Data use considerations
Priority is given to identifying those relationships which form a ‘family group’, i.e. partnerships and parent/child relationships. For many households, identifying relationships to assist the coding of family or household structure for that dwelling is quite straightforward.
For dwellings containing blended families or multiple generations of families, the nature of the family structure can be complex. The quality of family data in the Census is partly dependent on people’s ability to describe relationships within the constraints of the questionnaire format required by a Census.
The reporting of relationships to Person 1 can sometimes mean that closer relationships between other people in the household are lost. For example, reporting ‘niece’ (of Person 1) instead of ‘daughter’ of Person 2. In other cases, respondents have reported a relationship that is the reverse of what the question is intended to capture, for example, reporting 'grandparent' instead of 'grandchild'. While in many cases these errors are recognised and rectified, some are automatically accepted and cannot be reviewed.
More complex or unusual relationships are not automatically accepted by the processing system and are presented for manual coding. In some cases, additional information is used to determine relationships, such as:
- name
- usual residence
- marital status
- number of children given birth
An overcount of step children and an undercount of natural or adopted children for couple families has been identified for Western Australia, New South Wales, Queensland, Northern Territory and Other Territories. Victoria was partially affected. South Australia, Tasmania and the Australian Capital Territory remain unaffected. Only step children in couple families where the spouse was temporarily absent were impacted. These children have been incorrectly categorised to the step child of male parent or female parent categories for both Child type (CTPP) and Child type (including grandchildren) (CTGP) variables. These children have also been incorrectly categorised to step child under 15, student step child or non-dependent step child for both Relationship in household (RLHP) and Relationship in household (including grandchildren) (RLGP) variables. This also means these families have been incorrectly categorised as step families in the Family blending (FBLF) variable.
This variable does not have a non-response rate as it is created during Census processing by using responses from more than one question on the Census form.
