Is it possible to know
how many people in Canada and the U.S. are religious? Ryan Burge, who studies
religion in the U.S., says no. But some Canadian scholars disagree—although they
note we could do a better job at asking questions about religiosity and
spirituality. And they agree that the number of people who claim to be
religious is declining. Read
about it in my recent column.
Also find below the
full answers from the scholars who responded to my query about this topic.
Sam Reimer,
Crandall University
I think Ryan is saying
what we all know—that different polls/surveys give different results, and it
probably good to find some middle ground based the different results. What we
do know is that these results, though never perfect, are better than our best
guess! What is clear is the percentage of people who have no religious identity
is increasing in both NA countries, and that younger generations are more
likely to be nones than older generations (see Sarah WL new book).
Obviously, different
polls use different methodologies, different wording in the question, which
will give different results. Differing results can also be related to which
religious groups are listed, the order of the questions on the survey, and
other factors. It seems like Canadians
are more likely to select a religious identity in the census than on an online
survey. The census has less nones. This may be because they are more likely to
want to identify according to their family's historical religion on the census,
or panels tend to have more young respondents, not sure.
Joel Thiessen,
Ambrose University
I agree that we lack
objective metrics for ‘religiosity.’ Still, I’m partial to a relatively
objective metric of ‘religious nones’ to simply mean someone who does not
identify with a religion or religious group. This approach privileges
affiliation as the lowest common denominator, separate from the varied beliefs
or behaviours that may or may not accompany a religious none.
However, even when
measuring ‘religiosity’ as a category, consistent measures – however imperfect
– still yield useful descriptions and comparisons to track changes over time.
And yet, as Burge intimates, we still need to be cautious in our interpretations
of what these data reveal or not across time and space, mindful along the way
of different forms and expressions – and thus measures – of religiosity over
time.
Ideally, we should aim
to compare the constant measures of religiosity that survey researchers have
used over time (e.g., belief, behaviour, belonging, and even salience),
accounting for new/evolving measurements plus variations across religious
groups/traditions whenever possible and applicable.
Moreover, we must
gather and pay attention to high quality and multi-layered forms of qualitative
research that can offer a far more granular understanding of what quantitative
indicators mean or not. Nancy Ammerman and Lori Beaman are two great examples
of what difference this approach can have on our scholarly understandings and
interpretations of (non)religion.
I’m far less
comfortable with the assertion that “There’s no objective measuring stick for
us. So, everything is just a series of best guesses.” Guesses is an unfair and
misleading descriptor, reducing the role of legitimate experts in this space
(all of us!) to armchair fill-in-the-blanks. Just because we cannot draw upon
objective measures does not imply we’re just guessing.
Our work as scholars of
religion is set apart both in our ability to measure different aspects of
(non)religion and religiosity, however imperfect, and then to interpret those
data carefully and robustly. In the process, we must – and no doubt are – cautious
to not overstate what the data potentially mean or not. This latter
contribution of interpretation is part of what sets us apart from the armchair
guesser.
Sarah
Wilkins-Laflamme, University of Waterloo
I agree and echo Sam’s
and Joel’s thoughts. Any measure of a concept, including of religion, will
never be perfect. Additionally, there are many dimensions of religiosity to
measure. So it’s important to build good-quality measures, and to clearly state
what they are when presenting results from them.
And even though no one study
is ever perfect, a good-quality empirical study of (non)religion with a clearly
defined measure and a pretty representative sample is much, much better than
just a guess.
I proceed with survey/Census data on
(non)religion in the same way as many recommend working with election polls:
don’t rely on just one poll alone, but instead look at all the data being
gathered (and we’re fortunate that there is quite a bit out there), and
identify the trends that emerge (along with any discrepancies in results).
This
allows us to get a good sense of where trends appear in the various ways we
measure religiosity dimensions, and where findings are still mixed and more
study is needed.
John
Stackhouse, Crandall University
A few small points,
then, if I may.
1. In my work on
evangelicalism, I’ve long chided pollsters—going back to George Rawlyk’s 1990s
poll with son-in-law Andrew Grenville, working with Angus Reid at the time, and
recently with Cardus/Canadian Bible Society—that definitions of a religious group
should be “own-able” by representative members of that group. If the observer’s
definition, that is, materially differs from the participant’s definition—and
particularly from that of the leaders and spokespeople for that group—then a
yellow flag should go up.
(I published an article
a few [!] years ago to show that the massive scholarship on conversion that
both drew from and applied to Billy Graham generally misunderstood conversion
as taught by evangelicalism in general and by Graham himself—such that most sociologists
literally didn’t understand what they were looking at when they looked at Billy
Graham: "Billy Graham and the Nature of Conversion: A Paradigm Case,"
Studies in Religion/Sciences Religieuses 21 (1992): 337-50.
When it comes to
evangelicalism, to illustrate, I insist (most recently in Evangelicalism: A
Very Short Introduction) that all six descriptors ought to be present if one is
confidently identifying evangelicalism: Trinitarian theology; the centrality of
the Bible to theology, ethics, and piety; conversion as both gift and quest;
missional vocation; pragmatic flexibility beyond core convictions; and populism
in both theological authority and ecclesiastical politics.
These, I am
confident, can be reduced to good survey questions and—my point here—all must
be present, not just some, to be sure you have found evangelicals, rather than
this or that other kind of Christian.
Even more basically,
with church attendance statistics continuing to decline, pollsters are, in my
view, caving in (and have for thirty years) to count
evangelicals-in-good-standing among those who go to church only monthly.
But
mere monthly attendance, outside of special circumstances (like frontier
conditions in which a church is a twenty-mile horseback ride away, or Chinese
conditions in which precious pew seats are rationed), is not consistent with a
form of religion that emphasizes vitality. John Wesley, in short, would not
have approved of any Methodist attending church only monthly. Any good
evangelical pastor would feel the same.
Thus many things are
attributed to “evangelicals” that are true only of people who for one reason or
another happen to be caught in a too-wide net.
2. Having said all that
(!), I’m reminded of Michael Lindsay’s sociological work on American
evangelical elites (published eventually as Faith in the Halls of Power, OUP
2008) in which he reported that fully one-quarter of these successful people
identifying as evangelicals didn’t go to church. And I know a number of keen
evangelicals who struggle with churchgoing.
So do I contradict
myself? Then I contradict myself! But disillusionment with contemporary church
life is actually characteristic of the evangelical heritage—indeed,
characteristic of its “feeder” movements of Puritanism and Pietism.
What
Lindsay also reports is that these disaffected evangelicals nonetheless met
regularly with intimate and sustaining small groups of fellow believers. And
that is a page right out of the Pietist playbook, taken up by, yes, Wesley
himself. So then the pollster must ask a question about some other form of
sustaining religious fellowship. And Bob, as they say, is your uncle.
3. I think likewise,
then, of the massive decline of church attendance among Canadian Roman
Catholics. Given that form of religion’s understanding of the sacraments, you
just can’t be a “good Catholic” and not regularly attend mass. That’s just not
how that religion works, so to speak.
So, again, pollsters need to hear from
spokespeople in that religion and informed experts on that religion to draft
their questions to make sure they are finding who and what they want to find.
And the resistance of sociologists as a group even to read us historians, let
alone us theologians (as is persistently evident in sociological bibliographies
and footnotes), is sometimes a truly fatal flaw in the sociology of religion.
In sum, the right
questions will come from a proper grasp of the thing to be found. Part of what
is making the results vary in these disparate polls is more-or-less accurate
definitions.
Rick Hiemstra,
Evangelical Fellowship of Canada
I agree that some
measure is better than no measure, but John Stackhouse makes some very good
points. In my world, I look at evangelical Christians. As John points out in
this paper, Stackhouse Jr, John G. “Defining ‘Evangelical.’” Church & Faith
Trends 1, no. 1 (October 2007): 1–5. Evangelicals can be operationalized by
type or by movement.
On a recent Angus Reid
survey, we asked a religious affiliation question (movement) and the Christian
Evangelical Scale questions (type), and we asked respondents straight up
whether they considered themselves to be evangelical Christians. I realize that
the last question could be considered a movement question, but our religious
affiliation question found a different set of people than our
do-you-consider-yourself-to-be-an-evangelical-Christian question.
So, who are the real
evangelicals then? Is it those who claim to be evangelical but have nothing to
do with an evangelical church and who don’t evidence the beliefs and behaviours
of evangelicals? Is it those found by the Christian Evangelical Scale who don’t
consider themselves to be evangelical, nor do they affiliate with an
evangelical church. Is it those who affiliate with an evangelical church, but
don’t consider themselves to be evangelical, nor do they have the beliefs and
behaviours that the CES measures?
Let me present the CES
and affiliation data another way:
Here you can see that
evangelical affiliates are found across the CES scale.
My colleague Lindsay
Callaway and I are working on a survey of the ways evangelicals are
operationalized in North America. In the U.S. it is almost always doctrinal
scales, not behaviours.
As Sam Reimer points
out in his recent book Caught in the Current, and as Abby Day points out,
statements of belief are now often performative. This means they’re more about
belonging than about someone having considered a doctrinal proposition and
decided that this is what they believe. This means that the American
operationalizations are more likely to indicate a tribe than a conviction.
The EFC has used the
Christian Evangelical Scale, however, we’re working towards holding a
consultation on revising it because, as the first figure above shows, it
doesn’t do a great job of finding evangelicals. As Joel and Sam point out, it
does find something, and that’s better than nothing, but I think we can do
better.
As Lindsay and I are discovering, some of the American
operationalizations are baffling. For example, Barna has 4-point and 9-point
evangelical definitions (all doctrinal statements), and yet the 4-point
definition is not a subset of the 9-point definition. You can see for yourself
here: https://www.barna.com/glossary/
And yet, these are the
stats many rely on when they talk about evangelicals. In one definition we
found, Evangelicals are defined as ethnically white?! It’s no wonder people are
finding a link between evangelicalism and white-ness. They just started with
that assumption.
So, I agree with John
Stackhouse that there need to be measures of piety (behaviours) to substantiate
measures of doctrine. Doctrine alone just isn’t good enough to find evangelical
Christians. The CES asks just one behavioural question (frequency of religious
service attendance). That one question, however, is one more than other scales
use.
The CES has an evangelism question, but it’s not about whether people
evangelize, rather it asks if you think evangelism is important. If there ever
was a question that begged for a performative answer from evangelical
affiliates, this is it.
I personally find a lot
of media representations of evangelicals (that are usually bad) frustrating
when they’re relying on measures that may not actually be finding my community.
Evangelicals certainly have our sins to answer for, but often the ways we’re
portrayed are barely recognizable to us.
We should be more
skeptical about measures of religiosity than we are. I suspect most of us accept uncritically the
stats that confirm our own biases, but this is dangerous in a culture awash in
stats and fragmenting under the considerable social pressures of the moment.
Kevin Flatt,
Redeemer University
I have little expertise to add to the original
question. I'll just note that as the number of "nones" grows, I
become more and more interested in studies and surveys that can begin to
discern some of the different subsets of this category. Of course, nones have
in common that they do not affiliate with or identify with a
"religious" tradition.
But just as there are many different varieties
of religion and ways of being religious, some of which lead to meaningfully
different ways of life, there are differences among the non-religious too in
terms of the values, beliefs, and behaviours that shape their lives.
I'm aware of a few
books and studies that have started to ask these questions, but several of you
are experts on this, and I'd love to hear your pointers for the best places to
start to learn more about this question.
Peter
Schuurman, Redeemer University
Shipley and Young have
a recent book on sexuality, gender, and religion (Identities Under
Construction) and they have a nice section on how religion could learn from
gender studies about the wide variety of lived experience. In this sense,
sexual studies are "ahead" (I hate that word) of religious studies.
They say we need a new grammar for religion. You can see from the ol' SBNR term
that we do have our own acronyms in the world of religion (!). I have a book
coming out with Angela Bick this spring (we hope) introducing another term.
We'll see if it catches on.
Secondly, I have a
friend who is a pagan Catholic Jew. She said surveys never work for her. Does
she contradict herself? Then she contradicts herself! These are the days of
blessed incoherence. Polls just can't capture those layers (paradoxes?
jumbles?) The multiphrenic self said Gergen in The Saturated Self. Or you might
say, we are legion, to put a more ominous pall over the issue. Is it Advent
soon? I'm feeling more liturgical--almost Catholic. We'll see.
