals in the statistics.
Even when variables are more evenly distributed in the classifications, the problem still occurs. The more detailed the classifications, and the more of them that are applied in constructing a table, the greater the incidence of very small cells, and the greater the chance of individuals being identified in census output.
Considerable care is taken in the specification of tables to minimise the risk of the identification of individuals. In addition, a technique of unbiased random adjustment of cell values has been developed which allows very large tables for which there is a strong demand to be produced, even though they contain numbers of very small cells. These adjustments are small introduced random errors. They result in cells with very small values being insufficiently exact for any identifiable data to be exposed, while the information value of the tables as a whole is not impaired.
Because the column, row and person totals, sub-totals and overall totals in summary tables are derived after the random adjustment process is applied, the total would also contain random error if any of the cell components in a table were adjusted. Although each of the tables of this kind is internally consistent, comparisons between tables which contain similar entities may show some minor discrepancies.
In addition, because the tables at different ASGC levels are adjusted independently, tables at the higher ASGC level are not be equal to the sum of the tables for the component ASGC units.
It is not be possible to determine which individual figures have been affected by random error adjustments, but the small variance which may be associated with derived totals can, for the most part, be ignored.
Random adjustment of the data are considered, at present, to be the most satisfactory technique for avoiding the release of identifiable census data. All data output is useful, but it is subject to disturbance from original reports, in such a way to conserve information value while preserving anonymity.
No reliance should be placed on small cells. Even without random error, firm reliance should not be placed on small cells since possible respondent and processing errors have greatest relative impact on small cells.
Some small cells have been randomly altered to zero. Caution should be exercised in deducing that there are no people or households in an area having certain characteristics.
Many different classifications are used in census tables and the tables are produced for a variety of geographical areas. The effect of the introduced random error is minimised if the statistic required is found direct from a tabulation rather than from aggregating more finely classified data. Similarly, rather than aggregating data from small areas to obtain statistics about a larger standard geographic area, published data for the larger area should be used wherever possible.
When calculating proportions, percentages or ratios from cross-classified or small area tables, the random error introduced can be ignored except when very small cells are involved, in which case the impact on percentages and ratios can be significant.
See also Confidentiality, Random Perturbation of Table Cells, Table.
See Child, Family, Number of Children Ever Born (TISP).
See Child, Family, Dependent Child, Number of Children Ever Born (TISP).
This page last updated 20 January 2006