The question on religion has always been optional in the Australian Population Censuses, and there have generally been between 10 and 12 per cent of the population who have chosen not to answer the question. It may be surmised that their reasons for not answering are many and varied. Some people with high commitment to religious groups have not answered the question because they believe in the separation of religious affairs and government affairs, and thus they believe the government has no business knowing their religious involvements. Many others have failed to answer the question because they have no specific religious affiliation or involvement in a religious community.

The numbers not answering the question rose sharply between 1911 and 1933. In some ways, World War I marked the end of Christendom. The brutality of the war between nations with a Christian heritage convinced many that the Western world had not reached the heights of civilisation and moral superiority claimed for it at the end of the nineteenth century. For many, World War I raised questions about the validity and effectiveness of Christianity as the moral foundation of the Western world.

Throughout the century there have been small numbers describing themselves as humanists, rationalists, agnostics or atheists. In the 2001 Census,
17 565 described themselves as agnostics
24 466 as atheists
5 041 as humanists, and
1 618 as rationalists.
These have been coded by the Australian Bureau of Statistics as ‘no religion’ and are included in the ‘no religion’ group in the graph of change over time.

In 1971, the instruction to describe oneself as having ‘no religion’ if one had no religion was explicitly included in the Census. This was the major reason why the numbers of ‘no religion’ climbed steeply at that point, while the numbers not answering the question fell.
Between 1971 and 1996, the ‘no religion’ group rose risen steadily both in numbers and as a proportion of the population while the numbers not answering the question have varied somewhat from one census to another.

The increase in the ‘no religion’ group has sometimes been taken as an indication of growing secularisation in Australia, that people are turning more to science and technology and away from religion and other spiritual resources. Other recent research has shown that this is not a valid interpretation. The World Values Surveys between 1983 and 1995 showed that the proportion of the population describing themselves as ‘religious persons’ actually rose in that period, despite the fact that more described themselves as having ‘no religion’ and larger proportions said they never attended church services.

The rise in ‘no religion’ can be explained more adequately in terms of the proportion which does not identify with a particular religious group or denomination. Some of these people see themselves as ’spiritual’ although not belonging to a religious organisation. Some see themselves as using a range of religious or spiritual resources while not identifying themselves with a particular group. There has been rise in the social acceptability of describing oneself as having no religion.

In 2001 although the numbers that said they had ‘no religion’ increased, the group declined as a proportion of the population representing 15.9 per cent which proportionately 1.1 per cent fewer than in 1996. Thus while the ‘no religion’ community has not decreased in itself, a higher proportion of the overall population now identifies with a religious tradition. Part of this proportional decline is probably owing to higher levels of religious affiliation among immigrants to Australia, particularly to Buddhism, Hinduism and Islam. There is also a great deal of evidence to suggest that the decline in institutional religion and the rise of a more individual–focussed society has led to more people exploring alternate spiritualities and picking and chosing among their different elements and beliefs. These people would not classify themselves as having no religion but may struggle to define what their religion was (Hughes 2002).