03 Section B Option 3, Key question 3 - Age, cohorts and religious change (part 3)

Measuring religion by age

An age pattern depends on which aspect of religion we measure.

Religious affiliation means identifying with a religion; religious practice means activities such as worship or prayer; religious belief concerns convictions such as belief in God or an afterlife. A person can identify as Christian because of family heritage while rarely attending church. Someone without a religious affiliation can hold supernatural beliefs. These are different dimensions of religiosity, or religious commitment, rather than interchangeable counts of the same thing.

The following figures are from the 2018 British Social Attitudes survey of adults aged 18 and over in Great Britain, reported by David Voas and Steve Bruce in 2019. They are percentages within each age group, combining men and women.

The attendance measure excludes special occasions and includes attendance less often than annually; it is not a weekly-attendance rate. The God-belief measure combines respondents saying they have always believed with those saying they believe now but did not previously. NatCen, BSA 36, Table 9 and Appendix Table A.3.

The oldest group's affiliation exceeds the youngest group's by 22 percentage points, compared with four points for attendance and nine for belief. The direction is similar, but the size of the gap depends on the indicator. Reporting a 22-point gap in weekly worship would therefore misrepresent this evidence. Neither a religious label nor this broad attendance category measures the intensity of every person's faith.

These estimates come from a probability survey of private-household adults, using a design with known selection chances. Non-response and the exclusion of institutional residents limit coverage; the belief question also had a smaller sample. We should use the figures as dated estimates, not exact descriptions of every member of an age group.

Age, cohort and period effects

Three different processes can produce an association between age and religion.

An age or life-course effect is change associated with growing older or passing through life stages. For example, becoming a parent might encourage someone to return to a religious community for their child's upbringing. Retirement might offer more time for attendance, while limited mobility could reduce it. These are possible mechanisms, not a rule that ageing increases religious commitment. Bereavement may lead one person towards religious consolation and another towards doubt.

A cohort effect is a lasting difference associated with the circumstances in which a generation grew up. People raised when religious institutions were more central to family and community life may retain more religious identification than people socialised in a less religious setting. The older generation can remain more religious even if neither generation changes its level of commitment in adulthood.

A period effect is an influence of a historical time or event on people across age groups. In an original hypothetical example, a new restriction on gatherings reduces attendance among both young and old during the same year. This does not establish a fall in their beliefs. Period effects need not affect every group equally or last forever.

Why a snapshot cannot decide

A cross-sectional study compares people at one time. In an original illustrative survey, 60% of people aged 70 and 20% of people aged 20 report a religious affiliation. This is compatible with a life-course account: affiliation might rise as people grow older. It is also compatible with a cohort account: the two groups may have started adulthood at different levels and retained them. The 70-year-olds are not the same people as the 20-year-olds followed for 50 years.

Age, historical period and birth cohort are linked: someone aged about 20 in 2020 belongs to a different birth cohort from someone aged about 20 in 1980. Following people as they age also means observing them in a later period. Researchers therefore need comparisons over time and an argued explanation, not a mechanical assumption that any association is caused by ageing.

Socialisation and generational replacement

Religious change can occur because new generations enter the adult population with different experiences and commitments.

Socialisation is the process through which people learn norms, values and ways of understanding the world. Religious socialisation can involve family prayer, festivals, attendance, stories and relationships with a congregation. Repeated participation can make a religious identity familiar and meaningful. Transmission is not automatic: children can interpret traditions differently, reject them or adopt a religion their parents did not follow.

Institutions beyond the family can reinforce or weaken this process. Consider an original hypothetical town where a congregation once organised children's groups, leisure activities and support networks. If those activities close and families develop other networks, children may have fewer everyday connections with religion. Later, some of those children may raise their own children without religious routines. This is a possible feedback mechanism linking institutional change to family transmission. It must be investigated rather than assumed from a single age table. Religious youth groups, schools or online communities could instead sustain commitment in another setting.

Generational replacement means that the population's composition changes as older cohorts die and younger cohorts enter adulthood. If the entrants are less religious than those leaving, the overall rate can fall without any surviving adult abandoning religion.