In this post Katherine Sissons reflects on the Census 2011. She discusses the statistically significant relationship between gender and reliosity in female and male babies and young children (aged 0-4). What impact might this have for understanding the ‘nones’ and the way that the gender balance of adult religiosity and non-religiosity has on the next generation?
The 2001 Census for England and Wales was only the second census in Great Britain to include a question about religion (the first was in 1851). Analysing the data from the 2001 Census, sociologists found that baby girls were more likely than baby boys to be ‘Christian’(Voas & McAndrew 2012). Does this pattern persist in the 2011 Census?
The question on religion is the only voluntary question in the Census, and in 2011, 7.2% of the population chose not to answer it. However, this still means that we have data on religious affiliation for the majority of the population, including children. Here I focus on the two most popular responses: 59.3% of the population identified as Christian, and 25.1% selected “no religion”.
The table below shows the number of male and female children in the 0-4 age group who were recorded on the Census as either “Christian” or “no religion”.
At first glance, it seems that there are more males in every category – there are more male “nones” as well as more male Christians. That’s true – but only because the 0-4 age bracket includes about 80,000 more males than females overall. We generally think of the human sex ratio as 50:50 (an equal chance of giving birth to a boy or a girl). However, in most human populations about 104 males are born for every 100 females. The sex ratio usually remains uneven – with slightly more males than females – throughout childhood.
What we really need to know is the proportion of male children assigned to any one religion, compared with proportion of female children assigned to that religion. In the 2011 Cesus, 43% of male children aged 0-4 were reported to be “Christian”, compared to 44% of female children of the same age. 35% of male children aged 0-4 were reported to have “no religion”, compared to 34% of female children of the same age. In other words: if female children were treated the same as their male counterparts, there would be 10,000 fewer females in the Christian” category (and 10,000 more with “no religion”).
These differences may seem too small to be important. However, over a large population (3.4 million children in this age bracket) those small differences between the male and female children are statistically significant: there is a real difference between the male and female children, and that difference is too large to be attributed to chance.
To understand why this pattern occurs, we need to understand how the Census data is collected. Residents of England and Wales were asked “What is your religion?” and invited to respond by selecting a religion from a list, shown in Figure 2 (below).
Clearly, children aged 0-4 do not complete census forms for themselves. Rather, parents or guardians complete the census on their behalf. It is not necessarily the case that young girls are really more likely to be Christian than baby boys. Rather, it seems some parents are slightly more likely to identify young girls as Christian, compared with baby boys, who are slightly more likely to be assigned “no religion”.
Voas and McAndrew suggest that the religion assigned to a young child is influenced by the religion of the child’s parents. They suggest that in a family where both parents are Christian, male and female children are likely to be identified as Christian. Likewise, in a family where both parents select “no religion”, both male and female children are likely to be assigned “no religion”. However, when the parents have different religions, children tend to be assigned the religion of the same-sex parent: young girls are assigned the same religion as their mother, whilst young boys are assigned the same religion as their father (Voas & McAndrew 2012) [i].
So far, we have only considered the religion of males and females in the 0-4 age group. Figure 3 (above) shows that males in every age group are more likely than females to select the “no religion” option on the census form. This means that if a child has one religious parent and one Christian parent, more often than not it is the mother who is the Christian parent. If parents assigning a religion to their children apply the rule “like mother like daughter, like father like son”, young girls are more likely than their male counterparts to be identified as Christian because their mothers are more likely than their fathers to be Christian.
Are parents being unfair?
Since there are statistically significant differences between young boys and young girls at the population level, it is tempting to conclude that some parents treat their children differently. In fact, we can’t know for sure whether this is the case. The average family size in the UK is relatively low – 1.7 dependent children in each family (Office for National Statistics 2013). This means that many families will only have children of one sex (all girls or all boys). In those cases, it is not that one set of parents treats their boys and girls differently. Rather, parents who only have boys behave differently to parents who only have girls. Even in families that do have male as well as female children, the parents may feel justified in assigning different religions to those children. Barring multiple births (twins, triplets), there will be an age difference between children, and parents may feel that, for example, a four year old girl has expressed a belief in Christianity that her nine month old brother cannot. In those cases, parents are unlikely to be aware that their behaviour forms part of a wider sex-related pattern.
Of course, this raises yet another question: why are men more likely than women to say that they have “no religion”? This phenomenon is not confined to Britain but has been identified in numerous other surveys (Hayes 2000). Some researchers have looked to evolution to explain these sex differences (Norenzayan et al. 2012), whilst others favour more social or cultural explanations (Trzebiatowska & Bruce 2012). Without a consensus on this issue, the puzzle of why young boys are more “non-religious” than young girls remains, at present, only partially solved.
[i] Unfortunately, same-sex couples with children make up a relatively small proportion of households in England and Wales, and so could not be included in Voas and McAndrew’s analysis.
Hayes, B.C., 2000. Religious independents within Western industrialized nations: a socio-demographic profile. Sociology of Religion, 61(2), pp.191–207. Available at: http://socrel.oxfordjournals.org/cgi/doi/10.2307/3712285.
Norenzayan, A., Gervais, W.M. & Trzesniewski, K.H., 2012. Mentalizing deficits constrain belief in a personal God. PloS one, 7(5), pp.1–8. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3364254&tool=pmcentrez&rendertype=abstract [Accessed January 29, 2014].
Office for National Statistics, 2013. Family Size in 2012, Available at: http://www.ons.gov.uk/ons/rel/family-demography/family-size/2012/family-size-rpt.html.
Trzebiatowska, M. & Bruce, S., 2012. Why are women more religious than men?, Oxford: Oxford University Press.
Voas, D. & McAndrew, S., 2012. Three puzzles of non-religion in Britain. Journal of Contemporary Religion, 27(1), pp.29–48. Available at: http://www.tandfonline.com/doi/abs/10.1080/13537903.2012.642725 [Accessed January 22, 2014].
Katherine is a doctoral student in the department of anthropology at the University of Oxford. She is interested in the variety of lived experiences of nonreligion in the UK. Before starting her doctorate she lived for a while in Vanuatu and is still interested in religion and religious change in Melanesia. Katherine is also Assistant Editor for NSRN Online, with responsibility for the commissioning and publication of book review and media related blog posts for the NSRN blog.