There Are Less of Us Than You Think

There Are Less of Us Than You Think September 17, 2012

According to Grey Matter Research, Americans think our country is far more religiously diverse than it actually is. In a survey of 747 adults the research and consulting firm found that most underestimated the size of Christianity and over-estimated the size of atheists, Muslims, and other religious minorities.

The Mount Soledad Cross.

“The typical American adult pegs religious affiliation in the U.S. as follows:  24% Catholic, 20% Protestant, 19% unaffiliated, 9% Jewish, 9% atheist or agnostic, 7% Muslim, 7% Mormon, and 5% from all other religious groups. In reality, according to the 2008 U.S. Religious Landscape Survey from the Pew Forum on Religion and Public Life, Americans are right on target with the proportion of Catholics and the “all other” category, but way off target on the rest of the landscape. The typical American badly underestimates how many Protestants there are in the country, and way overestimates the presence of religious minorities such as Mormon, Muslim, and atheist/agnostic.”

In fact, if you check the Pew Forum data from 2008, you’ll see that Muslims in America only comprise 0.6% of the population. In contrast “Unitarians and other liberal faiths” comprise 0.7% and “New Age” faiths (ie Pagans) comprise around 0.4%. There are more Buddhists in the United States than there are Muslims. Likewise, respondents guessed large for  atheists, agnostics, and the unaffiliated. Speaking with the Religion News Service, Grey Matter president Ron Sellers noted that media attention is a likely reason for the over-inflated guesses of non-Christian or non-religious populations.

Sellers also mentioned that with Mitt Romney running for president as a Mormon and the current emphasis on Islamic-American relations, “smaller faith groups also may be getting disproportionate media coverage.”

Likewise, younger Americans, who tend to have more friends who are atheists or religiously unaffiliated, guesses in favor of their own experience. Also unsurprising is the news that adherents of a particular tradition tend to guess high on their own numbers.

Not going to become the 3rd largest religious group any time soon.

“One thing that is clear from this research is that people tend to overestimate the proportion of their own faith group.  Among people who identify with the Catholic Church, the average estimate is that 39% of the country is Catholic.  Not only is this estimate much higher than it is among non-Catholics, it is far higher than the reality of 24%. Similarly, among people who identify with a Protestant faith perspective, the average estimate is that 27% of the population is Protestant.  While this is far higher than the numbers among non-Protestants, it is still almost half the correct figure. Among people who identify as atheists or agnostics, the average estimate is that 16% of the American population is atheist or agnostic.  As with Catholics, not only is this estimate far higher than among any other group, but it is much higher than the reality.  Finally, among people who express no particular faith identification, the average perception is that 35% of Americans believe in God but have no actual religious preference.  Again, this is nearly double the average American’s perception, and far higher than the real figure in the U.S.”

So what’s the take-home message of this data? Sellers says that “this skewed perception of religion in America may benefit smaller faith groups and be detrimental to Protestants.” In other words we are over-estimating the influence of religious minorities, and under-estimating the influence of Protestant Christians. This may seem like a good thing, a hastening of the demographic shifts many of us existing in religious minorities have been waiting for, but it could also feed into the fears of certain Christians who are increasingly uneasy with our mere existence. Then again, maybe feeling like a religious minority could teach a valuable lesson to those willing to encounter it.

Being a minority tests the temper of a faith, its resilience and fiber […] Being a member of a minority entails the ability to bend and to negotiate. This, in turn, demands a deep understanding of the majority and local conditions, deeper than the majority may have about the minority; respect for them whenever possible; diplomacy; patience; and the building of relationships, infinitesimal gesture after infinitesimal gesture.”

People are over-estimating religious minorities, and those with no religion at all, but maybe this misconception will instill a willingness to embrace secularism once more, to re-enforce those church-state separations so that the “others” don’t exert undue influence. In which case, beware Christians, Pagans are growing at an alarming rate! Quick! Everyone join Americans United for The Separation of Church and State, it’s your only hope!

Browse Our Archives

What Are Your Thoughts?leave a comment

60 responses to “There Are Less of Us Than You Think”

  1. Having been a pagan for what feels like a long time and having so much of my spare time involved in some form of pagandom I always forget how weird and small we are. I say things like “Oh I can’t make it tomorrow, my coven meets on Fridays” and then realize “Oh… that’s not a thing people say is it?”

  2. The chart doesn’t break out atheist/agnostic. They seem to be subsumed under unaffiliated.

    At the middle of the twentieth century, Jews were about three percent of the American population. This survey indicates Jews have dropped to 1.7 percent. I believe this is a drop in both absolute and relative numbers.

  3. I’m part of a non-denominational Pagan church, and I have the same problem. It always surprises me how many people assume Christian when they hear “church”. It shouldn’t surprise me, but whenever it happens, it does.

  4. I realize Jason is writing with tongue firmly in cheek, but imho it’s never good for anyone to take action based on faulty information. Even armed with correct information it’s still salutary for everyone to take those inward steps toward living with people of other religions on the same land and in the world. The violent panic currently sweeping the Moslem world over a cheesy film trailer is, if nothing else, an example of the failure of such accommodation.
    I was secular long before I became Pagan, and still am secular in my politics. I would like to see my religious beliefs on our relationship with nature reflected in law, but only because they are defensible in secular terms, not because they are articles of my faith.
    Many of our Christian friends could benefit themselves and society by taking one baby step toward accommodating diversity: Stop saying “religion” when you mean “Chrstianity.” It would clear the air enormously and settle some disputes without further debate.

  5. I wonder how they get this info. I do not remember having been quizzed about my religion…so how would they know?

  6. This reminds me of when I was in high school and first found out that black people make up about 11% of the total population of the US. Everyone in my (mostly white suburban) class seemed shocked, and had thought the actual number was more like 20-25%. I live in Texas and also see a similar thing with Hispanics, and white people feeling threatened that Hispanics are taking over Texas. (I’ve been hearing about the Hispanics taking over for many years, and just got data saying they expect TX to be majority Hispanic by 2014 or 2015, so it hasn’t happened yet, but there are a lot of people who seem to think it already happened a long time ago.)

    I wonder if we’re seeing a similar tendency here with this overestimation of religious minorities, like I have seen with racial minorities. I wonder if it’s just so alarming to them to see a few non-white or non-Christian people that they way overestimate the “threat”. Maybe someone who knows more about psychology than me can give more insight. I also have worked at state parks, and people also way overestimate how many poisonous snakes/spiders/scorpions/etc. there are out there. They see one, and suddenly they think the place is crawling with them. I wonder if this is a similar thing.

  7. Surveys don’t need to query everyone in a given population in order to get a representative sample. That’s why there’s a margin of error on any poll; it’s an estimation and the actual results for any category could be higher or lower, within that margin of error.

    No survey of this nature is perfect, but given a large enough population and a good enough sample, they do a very good job of estimating the actual.

  8. I don’t know about that. I’ve seen a lot of research studies based on representative samples that turned out to be wrong. I guess I’m just not a “poll” person.

  9. “Love is letting go of fear.” That’s from Jerry Jampolsky. He wrote a great book with that title, too!

  10. we are few and far between when compared to the Big Three. Isn’t it a good thing they overestimate our numbers?I do hope this drives them to Americans United , that i belong to btw.Any member of any minority Religion/Faith should be a member of AU.I do have to agree this increases the paranoia of the RR, Nar types tho, those people scare me.Altho Christians are the overwhelming majority about 75 %, these folks see thier majority and percieved control in American slipping away .Seems most religious groups , us included tend to overestimate their own nunbers , just as they overestimate our numbers. I do believe we are headed for a post Christian America , just hope we survive to enjoy it . As the radical fringe of Christianity feels more treatened they will be more likely to strike out at thier percieved enemies .We are at the top of their list . On this point i beleive we are in for stormy weather .Like i said earlier hope we can weather this upcoming storm, i believe we can and be stronger for it .One of my fav sayings is ” that which dosn’t kill you, makes you stronger” Kilm

  11. True. I have to admit, though, that ~750 people sounds like an awfully small sample for the sweeping conclusions they’ve got. Not my field, though, so it could well be representative– it just seems rather tiny.

  12. I used to be manager of election polling for one of the largest polling companies in the U.S. I have a bit of expertise in the area. 🙂

  13. According to the survey methodology, Pew surveyed 35,556 people by phone for the survey, and have an 0.6% margin of error. That’s a pretty respectable number.

  14. Things become very interesting very fast when one starts to look at specific religious beliefs that are very non-Christian, such as reincarnation and astrology. Belief in reincarnation is consistently at 25% or more throughout the Western world.

    The thing is, most religions are not like bowling teams. You don’t join a team, get the shirt, and that’s your team, and you can’t be on any of the other teams. Only monotheisms work like that, and, therefore, counting up “adherents” according to the monotheistic model of religions-as-bowling-teams totally skews the results.

  15. But suffering leads to pain, and as a great poet once wrote:

    “And though we love to numb the pain

    We come to learn that it’s in vain

    Pain is our mother

    She makes us recognize each other.”

  16. The head of the Pew Forum on Religion and Society is a far-right evangelical political activist named Luis Lugo. Lugo is deeply committed to using public opinion research and other applied social sciences for spreading the gospel.

  17. Please point to where, exactly, in the methodology used to produce the poll, you see bias. Is it in their sampling? Their questionnaire? Do you have anything concrete you can cite, or just vague accusations of bias?

  18. I am not making a “vague accusation of bias”. I am stating a very specific fact about the man who is the Director of Pew Forum on Religion and Society.

    Luis Lugo’s last job before coming to Pew was working for the right-wing evangelical think tank Center for Public Justice (link). He was also an activist in the group Christians in Political Science, whose national meeting he addressed in 1999, when he stated that “Christian
    political scientists are, before anything else, at the service of God’s
    people, providing intellectual leadership to help them as they seek
    faithfully to carry out their temporal calling as citizens,” and also that “we must make the Christian tradition our primary intellectual community.” For more on Lugo and Pew, as well on Gallup and Templeton, see: A Form of Ministry (Push Polling for Jesus).

  19. And I am asking for specific instances where you are claiming that this particular survey is biased, either through the sample, the questionnaire, or some other basis. Thusfar, you’ve given some biographical information on the head of the Pew Institute, but nothing that indicates there’s anything wrong with the data in this particular survey.

  20. I’m somewhat confused about how this survey of what seven hundred or so people *think* really changes the same Pew numbers we’ve been looking at for all this time?

    (I do think the Pew survey kind of busts smaller groups up till all are below the ‘margin of error,’ but I still don’t know what place in the world I’m more likely to meet a JW than another Pagan. But even by our most-expansive estimates of our own numbers, if you poll seven hundred people, you’re only likely to catch a few of us, assuming your demographics are truly random. (And being Pagan doesn’t necessarily mean you’d answer the questions in question so differently than anyone else,) So I’m not sure what this means, here. 🙂 )

  21. I think that the issue is that there may be what Nate Silver at fivethirtyeight calls a “house effect”. No matter how good a single survey looks on paper, it’s not a good idea to put too much faith in it without multiple replications to compare the data.

  22. Damn lies and statistics.

    I would not read too much into anything. Statistics are not facts. Surveys are unreliable.

    Compare that to the autumnal feel in the woods this morning, the golden sun, the cool air and I know what I am, where I am and I don’t need labels , statistics or any other validation.

  23. That is probably because of this:
    “Church is an English word for a Christian religious institution or building”

    Why not use ‘temple’? It is a less ‘owned’ term. Alternatively the Englisc term ‘hearg’ could be used, if you want to avoid something Latin derived.

  24. And what would be ‘a large enough population’?

    The USA has a population of 311,591,917 (2011, Google Public Data), how many people would need to be polled to get a representative sample? And from where would you draw it?

    I’d say at least ten percent would be required to minimise bias, from each State/major settlement (large urban areas can skew data by not being representative of the larger region).

    So, that would be a minimum of 31,159,191 from all over the USA. Not just 747 people that (I can imagine) have been polled from, perhaps, a couple large, cosmopolitan, urban areas.

    I hear that the Twin Cities of Minneapolis and St. Paul, in Minnesota are something of a Pagan hub.

    Take a sample from there and I imagine the religious demography will be a lot different to Houston, Texas.

  25. That is 1/8763 of the population, or 0.0114% (roughly)

    You honestly think that is going to give an accurate representation of the whole? I’m not convinced.

  26. I always forget how weird and small we are.

    Charles, I have that problem politically. Among my friends – and the local Pagan community – I’m on the conservative end of the spectrum. When I get among mainstream folks and make what I think are pretty middle-of-the-road remarks, I get exaggerated reactions to remind me that I’m not, in fact, conservative. ::laugh::

  27. I’m afraid I don’t have the time to give you a complete primer in statistics and polling right now, but if you take the time to look up some basic information on the subject, you’ll find that samples far less than 10% are able to produce reliable and representative results of a much larger population.

  28. And ‘hearg’ would be “harrow” in modern English. That’s the term I use for the altar I have in my back yard, and for the space I mark off around it on holidays.

  29. I suspect a bigger problem than sample bias in this instance is the difficulty in defining religious identity. There are vast differences in people’s religious identity based on what they were raised in, what they self-identify with, what they actively practice, and what they believe. As just one example, there are millions of Catholics who are still nominally Catholic but who dropped out, and other Catholics who consider themselves devout but hold core beliefs which are way out of bounds with official doctrine etc. There are polytheists who don’t want to be called “pagan” for one reason or another etc.

  30. Let me try.
    If Zoroastrians represented one person in 1,000, then a polling sample of 500 might miss them altogether. A sample of 20,000 would have a much better chance of including some, and thus be more accurate.
    Note that the increase in accuracy has *only* to do with the size of the sample and *nothing* to do with the size of the total population.

  31. Besides being a CR, i am part of an ADF grove .We unlike most have a perminant location and sacred space we call our Santuary. Kilm

  32. Not true at all; margin of error can be moved either by increasing the sample size *or* decreasing the population. So if you had a sample of 1000 out of a population of 350 million, you’d have a MoE of 3.1. But with that same sample of 1000 out of a population of, say, 10,000, you get a MoE of 2.9. Just the way the math works.

    But this is academic to the point you obliquely made; the survey wasn’t intended to be granular enough to capture the number of Zoroastrians, or Wiccans, or Asatruar, or Neo-Pagans-Who-Don’t-Want-to-Call-Themselves-Pagans. To Jason’s original point, they’re so marginal, and so small, that they simply aren’t the focus of the study (it would be possible, but very expensive, to conduct such a “study of the margins”– if someone wants to fund such a thing, let me know and I’ll be happy to set up a methodology and a survey instrument).

    People in general are much more interested in relative numbers of Catholics and Protestants than they are of relative numbers of Alexandrian Wiccans vs. Gardnerian Wiccans.

  33. I tried to point this out to a group of pagans several years ago, but was roundly booed and promptly ground into dust for being, “too mainstream & buying into propaganda”. Explain as I might that this in no way lessened the legitimacy of their (or my own) beliefs, they simply refused to see they were outnumbered.
    My whole point was that, while we are free to believe/worship/practice as we please, we have to remember there are a large number of people out there who fear & revile us out of close mindedness or simple ignorance and because of that, we face hurdles socially, politically and religiously.
    Sometimes, as a minority, you have to tread lightly….not because you SHOULD have to but because strident revolution frightens people. The best we should do is set an example for religious tolerance while not condemning people for their own beliefs, as ignorant or backwards we may see them

  34. And to those of us Witches who have ridden motorcycles our whole adult lives (and part of our younger years as well) the statement going to “church” has a whole ‘nother meaning of its own! Just sayin’ … BB!

  35. As I am sure you know (from reading your bonifides below) statistical surveys, and polls of ANY kind, can be worded, phrased, and lastly interpretted to reflect the point the reader (or publisher) wishes to prove. I am by no means an expert, but I did take a year of statistical studies many years ago in college, and basically, the course taught us exactly that! How the question is worded, sometimes specifically FOR the type of folks you are polling, often guides the responses you are given, and multiple choice answers are even easier to skew to your point with. I have learned, yes, they are often right … but by producing “facts” from a well respected pollster about any given subject often will affect – DIRECTLY – the same poll later taken from the same sample group to more positively reflect the original poll. EX: Todays poll states “Romney is the landslide winner of the next election”. All else being equal, or similar, 2 weeks later, the same poll often reflects a larger percentage of the same group, after reading the results of the first poll, who support Romney! People DESIRE and WANT to be on a winning side, so when asked again, they side with the previous winner! This is not random conjecture but proven facts that the public is never informed of. Knowing this, I take most polls ‘with a grain of salt’, unless I can read the actual questions, and research the folks who were sampled. Just sayin’ …

  36. Informative? Absolutely. However, I know, and have witnessed those I know personally who are in the ‘Craft, claiming ‘agnostic’, when they think there are those asking, or can hear who might demean them for being so! Our society is much more open, and accepting of us, NOW than they were when I started in the ‘Craft, but believe me, there are still those who we terrify by our very existence! Mainly from lack of knowledge … but they still often recoil when I tell them I am a Witch!

  37. GOPagan, are you incapable of acknowledging the simple fact that it is important to know as much as possible about the source of any given set of data? Data doesn’t just fall out of the sky. It takes a serious commitment of resources to do any kind of polling on a significant scale. People do not commit those kinds of resources for no reason, nor do they do it out of the kindness of their hearts.

    Medical researchers who receive money from pharmaceutical companies are required to reveal those ties. I think the same basic principle should, and quite obviously so, apply to those engaged in public opinion research. If one takes a look at the three biggest names in that game in the U.S. (Gallup, Pew, and Templeton) one find two common factors: right wing politics and fundamentalist Christian beliefs.

  38. If my math is right, .03% of 311,591,917 is less than 93,478, and that seems awfully low. That’s way below the 300,000 estimates current back in the early-mid ’90s, and even further below Fritz Jung’s guesstimate of about a million Pagans in the United States, which he made in the early 2000s. It’s hard to imagine a group that size supporting publishers and book stores, which Pagans have done for decades.
    The poll could under-report Pagan numbers for reasons that have nothing to do with deliberate bias. As several people have noted, a lot of Pagans will not necessarily identify themselves as such, especially to a stranger. If a stranger called you and asked your religion, even if they identified themselves as a pollster, would you really tell them you were Pagan? Could you be sure who they really were? I would not even consider doing so. I’m way too deep in the closet to take such a chance. And I very much doubt I’m alone.
    As some others have noted, the poll’s findings of the size of Pagan numbers are much less than the margin of error, probably by a factor of 10, if the poll is typical. So, the real upshot of the discussion is that we still don’t know how many Pagans there are. That we are a tiny minority compared to the Christians is something we have known for decades, and that many of us experience daily.

  39. Not 0.03%. It’s <0.3% (doubled, because that is the number for "Wiccans" and also the number for "Pagans"). <0.6% total of 311,591,917 is less than 1,869,551. How much less is anyone's guess, but that would seem to corroborate the one million estimate.

  40. Here are some links to other estimates of the Pagan population. They vary widely, with most falling between 100,000 and rather less than 1 million, suggesting that this particular Pew survey is an outlier when compared to other surveys. In the Wikipedia article, scroll down for demographics. And of course, the Wikipedia article should be taken with the usual massive grain of salt one uses for Wikipedia, but the demographics section, at least, appears to be well sourced. Also scroll down to get the estimate of 450,000.
    These other estimates and data sets still tell us that we don’t really know how many Pagans there are. But they suggest that this latest Pew survey is indeed an outlier, an unusually low estimate. On the other hand, it is still perfectly true that we are vastly outnumbered by the Christians, a fact that I, certainly, cannot forget.

  41. Well, ok. Then we have this Pew survey corroborating Fritz’s one million, more or less. To put it another way, there are indeed as many of us as we think. And even so we are severely outnumbered, which we also already know.

  42. As Faoladh pointed out, I read the Pew survey quite wrongly, adding an extra “0” before the “3”, just due to poor eyesight. It’s actually estimating Pagan and Wiccan numbers at more than 1.8 million, which actually puts it at the high end of estimates of Pagan numbers. So my comments in the above post are way off. But the links are still useful, giving the context of a variety of other estimates of the number of Pagans.

  43. Less than 1.8 million. I was working with the numbers as given. However, the survey gives 0.4% for all “New Age” religions, which includes both “Wiccan” and “Pagan” (calling both of those “less than 0.3%”). 0.4% would be 1,246,368 or so, for all so-called “New Age” religions. That actually roughly approximates to ~800,000 for all Wiccans and Pagans together (assuming approximately 1/3 for each of the “New Age” subcategories listed, so a very rough estimate indeed).

    The point being that it is easily within an order of magnitude of the one million estimate, making that a really good one if the survey data are accurate.

  44. I’d tell them my religious leanings without hesitation. Hel, I have a hobby of engaging street evangelists when they ask the question ‘Do you have a moment to talk about God/Jesus?” My personal best is almost two hours.

    I see it as a public service. Whilst those bible bashers are harping on at me, they aren’t bugging everyone else and, sometimes it can even draw a crowd.

  45. If ‘Pagans’ don’t want to be outnumbered, then there is only really one choice available. Getting out there and ‘spreading the word’.

    Of course, that means agreeing on what word to spread…

  46. I think what matters are results. You’re making insinuations that the study is somehow flawed or biased, but you give no specifics. As I pointed out yesterday, Pew was tied for first as the most accurate pollster in the 2008 election. That’s an empirical measure; their methodology produces sound results, as compared against a predicted actual outcome.

    I think actual results are infinitely more significant than the political affiliation of the owner of the company. Or don’t you believe that even those with conservative political leanings can produce sound, impartial results?

  47. Yeah, could be just that. “Normal” people are white Christians. Anyone else is abnormal, so they stick out.

Close Ad