The Rise of the Nones

Due to a combination of laziness and everything blowing up at once at work, I wasn’t able to post about this earlier, so by now you’ve no doubt read all about the Pew forum’s poll report about the recent increase in the number of Americans with no religious affiliation, or “nones”.

You’ve seen this graph:

Growth of the Religiously Unaffiliated, from the Pew Research Center.

You’ve heard that for the first time, Protestants make up less than half of the US population. That there are more atheists, agnostics, and none-of-the-aboves than before, and that organized religions are losing members. That the nones lean toward the political left. That according to Bill Donohue we’re all a bunch of selfish, self-absorbed brats. So let me mention something else.

According to Pew’s chart, 2.4% of Americans adults call themselves atheists, 3.3% call themselves agnostics, and 13.9% are nothing in particular, neither atheist/agnostic, nor members of an organized religion, for a total of 19.6%. But the report also notes that

many of the country’s 46 million unaffiliated adults are religious or spiritual in some way. Two-thirds of them say they believe in God (68%).

which means that 32% of the unaffiliated don’t claim to believe in God. If my math is correct, the atheists comprise 12.2% of the unaffiliated, and agnostics are a further 16.8%; so atheists and agnostics together make up 29.1% of the unaffiliated. So even if we assume that none of the atheists or agnostics believe in God (probably a good assumption for atheists, perhaps not so good for agnostics), that still leaves us with 32% – 29.1% = 2.9% of “Nothing in particular”s who don’t believe in God, which presumably makes them atheists, even if they don’t call themselves that.

The last time I wrote about the American Religious Identification Survey, I noted a similar phenomenon. Except that then, the discrepancy between “people who don’t believe in God” and “people who call themselves atheists” was significantly larger.

Two obvious possibilities present themselves. It’s possible that ARIS and Pew are measuring different things, and that the numbers they report are only loosely related. But it’s also possible that people who don’t believe in God are more comfortable calling themselves atheists or agnostics; that the stigma attached to those words is disappearing. The difference between the ARIS and Pew surveys is the sort of thing I’d expect to see if that were the case.

More on ARIS’s “Nones”

I’ve taken a second look at the ARIS results, particularly the “Nones” which have attracted so much attention. Here’s a graph of various Nones through the three surveys, as a percentage of the US population at the time:

ARIS's "Nones"
(click to see a little larger.) This graph is drawn from data in Tables 1 and 3, and 4 of ARIS 2008.

Here, the topmost line represents what ARIS 2008 calls the “Nones”: atheists, agnostics, “anti-clerical theists”, nonreligious, and so forth. As expected, it’s the largest group.

It’s also the group that has grown the most since 1990, when NSRI 1990 (the survey to which ARIS 2001 was a followup) was conducted. However, its growth has slowed down substantially since 2001.

The red line at the bottom shows self-described agnostics, and the purple line just below that, self-described atheists. NSRI 1990 lumped atheists and agnostics together, so the leftmost data point actually shows the sum of both. This also explains the dip in 2001. The sum of self-described atheists and agnostics is 0.7% in 1990, 0.9% in 2001, and 1.6% in 2008, so the trend is actually increasing, and has apparently picked up steam (there were 29% more atheists+agnostics in 2001 than in 1990, and 78% more in 2008 than in 2001).

It’s interesting to contrast this to the slowing growth in Nones overall. I note that the “new atheist” bestsellers all came out between the last two surveys: The End of Faith in 2004, The God Delusion, Breaking the Spell, and Letter to a Christian Nation in 2006, and god is not Great in 2007.

It would be nice to say that the Four Horsemen led to the growth in nonbelief, but there’s not enough data in here to jump to such a conclusion. At most, I think we can say that Harris, Dawkins, Dennett, and Hitchens haven’t killed atheism; atheists aren’t going back into the closet.

Lastly, the three triangles with no lines, in 2008, represent the answers in Table 4:

  • Atheist A: there is no such thing as a god.
  • Agnostic A: there is no way to know whether there are any gods.
  • Agnostic B: don’t know whether there are any gods.

The obvious thing to notice is that there are far more of them than either self-described atheists, agnostics, or both together. Doubtless this includes some Buddhists, Taoists, Scientologists, etc., but there are too few of those to account for these numbers. I suppose that some come from the “generically non-religious” pool, while others identify themselves as members of some mainstream religious group, but don’t accept all of the group’s tenets. “Cultural Catholics”, if you will.

All in all, it’s a bit disappointing that with all the hoopla about “the new atheism” and uppity unbelievers, out-of-the-closet atheists still haven’t cracked the 1% mark.

Another point that’s been talked about is the growth in “generic” Christianity (“Christian unspecified” and “Non-denominational Christian” in Table 4 of ARIS 2008). Some of this is due, I’m sure, to smaller churches shutting down and their parishioners migrating to generic megachurches. But the growth of the Nones leads me to suspect that it’s also due to a growing disenchantment with organized religion.

It’s not uncommon to hear a sentiment along the lines of “I’m a Christian, but religion is bullshit”. People who feel the divine, but feel that organized religions are corrupt, or self-serving, or otherwise undeserving of their membership. These people are not swelling the ranks of rationality; but at the same time, they’re diminshing church rolls, and helping to reduce the power base that the Pat Robertsons and Ted Haggards of the world can mobilize and use to wield social and political (to say nothing of monetary) power. So they’re a net gain.

In other words, what this data seems to show is that a) Americans are rejecting religion, and b) Americans are rejecting superstition. These are two separate issues, but in both cases, we’re moving in the right direction.

The New ARIS Is Out!

ARIS 2008,
the American Religious Identification Survey, has just been released,
and the atheosphere is as giddy as a bunch of schoolgirls upon the
release of Harry Potter and the Adjective Noun.

The most significant finding, IMHO, is that the “Nones”
(atheist/agnostic/no religious preference) are up since 2001,
although nowhere near as dramatically as betwen 1990 and 2001. But
still the fastest-growing segment of the population.

One new feature is that, unlike the 1990 and 2001 ARISes, the 2008
survey asked “what do you believe?”-type questions, rather than just
“what do you call yourself?”

Table 3
in the
full report
(p. 5) lists atheists as comprising 0.7% of the US population, and
agnostics at 0.9%. However,
table 4
(p. 8) lists people’s answers to
the question of whether there is a god: “There is no such thing”
(atheism) comes in at 2.3%, “There is no way to know” (proper
agnosticism) at 4.3%, and “I’m not sure” (common agnosticism) at 5.7%.
(IMHO it might be interesting to see how many people haven’t really
thought about it. Maybe next time.)

In other words, there are a bunch of people who are atheists and
agnostics, but don’t call themselves that. Presumably they either just
call themselves “None of the above”, “No religion”, or “Spiritual, but
not religious”; or else they’re members of religious groups that allow
that kind of latitude, such as Buddhism or Taoism.

Table 7
(p. 11) shows that religion is stronger among women than men. In most
religions listed, there are more women than men. “None” is an
exception to this rule (as are “Eastern Religions”, “Muslim”, and “New
Religious Movements/Other”), but the skew is most pronounced: 60% of
Nones are male vs. 40% female.

Update, 16:14: USA Today has a nice
interactive graph
of the survey results.

I’m guessing that the sharp drop in “Other religions” in Wyoming, and of “Don’t know/Refuse” in Delaware represent statistical anomalies (i.e., they happened to get a bad batch of poll respondents), rather than real demographic trends.