In my November 26 Winnipeg Free Press column, I wrote about how the recent census shows it’s not only the mainline Protestant denominations that are declining, but also Pentecostals and Baptists—two evangelical denominations tracked by Statistics Canada. What’s happening? And does the decrease extend to other denominations? I asked some scholars from religion who are part of the evangelical tradition to share some thoughts. Their full responses are below.
Rick Hiemstra, Director of Research, The Evangelical Fellowship of Canada
I think the answer to your question lies in how the census question has evolved over time and in how the religion question is coded.
For a chart showing how the census questions have change from 1951 to 2001 see the chart on page 5 of “Evangelicals and the Canadian Census.”
Extending this chart, I’ve included the 2011 National Household Survey question and the 2021 census question below. The differences between these two questions have been highlighted in yellow.
2011 National Household survey which replaced the 2011 long-form census
22. What is this person’s religion?
Indicate a specific denomination or religion even if this person is not currently a practicing member of that group.
For example, Roman Catholic, United Church, Anglican, Baptist, Lutheran, Muslim, Presbyterian, Pentecostal, Jewish, Buddhist, Hindu, Sikh, Greek Orthodox, etc.
(Specify one denomination or religion only, or no religion.)
(https://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvInstrumentList&Id=75586)
The 2021 long-form census
30. What is this person’s religion?
Indicate a specific denomination or religion even if this person is not currently a practicing member of that group.
For example, Roman Catholic, United Church, Anglican, Muslim, Baptist, Hindu, Pentecostal, Lutheran, Presbyterian, Sikh, Buddhist, Jewish, Greek Orthodox, etc.
(Specify one denomination or religion only, or no religion.)
(https://www.statcan.gc.ca/en/statistical-programs/instrument/3901_Q2_V6)
Note the declining position of evangelical denominations in the question.
If you look at tables 2 and 3 in the “Evangelicals and the Canadian Census” paper linked above, you can see that the direction of census numbers between 1991 and 2001 were sometimes at odds either in the direction or magnitude of denominational statistics.
For example, the 2001 census reported a decadal decline for Pentecostal of 15.3% whereas the Pentecostal denominations reported a membership growth of 23.3% over the same period. The census and membership numbers are not the same thing.
As I note in the paper, Pentecostal did not appear in the 2001 census question, and I think this accounted for at least some of the discrepancy. So questions do make a difference in data.
Baptist and Pentecostal are the only evangelical denominations that remain in the census question. Their share fell between 2011 and 2021, and while there may have been some decline, there has been a somewhat corresponding rise in “other Christian.”
I’ve had some correspondence with Jarod Dobson from StatCan about how they’re coding the religion variable on the 2021 census. He provided this link:
https://www23.statcan.gc.ca/imdb/p3VD.pl?Function=getVD&TVD=1311612&CVD=1311613&CPV=2&CST=01042021&CLV=1&MLV=4
You’ll note that “Other Christian” isn’t broken down any further. In the past, census coders were provided with a codebook that essentially told coders what buckets to put different religious code values (answers) into. I asked for the equivalent document for today and, while I was presented with a spreadsheet with more detail, I was cautioned not to treat it as a code book.
So, we don’t really know how “other Christian” was coded, although I would really like to know. I suspect that there are many in this category who would be considered evangelical by type or movement definitions (see Stackhouse “Defining ‘Evangelical’”;) but who would be reluctant to identify either with the more comprehensive “evangelical” label or one the denominations associated with the movement.
I think this is more than just a case of their being “shy evangelicals.” It’s likely an attempt to put distance between themselves and the common characterizations of evangelicals in the media and culture.
Until StatsCan releases a codebook for the 2021 census, we won’t know the composition of “other Christian.” My guess is that they’re mostly Evangelicals.
Kevin Flatt, Redeemer University
Rick's points about the limitations of census data regarding evangelical affiliation are important ones, and I agree with his analysis. I'll add three additional thoughts.
First, the "Other Christian" category may contain evangelical denominations or traditions among new Canadians, such as African Independent Churches, that do not easily map on to more well-known (to Canadians) evangelical groups such as Baptists and Pentecostals.
Such "new to us" groups tend to get under-counted and under-noticed, but with immigration trends, are likely becoming a larger part of the evangelical mix in Canada.
Second, the Christian, Not Otherwise Stated (n.o.s.) category (as distinct from the "Other Christian" category) probably has some mix of evangelicals and "cultural" Christians who retain some sense of Christian identification but have lost touch with any particular church community or tradition. It thus combines the two Christian subgroups with the highest and lowest levels, respectively, of measurable religiosity.
There are probably more cultural Christians than evangelicals in this census category. But given the large sociological differences, it would be nice to tease this out further. My understanding is that others who are more well informed than I am (e.g. Sarah Wilkins-Laflamme) have educated guesses about the relative proportions of evangelicals and cultural Christians in this category based on non-census data.
Sam and his colleagues have done some great work using denominational statistics that suggests that some evangelical groups stalled or began to decline numerically after 2000 or so. This represents a change from the situation in the last half of the 20th century, which denominational statistics and other detailed research suggest were a period of fairly robust growth for evangelical churches, matching or exceeding population growth.
As Rick notes, the census has never really done a good job of capturing these trends, but if that numerical transition has really happened "on the ground," one would expect it to show up in broad strokes in the census eventually. If "the great evangelical stall-out" is really happening, my armchair hypothesis is that it has three main drivers:
Number 1: The drying-up of defections from mainline Protestantism as a source of growth. I think the switch of theologically and morally conservative Protestants out of liberalizing mainline Protestant churches was an under-appreciated contributor to evangelical growth in the 1960s-1990s period.
But most evangelically-inclined mainline Protestants who wanted to switch over to evangelical churches had already done so by the 1990s, and of course as the mainline Protestant churches have dwindled away (United Church affiliation fell by about 40% from the 2011 to 2021 census) there are fewer potential defectors.
Number two, Canadian and broader Western popular culture and institutions are less compatible with evangelical faith today than they were 40 years ago. The cognitive dissonance for evangelical children and youth in public schools, for example, is now much greater due to both the waning of a residual generic Protestant ethos (remember that Ontario public schools could still begin the day with Christian devotions until 1988 or so) and the emergence of a new moral/ideological ethos, especially around sexuality.
Social media use has exploded since about 2012; this tends to have a secularizing impact on balance in terms of young people's reference points. Of course, many evangelicals retain their convictions or even have them strengthened as they age, and pass them on to the next generation, but the headwinds are stronger than they used to be.
And number three, changes in the character of evangelical churches. In explaining church growth or decline, we need to look not only at the external environment, which is largely shared by all churches in the same geographical/cultural setting, but also the varying characteristics of the churches themselves.
Bob Burkinshaw argued in Pilgrims in Lotus Land that a combination of theological/moral conservatism (for lack of a better word) and innovative techniques underpinned a lot of evangelical growth in 20th-century British Columbia. If that is correct, and generalizable to the rest of the country—and I think it largely is—then a change in those characteristics could lead to numerical reversals.
I don't have good data on this, but my anecdotal impression is that the larger evangelical groups in Canada today are by-and-large somewhat less theologically and morally conservative than they were in, say, 1980. If that hunch is correct, it could be a factor too. But this last point will be controversial.
Sorry to pull the old preacher's trick of sneaking in three sub-points under my final point.
John Stackhouse, Crandall University
As others have said, we do need to bore into “Other Christian” and “Christian Not Otherwise Stated” to see who’s there.
1. What’s the difference between those two categories? Ought there to be two?
2. Boring into them further will help test the thesis that denominationalism is waning (a thesis going back a full generation) such that Christians, and especially evangelicals, don’t care about denominations and instead search for a local church according to an individual/familial list of concerns (rather than a “tribal” loyalty to denomination or a doctrinal/sacramental/ethical loyalty to a particular tradition): good preaching (as they themselves judge “good”), good Sunday School, good youth group, etc.
I wouldn’t condemn this attitude as necessarily consumeristic, but it is utterly individualistic.
3. Boring into them further will also test the thesis that evangelicals have been declining for at least three decades—in every metric, from regular churchgoing to giving to affiliation—and somehow not all the relevant census and poll data show that decline.
If that thesis is true, then what has happened to prompt evangelical decline? If that thesis isn’t true (per Reimer and Wilkinson, who—if memory serves—suggest that evangelicals have not declined but have in fact retained their share of the market, thus actually growing in a growing Canada), then where are (all) the evangelicals in the national census?
Sam Reimer, Crandall University
A couple of comments:
First, when Reimer and Wilkinson wrote their 2015 book on evangelical congregations, our sense was that we were just passing over the peak and starting the decline, although some evangelical denominations were declining in the 1990s.
I think that evangelicals are more clearly declining now, but it will be slowed because of immigrants who are evangelical. As Kevin noted, these are harder to count. However, evangelicals are likely losing ground because the Canadian born are becoming nones.
Second, the census, because of the wording of their question—and maybe just a sense from respondents that they should answer in a way consistent with their family tradition—reports higher percentages of religious affiliates than polls do. More people are willing to say they have no religious affiliation in the polls. So, the recent Maru/EFC poll showed that Christians were a minority in Canada, whereas the census still reports a slight majority.
