Australia - Asia relations, Religion

Why the census fails to capture the religious identities of Asian Australians


The dismantling of the White Australia Policy in the second half of the twentieth century ushered in a new multicultural Australia, whose diversity is reflected nowhere more clearly than in the populations’ increasingly heterogeneous religious landscape. As Gary Bouma points out in his insightful book titled Australian Soul: Religion and Spirituality in the 21st Century, there were ‘more Buddhists than Baptists, more Muslims than Lutherans, more Hindus than Jews and more than twice as many Sikhs as Quakers’ at the beginning of the new millennium.

The Australian census is one of the primary tools used by academics and policy makers alike to track trends in Australian society. The census is carried out by the Australian Bureau of Statistics (ABS) every five years to collect data on the key characteristics of Australians and their place of residence at the time of the Census night. The census data is regarded by many people as one of the most valuable sources of empirical data for the analysis of Australians’ religious affiliations.

We benefited much from the census data during our study into the religious affiliations of the Chinese community in Australia. Our use of census data, however, also made us aware of some of its limitations, particularly for capturing the religious identities of Asian Australians. For those seeking to better understand Asian Australians, we must remember that while the census is very important, it is a blunt instrument, and certain shortcomings mean that it is unable to capture the real nuance of Asian Australia’s religious identity.

Coding problems

The Australian Census is expensive. Moreover, given that the census is designed to collect ‘key characteristics’ of Australian residents, their families, and their places of living, it is understandable that there are only so many questions can be included in the Census Household Form (i.e. the Census Questionnaire). In the census, there is only one single question on the topic of religion: ‘Q19: What is the person’s religion?’ This question alone has generated interesting data which informed academic publications and public discussions. However, coding problems have meant that we are significantly disadvantaged when it comes to understanding the religious identities of Asian Australians.

While any religion can be given as an answer to Q19 in the census, it is only data which fits into the predefined categorisation system for religions which is truly captured. Although this system covers a great number of well-known religions and denominations, it overwhelmingly reflects the Western social context of the Australian Standard Classification of Religious Groups, which was developed by the ABS to collect, aggregate and disseminate data relating to the religious affiliation of the Australian population in the Australian Census, the ABS surveys, and many relevant administrative data collections conducted by the federal and state governments in Australia. Unfortunately, this official classification of religious groups provides far more detail in the Christian code category and tends to neglect the important nuances in other religious traditions.

To be more precise, within the Christian category of the census’ coding system, there are almost 100 different four-digit codes, each code representing different denominations, from the well-known denominations like the Western Catholic Church and the Anglican Church of Australia to smaller groups like the Christadelphians and Gnostic Christians.

This level of detail stands in stark contrast with the codes for Buddhism, Hinduism, Judaism and Islam, where the system makes no denominational distinctions. That means, for example, if a respondent answers that they are a Zen Buddhist or Tibetan Buddhist, they will be recorded with the same code, even though the differences in theology and religious practice between Zen Buddhists and Tibetan Buddhists is just as significant as the difference between Catholics and Christadelphians for instance.

Without specific codes for the denominations of religions such as Buddhism and Islam, this data on non-Christian denominations is lost, making it difficult to assess the relative strength and size of these groups in contemporary Australia. This is particularly problematic for those studying Asian Australians because these communities tend to have a higher proportion of believers in non-Christian religions. Moreover, in an increasingly multicultural society, understanding minority communities is an absolute necessity, and the coding behind the census should be adjusted to reflect the true diversity of contemporary Australia.

Unfriendly to Multi-faithers

The Census Household Form does not explicitly stop people from listing more than one religion. However, upon analysis of the data we found that the number of respondents to Q19 matches exactly the overall number of entries for religious affiliations. That means one of two things, either that there are no multi-faith people in Australia whatsoever, or more likely, that they were not correctly recorded.

This is clearly problematic for people participating in the activities of multiple religions or religious organisations. Importantly, mutual exclusivity of religious affiliation is culturally specific. For example, in the case of the Chinese community, it was only in the late nineteenth century that this conception of religion was introduced. Even today, it is not uncommon for a Chinese Australian to express faith in both Buddhism and Daoism.

Designing a survey question which effectively navigates cultural differences is by no means an easy task. We know that even understandings of words like religion or belief differ culturally. Existing research shows that, in China for example, careless translation of belief (xinyang) or religion (zongjiao) can change survey results completely. While language and cultural problems seem more intractable, at the very least we think the quality of the census data could be improved by simply removing the outdated assertion of mutual exclusivity of religious belief, which does not reflect the beliefs and behaviour of all Asian Australians.

‘Low-resolution images’

Like all structured surveys, the census primarily focuses on straightforward and measurable information. As a result, Q19 simply gauges people’s self-identification with a religion, but there are no common benchmarks for what qualifies as having a religious belief. In other words, the current census data can only provide ‘low-resolution images’ on the religious identities of Asian Australians.

For instance, some may see attendance of religious services as a prerequisite for group membership, while others may never attend and still self-identify within the group. Information on the frequency of worship, the level of commitment, and data on financial contributions are some important indicators for the future prospects of the religion which are not captured in the census.

Without such detailed information about the religious lives of Australian residents, our understanding is likely to lag behind reality, resulting in a skewed sense of Australia’s religious identity. This, again, is particularly problematic for understanding the religious identity of Asian Australians. For example, it is not uncommon for an Asian Australian to claim themselves as ‘Buddhist’ without ever going to a temple in Australia. On the other hand, there are also many highly committed Buddhists in Australia who regularly visit temples and devote themselves in worship and service of Buddhist organisations.  Without the information on the frequency of worship or the scale of financial contributions, the census is unable to capture the true level of religiosity within different subgroups of the population.

What can be done?

To better capture the religious identities of Asian Australians there are at least two potential ways forward.

One straightforward approach for specialist researchers who  study the religious affiliations, practices and identities of Asian Australians is to focus on different resources. Rather than restricting themselves to the problematic census data, these researchers could focus efforts on other surveys that have specifically-designed components to capture the religious affiliations, identities, and behaviours of respondents. For example, as opposed to the single question on religion in the census form, the 2016 Australia Church Life Survey — an online survey conducted in early December 2016 — includes far greater detail on the religious and spiritual lives of Australians. However, this survey only covers a sample of 1258 subjects, leaving it susceptible to the risk of not capturing less commonly practiced religions.

The census, on the other hand, offers unparalleled comprehensive coverage of the Australia population thanks to its statutory nature. Therefore, in comparison with developing standalone surveys, a more effective approach is to improve how the information is captured and coded in the census.

Given the statutory nature of the census and the huge administrative cost associate with it, it is understandable that no decision on adding or altering a question in the Census Household Form can be taken lightly. It has become standard practice that new topics and questions can only be added to the Census Household From after an extensive process. That said, even if it is not practically feasible to add more questions on religion in the census, at the very least, major schools of Buddhism (such as Mahayana, Theravada, and Tibetan) and major denominations of Islam (such as Sunni and Shia) should be added as possible options alongside the six Christian denominations that are already on the census form (Catholic, Anglican, Uniting Church, Presbyterian, Greek Orthodox and Baptist).

Even if nothing can be changed on the census form due to financial and practical restrictions, a more sophisticated coding system should be developed if we want to capture the nuances of denominational breakdown within non-Christian beliefs. These changes will allow us to better understand the religious identities of Asian Australians, which constitute an important part of contemporary Australia’s multicultural society.


Featured Image: Australian census forms, from flickr user Ruben Schade.


About Yu Tao and Theo Stapleton

Dr Yu Tao is a lecturer in the School of Social Sciences at the University of Western Australia, where he teaches and coordinates Chinese Studies. His current research investigates the interactions between religious groups, civic organisations, and local governments in contemporary China and overseas Chinese communities.

Theo Stapleton is a research assistant in the School of Social Sciences at the University of Western Australia. His current research focuses on the strategies that religious groups apply to extend their social influence in contemporary China.

11th July, 2018

Summit a Comment

Your email address will not be published. Required fields are marked *