10 October 2012

More Americans Have No Religion

A new survey by the Pew Forum On Religion and Public Life profiles the rising ranks of the religiously unaffiliated (eighty page pdf with the full report here).

Compared to five years ago, many more Americans are religiously unaffiliated.

About 6% of of American adults surveyed are atheists and agnostics, and another 14% believe in god in some sense but have "no religion." Overall about one in five Americans (19.6%) have no religion. The percentage of Americans identifying as having no religion has increased by about a third, from 15% to 20% in the past five years.  About 26% of Americans in 2012 were not Christians, up from about 19% in 2007.

The ranks of non-Christians who have a religious affiliation has grown at roughly the same rate as those who identify as having no religion. The percentage of people in the subset of people who have no religion who identify as atheists or agnostics has risen about 50% over the past five year, about twice as fast as the rate of growth for people who have no religion but do not describe themselves as atheists or agnostics.

At whose expense has the "market share" of the religiously unaffiliated grown?

Almost all of the growth in the percentage of people identifying as having no religion is matched by a decline in the percentage of people identifying as white and Christian (the survey classifies Mormons as Christians and not "other" as a recent 5280 magazine article did).

Other studies have indicated that mainline white Christians and white Catholics have seen their ranks shrink fastest; the ranks of white evangelical Protestants have also fallen, although a fair amount of the decline in particular white evangelical Protestant denominations is attributable to the rebranding of white evangelicals as non-denominational Christians.  The percentage of Americans who identify as Mormon or as Orthodox Christian, respectively, has remained roughly constant over the last five years, but part of this is due to the limited power of the survey to discern trends in populations that make up roughly 2% and roughly 1% of Americans, respectively.  Atheists alone (2.4% of American adults) and Agnostics alone (3.3% of American adults) were more common than either Mormons or Orthodox Christians.  Of course, there are important regional variations in these percentages.  Religion is one of the strongest markers of regional differences in American culture, particularly among whites.

Growth in the ranks of Hispanic Catholics has masked decline in the number of white Catholics in the United States until the past few years when Hispanic immigration to the United States has leveled off to roughly zero net migration.  Also, the number of children born to an average Hispanic woman in the United States has fallen considerably in the last few years.

The religiously unaffiliated are not generally seekers. Only 10% of people who are not self-described atheists or agnostics, but who describe themselves as having no religion, are looking for a religion. They are no more, and no less, accepting of supernatural ideas and New Age religious ideas (ghosts, psychics, astrology, reincarnation, spiritual energy in mountains or crystals, yoga as a spiritual exercise, deep connections with nature and the Earth) than members of the general public. A minority of people with no religion believe in God and pray.  These are people who were seekers and found an answer by discarding a religious affiliation, who are mostly satisfied with that choice: 74% of people who say that they currently have no religion now were raised with some religious affiliation.

The Roman Catholic Church places a great deal of emphasis on at least annual church attendance at Easter. And, it is right, viewed in light of its own self-interest. Only 8% of people who attend church at least yearly view themselves as having no religious affiliation, while 50% of those who seldom or never attend church see themselves as having a religious affiliation.

This result is not a function of methodology or small sample sizes.

This is not a statistical blip. The percentage claiming no religious affiliation, an atheist religious affiliation or an agnostic religious affiliation, respectively, has risen or held steady in every successive year in response to identical questions using the same survey method conducted by the same organization from 2007 to 2012 with sample sizes of not less than 9,443 in any given year. This year's sample included 17,010 respondents. The margin of error in the 19.6% estimate of the percentage of people who have "no religion" in 2012 is about +/- 0.6 percentage points, and the margin of error in the survey's 2012 estimate of the percentage of people who are atheist or agnostic is about 3.6 percentage points. The study from 2007 through 2012 is based on a total of 121,788 survey responses over this entire time span.

This study's results are also consistent with the trend lines observed by other studies measuring trend lines in religious affiliation over time using methods that are not identical to Pew's.

The General Social Surveys (GSS), conducted by the National Opinion Research Center at the University of Chicago) which has seen the percentage of people identifying as having no religion as having roughly doubled from about 9% in 1990 to about 18% in 2010 (when Pew's figure was 17.4% a result consistent with the GSS result at the one standard deviation level given the margins of error in the two large survey samples).

Similar trends have also been observed in the state level American Religious Identity Survey (ARIS), the most definitive survey based study of religious identification in the continental United States in existence (ARIS does not survey either Hawaii or Alaska for reasons of cost and logistics) and in surveys by Gallup and Barna, to name a couple of examples.

The religiously unaffiliated are demographically distinctive.

The change is a generational one that is working its way up the age pyramid of people with no religion get older, with younger cohorts much more prone to say that they have "no religion" than older ones. About 32% of adults under the age of thirty have no religion while 9% of those sixty-five years old or older have no religion. But, every cohort born after World War II has seen increasing percentages of cohort members who identify as having no religion in the past five years.

Thus, while younger American adults are less likely to be religious than older American adults, Americans born since World War II have also grown more likely to view themselves as having no religion rather than more religious as they have grown older.

Education and income had only slight impacts on the tendency to identify oneself as having no religion.  Women were less likely to identify as having no religion and much less likely to identify as atheist or agnostic.  Blacks and Hispanics were less likely than whites to identify as having no religion.

Consistent with prevailing stereotypes, those who identify as having no religion are most common in the West (26%), then the Northeast, then the Midwest and make up the smallest share of people in the South (15%).

[T]he United States remains a highly religious country – particularly by comparison with other advanced industrial democracies. . . . The number of Americans who currently say religion is very important in their lives (58%), for instance . . . is far higher than in Britain (17%), France (13%), Germany (21%) or Spain (22%).


Secularization is arriving late in the United States, but it appears to be finally arriving despite a long delay relative to Western Europe. Indeed, the very high percentage rates of growth in the percentage of people identifying as non-religious, after many decades during which there was almost no change in this percentage, suggests that this is a cultural trend that started around the time that I went to college and may have reached a tipping point.

The religiously unaffiliated, and non-Christians generally, favor the Democrats

Almost one in four registered voters who are Democrats are lean Democratic have no religion, the most common single religious affiliation in that political party, compared to 9% who are white evangelical Christians and 16% who are black Protestants. Collectively, non-Christians (including, for example, people who no religion, Jews, non-Christian Unitarian-Universalists, Buddhists, Hindus, Muslims, and pagans) make up one in three registered voters who are Democrats or lean Democratic. Catholics make up 21% of Democrats (and 22% of registered voters who are Republican or lean Republican), while Protestants make up 45% of Democrats.

In contrast, non-Christians make up about 18% of all Republicans, while 34% of Republican are white evangelical Christians. Non-Christians are as large a share of Democratic voters as White Evangelical Christians are a share of Republican voters.

Race is a key factor in the political affiliation of Roman Catholics. One-third of Democrats who are Catholic are not white; about 18% of Republicans who are Catholic are not white. Race is also a critical factor in determining the political affiliatioon of Protestant Christians. More than one in three Democrats who are Protestant Christians are black. Only about 1.7% of Republicans who are Protestant Christians are black.

The tendency of people who have "no religion" to favor Democrats over Republicans in the Presidential race has increased. In 2000 and 2004, people who no religion favored the Democratic Presidential ticket over the Republican one by a 2:1 margin. In 2008, people with no religion preferred Obama to McCain by more than a 3:1 margin: "Obama’s margin of victory among the religiously unaffiliated was 52 points; McCain’s margin of victory among white evangelical voters was 47 points."

About 48% of all registered voters are self-identified Democrats or leaning Democratic, 63% of people with no religion are, while 43% of all registered voters identify as Republican or leaning Republican, compared to 26% of registered voters who have no religion. People who have no religious identification are also, unsurprisingly, also more liberal than registered voters as a whole. But the difference is much more pronounced on social issues (e.g. abortion and gay rights) than economic ones (e.g., whether government is too big or too small).

Growth in the percentage of Americans who are not Christian and their increasing propensity to favor Democrats is a long term trend that favors the Democratic Party for the indefinite future.  If the Republican party does not become more welcoming to ethnic minorities and/or non-Christians, it is on a path towards permanent minority status in American politics.  The perception that the public has of the Republican party as bigoted, even if this is not fully accurate, is deadly to its political prospects.

No comments: