PRRI Poll Numbers on Mormons

I love religion polls and surveys. I think they tell us a lot about what has been a rather transformative period in American religion the past 20 years. While I’m still eagerly awaiting the next ARIS survey, PRRI came out with a new religious study to wet our appetite.

There actually were a lot of surprises in this poll.

One thing I found quite interesting was the high rate they got for Mormon identification. Unlike Pew or older ARIS studies, they found us at 1.9% of the population. While they suggested this wasn’t much of a change, referring to the older Pew study[1], their figure is actually substantially higher than either Pew or ARIS. ARIS is increasingly out of date but had us at 1.4%. The last Pew study in 2014 had us at 1.6%. Unfortunately since PRRI is a new study we can’t compare like to like. That makes it hard to know how much that figure is an artifact of how they conducted the survey or if it indicates significant growth. It’s so out of line with both Pew and ARIS but using a fairly large sample that it’s hard to imagine it doesn’t indicate actual growth.

While I probably shouldn’t have been surprised, the age breakdown study did manage to undermine my expectations. I’ve always seen Mormons as primarily a young religion. Relative to most established American religions we are. Compared to other religions though we’re not as young as I thought.
In particular non-Christian religions swamp our youth. Unsurprisingly the Nones also are far more young. I wasn’t surprised by that though since each rising generation since the early 90’s has been less religiously affiliated than the one before. To the point that more than a third of the current generation are unaffiliated.

One nice thing about the PRRI study is how it broke out Protestants and Catholics by race (although oddly not black Catholics – of which there are quite a few in places like Louisiana). Hispanics unsurprisingly are much younger than white religions. Primarily due to immigrants having more children I suspect along with immigrants being more religious.

While Mormons either remained constant or more likely increased, the same isn’t true for our Evangelical friends. While there were hints of Evangelicals finally succumbing to the same attrition that hit mainline Protestants decades earlier, PRRI suggests this is in full effect now. Their graph is a bit misleading in that it combines PRRI studies with Pew figures. However it definitely shows a decline over time in Evangelical self-identification.

Given that Evangelicals were, for a fair time, keeping mostly up with Mormons this is a pretty interesting change. It also makes our stability and/or growth even more fascinating since it goes so much against trend.

Many of the other findings were more in line with expectations. It confirmed that Christians, particularly white Christians, are a minority among Democrats. While the last year there’s been a lot of attention to more left wing Christianity, Democrats continue to be less religious each year. Likewise despite the drop in Evangelicals, they continue to makeup a fairly stable percent of Republicans over the last decade. (Around 35%)

There were a few other interesting trends. Driven presumably by hispanic immigration, the South has become much more Catholic. I lived in Louisiana for a while so I was used to a strong southern Catholicism. I suspect in other southern states this is a bigger change though. Today 29% of Catholics live in the south compared to 13% in 1972.

Unsurprisingly the Nones continue to grow as a fraction of Americas. The rate of growth since the great recession of 2008 though has been surprisingly sustained matching only the trend in the 90’s. Seeing the graph though it becomes clear just how significant this trend is.[2]

[1] They wrote, “currently, 1.9% of the public identifies as Mormon, a number identical to findings from a 2011 study of Mormons in the U.S.” However the study they refer to was the 2012 Pew one. However that study I believe merely recontacted Mormons from an earlier 2007 study. I suspect they were going by the line “Mormons make up slightly less than 2% of the US public.” But that’s misleading both because of the date (it’s 2007 not 2011) but more significantly because the figure was 1.6%.

[2] We discussed the Nones last year around this time. The continuing growth strongly suggests that the rise of the Nones is ceasing to be merely a nominal change in self-identification of those with low religiosity to an actual substantial change. There’s no indication of an abatement in this change.

12 comments for “PRRI Poll Numbers on Mormons

  1. adano
    September 6, 2017 at 10:27 am

    Fascinating stuff. Just a minor quibble, but regarding this: “In particular non-Christian religions swamp our youth.”

    Saying that Muslims (etc) tend to be young is not the same as saying that young people tend to be Muslim. You appear to be saying the second thing here. The figure supports the first statement but not the second.

  2. JT
    September 6, 2017 at 11:08 am

    Very interesting. It would be interesting to see what role life expectancy plays into some of these numbers. If American Mormons live, on average, almost a decade longer than other Americans, it would not be surprising to see Mormons skewing a little older.

  3. JT
    September 6, 2017 at 11:09 am

    (There are obviously many other variables at play, but certainly a factor to take into consideration.)

  4. fbisti
    September 6, 2017 at 12:34 pm

    Interesting stuff, but should be interpreted with a bit of through a glass darkly vagueness.

    I read through the full report. Based on my 30+ years of experience in survey research, I am impressed by the robust methodology. However (in my overly pedantic world), at least one of the authors’ own statements about some aspects of the findings isn’t fully accurate:
    The study sampled the youngest adult currently residing in the household, so it is a sample of adults and cannot be used to conclude: “currently, 1.9% of the *public* identifies as Mormon.” The full questionnaire was not included in the report, but such a conclusion could only be made if there was a question determining what religious affiliation is claimed by each member of the household, adults and minors. So that conclusion should have read “of the adult public.”
    Other comments…
    -The key question used the term “Mormon” and therefore would include any fundamentalist or other historical spinoff religion that uses that moniker. I don’t know how this question has differed in terminology in other such studies.
    -Differences in sampling and instrument (nature and size of the sampling frame and weighting; differences in wording) between studies can easily account for differences in results (1.4% vs 1.9%) that are technically statistically significant due to the large sample sizes. Though actual differences in such statistics can also be actual and the result of actual changes over time. So except for differences/errors introduced by the ever-increasing non-response bias (a growing proportion of sampled respondents willing to take the survey), only a study that is identical in design can (with high accuracy) identify changes resulting only from time.
    Still, fascinating data. Thanks Clark

  5. Clark Goble
    September 6, 2017 at 1:41 pm

    (Sorry for the title getting screwed up – it’s a bug in WordPress)

    Fbisti, yeah, the summary and discussion of the poll was a bit underwhelming. As I mentioned they compare Mormon numbers with Pew but misrpresent the Pew numbers. Although as someone mentioned to me elsewhere the 1.9% and 1.6% are within the error bars of 0.4%. (Although barely) I definitely agree though that this could also be due to other artifacts. Which is why I’m a bit cautious pushing that 1.9% too much. Comparing different surveys with different methodologies is fraught with problems. So I’m not pinning anything on that 1.9%. I want to see the latest ARIS which I find to be the most reliable. (I have a lot of issues with Pew’s data as they get some pretty questionable results for some questions – there’s also questions of being biased by oversampling in Utah County with Pew.)

    Typically in these self-identification studies Mormon doesn’t always mean LDS. Looking at both Pew and ARIS stuff there are reasons to think a few non-LDS ended up in the survey although it’s ambiguous. (Primarily answers to questions on doctrine in Pew)

    Adano, yes, I just meant that Muslims and Buddhists skew younger than Mormons.

  6. September 6, 2017 at 5:52 pm

    Looking at the Pew numbers from 2007, I think I see how someone could accidentally come up with 1.9%.

    If you expand the Mormon section (, it shows “1.6%” for Church of Jesus Christ of Latter-day Saints, and “< 0.3%" for Other Mormon. If someone just took the numbers and added them together, they'd end up with 1.9%.

  7. Brian
    September 6, 2017 at 9:18 pm

    “To the point that more than a third of the current generation are unaffiliated.”

    I don’t understand how you came to this conclusion. From the data you provide, it appears that while more than a third of unaffiliated are of the current generation (the inverse of your statement) only about 25% of the current generation is unaffiliated (I’m guessing on where the values are on the far right end of the line plot).

  8. Clark Goble
    September 6, 2017 at 11:43 pm

    Brian. Sorry I chose the phrase “current generation” poorly. I meant 18-25 year olds of which 34% are unaffiliated. I should have said current new generation.

    Eric, that actually makes sense. Although that would imply the numbers here don’t break out types of Mormon. It’s weird as typically the header is he sum of the subcategories such as for historically black protestant which is then broken down into baptist, methodist, etc. But for Mormon the header is 1.6% and then two subcategories of 1.6% and < 0.3%. I assume you're right that they just added them together but I think < 0.3% just means so small that the total was still 1.6%. I'd never noticed that before with the Pew data. I'm not quite sure what fully to make of it. In any case, I think they read it wrong if I understand it correctly.

  9. James Olsen
    September 8, 2017 at 5:55 am

    Sorry if I missed someone else posting this, but it’s interesting that the Church just came out with a Newsroom article on Nones among our youth. They make it a society wide trend (which it is), but they do feature, quote from, and refrain from disparaging an LDS None. I can’t say much for the article’s analysis, it’s constant pairing of data or quotes by Nones with contrary quotes from church authorities or scholars, or the simplicity of its prescription (social media!); rather, the significance is the article itself and its tone (note the title):

  10. Clark Goble
    September 8, 2017 at 10:44 am

    Yeah that was a weird article when it came out. I remember people discussing it a little. It seems like it’s a discussion of the Nones and the high rate among 18-25 year olds of religious disaffiliation. (Is Gen Z really a thing now?) When did the Newsroom start doing a category called “commentary” which seem like op eds? In this case the main focus seems to be explaining the push to use social media as a strategy here. That appears to be a formal strategy of the church although I’m skeptical it’ll have a huge effect. My sense is that what’s driving people to the Nones isn’t the lack of church in social media.

  11. Wally
    September 8, 2017 at 11:13 am

    Or to “whet” your appetite perhaps?

  12. September 23, 2017 at 7:48 pm

    Interesting. I am not good at interpreting surveys. What I see by the graph is that the numbers with high 18 to 29 numbers are religions that have, for the most part, very strict upbringing and don’t usually leave the home of their parents until marriage. It might be unwise, therefore to state any other belief than your parents. Also, in my area of the country, Muslims have high birth rate. I noticed that the white catholic has the lowest number for the 18 yo 29 year old. Probably due to much more freedom that is tolerated by their parents.

    What I most notice is that those that had low 18 to 29, there are higher numbers for the child rearing years, which seems to suggest the wanting to raise the children how they were raised. If that is correct than I would suggest that we need to make good memories for our children so that they want to raise their children how they themselves were raised.

    Birth rates of the various groups need to be considered as well as the strictness or closeness of the family life, divorce rates, etc.. I think the best way to retain our youth is family closeness and parents always setting the example and trying not to be hypocrites (that is a total turnoff for our youth).

    What worries me most is the low numbers for the old and closer to death. How can that be?

    I admit I didn’t read the study, just looked at the graph.

Comments are closed.