The Best Charts About the ‘Rise of the Nones’ You’ll See All Year

Morning Edition on NPR is running a weeklong series on “Losing Our Religion” and they ran a nice story today about the “Nones” (atheists, agnostics, people who believe in “something” but don’t belong to an organized religion, etc) and it’s worth a listen. It’s nothing you haven’t heard before, but yay for more press about the topic!

Better than the piece, though, are the graphics by Matt Stiles appearing alongside the piece on the NPR website. They showcase the growth of the Nones (in general), the growth of young Nones, and the growth of both male and female Nones.

Somewhere, a Wall Street banker is salivating over those lines.

The trend, [Harvard professor Robert] Putnam says, is borne out of rebellion of sorts.

“It begins to jump at around 1990,” he says. “These were the kids who were coming of age in the America of the culture wars, in the America in which religion publicly became associated with a particular brand of politics, and so I think the single most important reason for the rise of the unknowns is that combination of the younger people moving to the left on social issues and the most visible religious leaders moving to the right on that same issue.”

And the Internet. Don’t forget the Internet.

Keep in mind a recent Gallup poll showed slower growth in our numbers than we’ve seen in recent memory, but one data point is hardly enough to suggest a trend.

About Hemant Mehta

Hemant Mehta is the editor of Friendly Atheist, appears on the Atheist Voice channel on YouTube, and co-hosts the uniquely-named Friendly Atheist Podcast. You can read much more about him here.

  • Sven2547

    I see it as a natural consequence of the Information Age. The more you know about the universe, the less you need to fill in unknowns with “God”.

  • Epinephrine

    Graphs should start at 0.

  • abb3w

    Missing would be a graphic showing the respective generational cohorts over time — though Pew has one. Those (as I recall) are near-flat — but with each generational decade tending more irreligious than the last. Most of the change shown on the above charts results from this.

  • ggsillars

    It would be remarkable if there weren’t a pause in the growth of the nones. Things rarely proceed in a straight line without any short-term ups and downs, but the overall trend is definitely up for us.

  • Bad_homonym


  • A3Kr0n

    Wake me up when its all over, or just let me sleep.

  • seriously

    Yeah, I don’t see how starting at 0 would help with this at all (or how the graphs starting higher than 0 somehow might skew any interpretation)

  • seriously

    Is there anything more than anecdotal evidence that the internet has anything at all to do with the rise in atheism? I know that there’s been a lot of evidence that the internet actually just tends to serve as an eco-chamber, so I’d be curious to see if Hemant’s insistance really is anything more than just an idiosyncrasy of his (especially if he’s going to act like he’s correcting a Harvard sociologist who specializes in religion)

  • Sven2547

    The internet provides people with access to communities that they may never have known existed. This is particularly true in communities where a high percentage of its members are “closeted” in some fashion.
    Also, Hemant isn’t contradicting Putnam, he’s just adding another point.

  • Sven2547

    Depends on the graph.

    For example, look at a graph of Apple’s stock prices last Friday. It fluctuated between $519 and $524. Having that axis start at zero would render it pretty useless.

  • Reginald Selkirk

    In the first graph, notice the noise level; there is a certain amount of boundcing around from year to year, although the long term trend is clear enough. This noise level should give you pause about the short time span of the last two graphs.

  • Joshua Zelinsky

    One issue here is how many of these people are people who a few years ago would simply have said they were Christian just as a default and are now comfortable self-identifying as “none”.

    It is also worth keeping mind that when one separates the demographics more, essentially splitting nones into ,atheist, agnostic, or none/none-of-the-above, then most nones look very different from the self-identified atheists and agnostics. One major example is with knowledge about religion. Atheists and agnostics know more about religion as a whole than any other group (with only Jews and Mormons coming close) . In contrast people who answer as “none in particular” or something similar when given the atheist and agnostic option are one of the least knowledgeable groups (See the above Pew link). More studies need to give the options of atheist, agnostic and none-in-particular as separate options.

  • seriously

    Oh sure, I don’t deny that. But it could very well be that the internet can make atheists more supported in their atheism. It could also be that people who are secure in their religion just ignore information that contradicts it (as we have a lot of psychological evidence to suggest) and that people who are questioning their religion actively seek out information to confirm their suspicions (as we have evidence to suggest). So it seems that if people just use the internet to reaffirm their biases (as we have evidence to suggest) then it just seems what’s probably happening is that people who question are going to be atheists, people who don’t won’t, and the internet doesn’t really make a shred of difference.

    So I’ve heard Hemant say a lot that the internet matters, but I’d like some evidence.

  • Joshua Zelinsky

    Because it makes them easier to read and easier to see trends and the rough size of trends at a glance. When you start a percentage graph at 5% like the second graph does, it visually looks like much more of a jump than it is if you start at zero. This is bad presentation of data.

  • RobMcCune

    At a glance a graph starting higher than 0 can make a trend look like a greater change than actually occurred.

  • Dangerous Talk

    The Media Is Doing It Wrong: Rise of the ‘Nones’ –

  • Blacksheep

    I’ve noticed for years that many people that I know who used the terms “Christian” or “Jewish” had no real connection or belief, it was just the customary thing to say. Maybe this is a more honest age – and maybe it will lead to a greater percentage of Christians who strive to follow Christ.

  • Bad_homonym

    Not really. We use scada displays at my place of work and not one trend I can think of is graphed from zero. You get better resolution on a graph with a narrower range as long as all the data is in range

  • ortcutt

    Atheist and agnostic aren’t religious affiliations. I would answer None if they asked me my religious affiliation, but I am an atheist. Affiliation and belief are orthogonal.

  • Anna

    The trend, [Harvard professor Robert] Putnam says, is borne out of rebellion of sorts. “It begins to jump at around 1990,” he says. “These were the kids who were coming of age in the America of the culture wars, in the America in which religion publicly became associated with a particular brand of politics, and so I think the single most important reason for the rise of the unknowns is that combination of the younger people moving to the left on social issues and the most visible religious leaders moving to the right on that same issue.”

    I’m not totally convinced by that. It could be one reason, but there are several other things that could be contributing to the rise of the Nones. As someone who was coming of age in 1990 (well, I was in middle school), I think it could be that my generation is the first to really grow up in a society with a great deal of religious pluralism. Previous generations didn’t go to school and meet children who were from Hindu, Muslim, or Buddhist families. Mine did. Perhaps a lot of people met those children and began to question the claim that Christianity was necessarily right and superior. The rate of interfaith marriage has also soared. Children of interfaith marriages are less likely to follow any particular religion as adults. And you have the fact that religion is seen as more malleable. It’s not as taboo to change religions as it was for previous generations. It’s also more acceptable not to have a label at all.

  • RobMcCune

    It’s a matter of who your audience is and what their needs are. A general audience who might simply glance at a graph could be mislead if they were to go by appearance only. So they should start at zero where practical, or tables should be used instead.

  • Bad_homonym

    I guess we agree to disagree. The graph should also range to 100% by that reasoning. The resolution ought to be maximized so as not to ‘dampen’ the data. Higher resolution means higher accuracy in my books so eliminating the portions of the graph with no data makes sense!

  • C Peterson

    I happen to think these surveys overstate the growth of the nones, just as they understate the actual number of atheists. As you suggest, there are almost certainly many people who have always been nones, and always been atheists, and our changing social norms now makes it more acceptable to identify as such (more for none than atheist, so far). So the rise in nones really decomposes into two parts, which the data don’t separate: those who have recently become nones (the actual growth), and those who are willing to admit what they’ve been for a long time.

    Fortunately, the societal effect is largely the same no matter where the growth is coming from.

  • Joshua Zelinsky

    Sure, but if you had to pick one of atheist, agnostic or nothing in particular, which would you pick?

  • ortcutt

    I would pick “nothing in particular”. I have no problem identifying myself as an atheist, but that’s not a religious affiliation.

  • Amakudari

    [T]he single most important reason for the rise of the unknowns is that combination of the younger people moving to the left on social issues and the most visible religious leaders moving to the right on that same issue.

    This is true, but there’s also the question of why people moved “to the left” (clearer: became more acceptant and tolerant of differences). It’s not enough to say that younger people are more liberal, because this country has had Great Awakenings and counter-Enlightenment movements. To me, the biggest reason is increasing access to information. If you grow up in a close-knit community with a few channels and near-mandatory attendance at one of a handful of churches, like many of America’s seniors, chances are that you’ve never heard a kind word about atheists, queers, etc., and would hardly know one if you met one. Now, you have a generation that can read about people from all walks of life, that attends college away from home, that by immigration and urbanization are much more likely to have very different peers in school.

    If it were simply about intergenerational differences, I think people would get off that bus at “Christians but not that kind of Christian.” It’s not enough to be on the wrong side of the argument, as slavery and women’s suffrage have shown. The “nones” have to be driven by something else, and I think it’s the the increasing difficulty of holding a religious viewpoint, even milquetoast liberal Christianity, without hearing challenges to it.

    (A knock-on effect is that it seems liberal more than conservative denominations are losing members to the “nones,” so the well-funded, more popular demagogues tend to be all the crazier.)

  • C Peterson

    Given the intent of the survey, “atheism” is a logical alternative to religious affiliations. In the same way that atheism is treated as a religion for legal purposes, it can be viewed as one for the purpose of understanding people’s views. By answering “none” when you are in fact an atheist, you would be needlessly skewing the results. “Agnostic” is the category that doesn’t fit, since it doesn’t separate theists and atheists, and isn’t even mutually exclusive of any of the listed religions. You can’t effectively be a Catholic and an atheist, but you can certainly be a Catholic and an agnostic.

  • ortcutt

    Someone can certainly be a Jewish atheist. In fact, many are. Answering “atheist” to your religious affiliation is like saying “5 foot 7″ when someone asks your weight.

  • C Peterson

    Sure, because “Jewish” is ambiguous in a survey asking for religious preference. It should be made clear that only those who consider their religion to be Jewish should identify as such. Catholic isn’t ambiguous. Lutheran isn’t ambiguous. And atheist isn’t ambiguous, and is logically included in a list of religious affiliations.

  • ortcutt

    You can’t be a Catholic or Lutheran atheist? That’s probably surprising to many Catholic and Lutheran atheists.

  • C Peterson

    I expect that those who are pretending to be Catholic or Lutheran for social reasons when they are actually atheists will not be confused by the survey. They will either tell the truth, or they will lie. That’s a factor with any survey that asks sensitive questions. But there’s no logical inconsistency. They can’t be both atheists and of some religion.

  • meekinheritance

    Just think how big the rise would be if we didn’t practice birth control!

  • abb3w

    There’s a long-term trend vs. cohort that predates public access to the Internet. The dawn of the trend looks likely to predate the Internet, but the GSS doesn’t go back past 1972, and there irrelgion vs. cohort correlation was already well developed.

    The rise across cohorts of the Nones appears more pronounced among those with higher internet access in the home (GSS 2006-2010 variable INTRHOME), so there’s at least SOME evidence. However, the number of Strongly religious appears to be lower among those who don’t have home internet access, which is not what a simple connection of the rise would lead to expecting.

    There’s also a recent trend shown up in the Pew study data in the last couple years, showing a shift within most (post-WWII) cohorts over time to irreligiousity. That might well be from the influence of growing WWW access, or possibly the post-9/11 “New Atheist” writers.

  • abb3w

    The decomposition can be approximated by lumping in the “not particularly” religious in with the Nones. The trends by cohort becomes much more gradual, but still persists; and there’s a significant-looking bump against time for the last few years.

  • abb3w

    The jump looks to be the confluence of two things: the end of the heyday of the religious right, and the logistic curve rise in the Nones versus Cohort growing in signal amplitude to well beyond the noise level (where new arrivals were more than 10% likely to be irreligious). The latter, however, is the bigger factor.

  • Anna

    Or we could end up like Scandinavia, where a lot of the people who self-identity as Christians are actually atheists, LOL.

  • baal

    ?! Birth control is a lot like a TIVO. Its main impact is to shift the timing of pregnancies. It doesn’t really do all that great a job of population control (unless forced or compelled by the State cf China). What controls population is more a question of educational opportunities and jobs availability for women as well as what support the country provides to support children.

    Also, I doubt atheist are in agreement with the religious that the best way to grow the religion is to make as many kids as possible.

    I otherwise find the charts heartening.

  • Patterrssonn

    “and maybe it will lead to a greater percentage of Christians who strive to follow Christ.”

    Now if they can only agree on what that actually means.

  • pagansister

    “Now if they can only agree on what that actually means”. Maybe like following the Pied Piper?