2901.0 - Census Dictionary, 2006 (Reissue)  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 13/07/2007  Reissue
   Page tools: Print Print Page  
Contents >> Short Definitions and Classifications - 2006 >> Relationship in Household (RLHP) - Characteristics 2006

Relationship in Household

On this page:
Image of Question
Quality Statement


This is a key variable at the person level. It is used to record the relationship of each person in a family to the family reference person or, where a person is not part of a family, that person's relationship to the household reference person. More Detailed Description

Image of Question

2006 Household Form - Question 5


Applicable to: Persons present in the household on Census Night

Husband, Wife or Partner
Lone parent
Child under 15
Dependent student
Non-dependent child
Other related individual
Non-family member
Visitor (from within Australia)
@@ Not applicable
VV Overseas visitor

Total number of categories: 31

More Detailed Description
Quality Statement - Relationship in Household (RLHP)

There are many aspects which can affect the quality of Census data; the following information should be considered when viewing data on Relationship in Household (RLHP).

In the 2006 Census, data on the relationships people have with others in the same dwelling is mainly derived from Question 5 on the Census household form, which asks for each person’s relationship to Person 1 on the form. This data is captured automatically as check box responses in 92% of cases, with the remainder obtained from written responses to the question.

During the processing of Census data, families and households are identified and created based around a ‘family reference person’, and Relationship in Household (RLHP) is derived for each person in the dwelling. In over 95% of cases Person 1 is the family or household reference person. For the remaining cases, a different person is selected to better allow relationships to be identified or because more than one family was identified as living in the household.

For many households, identifying relationships to assist the coding of family or household structure for that dwelling is quite straightforward. However, across the community, a wide variety of living arrangements exists and family structures can be complex and dynamic in nature, and so the quality of family data in the Census is partly dependent on people’s ability to describe these relationships within the constraints of the generalised questionnaire format required by a Census. Reporting relationships in respect of Person 1 only, can make it difficult to establish all the relationships which exist in a household, or to identify whether more than one family is living in the dwelling. In some cases, additional information such as name, usual residence and marital status is also used during data processing to help determine these relationships. Priority is given to identifying those relationships which form a ‘family nucleus’, i.e. partnerships and parent/child relationships.

In Census data, a maximum of three families are able to be identified within a single dwelling, in accordance with existing ABS standards. While this may generally have only a small effect on the identification of relationships within a dwelling, the impact may be more significant among population groups which are more likely to live in multi-generational households or with larger numbers of extended family members.

In cases where some members of a household are away from home on Census Night, members of the family nucleus (partners, parents and children) and unrelated persons who were temporarily absent on Census Night (and identified as such in Question 53 on the Census Household form) are taken into account when deriving RLHP. This allows for the identification of some families, and also for distinguishing between lone person and group households. However, during processing of 2006 data there were 29,985 dwellings where boarders/lodgers and unrelated household members were listed as temporarily absent on the Census form and were erroneously excluded when RLHP was derived. This resulted in an additional 29,985 persons incorrectly being recorded as lone persons instead of group household members. This error has also affected the dwelling level variables CPAD (Count of persons temporarily absent from household) and HHCD (Household composition). These dwelling level variables have been amended, reducing lone person households and increasing group households by 29,985. However, it is not possible to make a similar amendment to RLHP. This explains the discrepancy in numbers when comparing "Lone persons" in RLHP with "Lone person households" in the amended HHCD.

In addition the ABS has undertaken some preliminary evaluation of the quality of Census data which suggests that around 35% of family members and over 40% of unrelated people who are temporarily absent from their usual address on Census Night are not included in the relevant section of the Census form (Q53). This is only partly explained by the reporting limitation of three persons per household.

For the 4.1% of persons imputed into dwellings for which no form was received, relationship data is set to not applicable.

The 2006 Census shows a very small proportion (0.01%) of persons aged 80 years and over as living with a parent (that is, as a non-dependent child). The ABS plans to do further analysis for this group to identify any data quality issues which may be affecting this result.

The ABS aims to produce high quality data from the Census. To achieve this, extensive effort is put into Census form design, collection procedures, and processing procedures.

There are four principal 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.

When completing their Census form, some people do not answer all the questions which apply to them. In these instances, 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 processing of information from Census forms is now mostly automated, using 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.

The Census form may be completed by one household member on behalf of others. Incorrect answers can be introduced to the Census form if the respondent does not understand the question or does not know the correct information about other household members. Many of these errors remain in the final data.

More detailed information on data quality is available in the 2006 Census Dictionary (cat. no. 2901.0), in the section titled Managing Census Quality.

Previous PageNext Page