Church Attendance in the United States 93
very well attend services more or less regularly, if not weekly – has remained stable. If the
standard survey questions actually tap a person’s religious identity more than their lit-
eral church attendance, and if the true trend has been for people to attend less often but
still regularly enough to consider themselves religiously committed, then this would
produce stability over time in the standard surveys even in the face of real decline in
weekly attendance. The basic idea is that a real decline in attendance, if it takes the form
of many people shifting from weekly to, say, monthly attendance, might not register
in standard surveys.
This is, admittedly, speculation, but it is plausible speculation, and we find it difficult
to develop a similarly plausible account in the other direction – one that would explain
why time-diary evidence shows decline over time if stability is the true picture. All
in all, although it is not yet possible to say that the new research has definitively
established that religious participation in the United States has declined, our view is
that the evidence and arguments for decline are, at this writing, more compelling than
the evidence and arguments for stability.
The emerging picture, then, is of an American society in which, since the 1960s –
but not before – people engage in less and less religious activity. This is occurring, it
seems, without any decline in belief in the supernatural or concern about spiritual-
ity. Interestingly, this pattern is not limited to the United States. On the contrary, it
characterizes many countries around the world. Although advanced industrial societies
vary quite widely in their aggregate levels of religious participation and religious belief,
they show basically similar trends over recent decades: Down on religious participa-
tion, stable on religious belief, and up on thinking about the meaning and purpose of
life (Inglehart and Baker 2000). Some, although not all, ex-Communist societies show
increases in both participation and belief, but that is a subject for another essay.
CONCLUSION
The current state of knowledge about religious service attendance in the United States
should not comfort those who expected modernity to be fundamentally hostile to
religion. It seems that religious participation was either stable or increasing for two
centuries, including the late nineteenth and early twentieth century decades during
which the United States changed from a predominantly rural to a predominantly ur-
ban society. Moreover, many conventional religious beliefs remain popular and show
no sign of decline even now. At the same time, however, what we know about church
attendance also should not comfort those who believe that there has been no impor-
tant change, or that social changes associated with modernity do not have potentially
negative consequences for religious belief and practice. It seems likely – although not
yet definitively established – that religious participation has declined in the United
States, as in many parts of the industrialized world, over the last three or four decades.
Cross-national evidence also indicates that certain aspects of “modernity” – more in-
dustrial employment and higher overall standards of living – are indeed associated with
less traditional religious belief among people (Inglehart and Baker 2000). Beware simple
tales about secularization, but also beware wholesale rejections of secularization.
Although trends in church attendance are intrinsically interesting, we also know
that focusing exclusively on religious practice – or even on the combination of religious