Review of 2021 Census Topics

This is one of several submissions made on behalf of this agency.

The focus of this submission is the question on number of children ever born.

This submission recommends that the ABS

• retains the format of the question as used in the 2016 Census
• returns to asking the question every second census, so do not include in the census again until 2026

Data from the census on children ever born has been used extensively by researchers and policy makers to understand more about the socio-economic correlates of fertility in Australia.

This census question is the only source of information about women who had their children while living overseas and who have subsequently immigrated to Australia.

This Census question is also the only readily available source to examine childlessness in Australia. The policy implications for this issue are as important, in the context of population ageing and declines in fertility since the baby boom-peak in the 1950s and 1960s.

A better understanding of fertility differentials will greatly help in the development of assumptions for population projections developed by Redacted text

Conventional fertility analyses in Australia use births data is available from vital registrations and the perinatal data collections of Health Agencies throughout Australia. Collected for administrative and civil purposes, the shortcomings of both data sources are:
• they provide limited demographic information about mothers living in Australia
• no data is available for women who had their children while living overseas
• fertility patterns are only available for indigenous mothers. No data is available for women of other ethnic or social groups.
• This data does not readily allow us to look at cohort fertility across generations of women living in Australia.
• No information is available about the socio-demographic characteristics of all mothers living in Australia, including place of residence, country of birth, education, occupation and income.

A census question already exists for collecting this information

This question has been asked, in various forms, since 1911.

At the 2016 Census the level of non-response to this question (excluding imputed persons) was 3.1 per cent (see http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/by%20Subject/2900.0~2016~Main%20Features~Item%20non-response~10036 )

Other census questions had higher levels of non-response in 2016, which suggests that the women are willing to answer this question.

The question wording, and data processing has already been established by the ABS.

There is an ongoing need for research and analysis about childbearing patterns in Australia. Childbearing levels have changed considerably over the last century and may continue to change in the future.

This question is the best source of data for all women currently living in Australia

To my knowledge this detailed information is not available from any other source.

Births data is available from vital registrations and the perinatal data collections of Health Agencies throughout Australia. The shortcoming with both data sources is that they are administrative data. They do not contain detailed socio-demographic information about the child, the mother or the family into which these children are born.

This question is normally included in every second census, so is collected every 10 years. This is sufficient to monitor changes in the fertility patterns in Australia and its socio-demographic correlates.
However administrative decisions made for the last two censuses has meant it was included in the previous 3 censuses.
To re-establish temporal consistency in the time series of data on the number of children ever born I recommend that the ABS reverts to the pattern of every second census as is the normal practice. This means not including it in the 2021 Census, but asking it again in 2026.