The big caveat for polls like this comes right up front: “Unaffiliated” is not a synonym for “nonbeliever.” Agnostics and atheists are actually a small minority of the “unaffiliated,” many of whom say that religion remains important to them even if they no longer identify with any one denomination.

That’s the good news for the faithful. The bad news: Pretty much everything else, especially if you’re a mainline Protestant or Catholic.

In 2007, there were 227 million adults in the United States, and a little more than 78% of them – or roughly 178 million – identified as Christians. Between 2007 and 2014, the overall size of the U.S. adult population grew by about 18 million people, to nearly 245 million. But the share of adults who identify as Christians fell to just under 71%, or approximately 173 million Americans, a net decline of about 5 million…

Of the major subgroups within American Christianity, mainline Protestantism – a tradition that includes the United Methodist Church, the American Baptist Churches USA, the Evangelical Lutheran Church in America, the Presbyterian Church (U.S.A.) and the Episcopal Church, among others – appears to have experienced the greatest drop in absolute numbers. In 2007, there were an estimated 41 million mainline Protestant adults in the United States. As of 2014, there are roughly 36 million, a decline of 5 million – although, taking into account the surveys’ combined margins of error, the number of mainline Protestants may have fallen by as few as 3 million or as many as 7.3 million between 2007 and 2014…

Like mainline Protestants, Catholics appear to be declining both as a percentage of the population and in absolute numbers. The new survey indicates there are about 51 million Catholic adults in the U.S. today, roughly 3 million fewer than in 2007.

Where are all the lapsed Christians going? Some are going to the unaffiliated column…


…whereas others are passing away, replaced in the population by younger Americans who are much less likely to identify with a particular religion than their elders. Sharp differences between twenty- and thirtysomethings on the one hand and older people on the other are a staple of modern polls, but this is dramatic even by the usual standards:


Among those 70 or older, 85 percent are Christian versus just 11 percent who are unaffiliated. Among those 25 or younger, it’s … 56/36. Will the next generation dip below 50 percent Christian? Each and every Christian denomination has lost support over time, and the decline across the various age groups isn’t even. Starting from the left, you see a drop-off of seven or eight points every 20 years among all Christians and then suddenly a drop of double digits starting for those born in the early 80s and holding firm to the present day. What explains that?

While it’s true that not all unaffiliateds are nonbelievers, it’s also true that unaffiliateds include more nonbelievers than ever before. Compare the trend since 2007 among unaffiliateds who say religion is important to the trend among self-identified atheists and agnostics:


In 2007, unaffiliateds were 16.7 percent of the population, the fourth-largest “denomination” in America behind evangelicals, Catholics, and mainline Protestants. Seven years later unaffiliateds stand at 22.8 percent of the population, now the second biggest “denomination” behind only evangelicals at 25.4 percent. They’ve surpassed Catholics, who dropped from 23.9 percent to 20.8 percent. Atheists and agnostics are now 7.1 percent of the population — bigger than all non-Christian faiths in the U.S. combined, even though non-Christians have grown in numbers in the last seven years.

As noted in the last paragraph, the news is especially bad for Catholics. This brought me up short.


That’s … dramatic. I remember reading articles last year about whether, given the new pope’s apparent popularity, there’d be a “Pope Francis effect” among lapsed American Catholics that might rekindle their interest in the Church. Evidently the answer is no; on the contrary, if there’s a “Pope Francis effect,” it may be in the other direction. Some Catholic writers have wondered if Francis’s feints towards liberalizing Church policy might not end up driving away more conservative members of the faith, even potentially to the point of schism. Is that what’s happening here, or is this a byproduct of something else? Lingering fallout from the Church’s child-molestation scandal, maybe? Or are Americans, especially younger Americans, responding to the idea that they need to choose between gay rights and Christianity by choosing the former? (Unsurprisingly, at 41 percent, gays themselves are far more likely to identify as unaffiliated than straights are.)

Pew’s survey is long, but if you can spare five minutes, I’d suggest scrolling through Chapter 3 to see how religious affiliation lines up with other demographic data. One key fact: The unaffilateds, especially atheists and agonistics, are disproportionately male whereas Christians are (not quite as) disproportionately female. Another: If you’re “unaffiliated,” you’re as likely to have never been married as you are to be married. Among Christians, you’re more than twice as likely to have a spouse than to have never been married. And one more: Atheists and agnostics are more likely to have bacherlor’s degrees or postgrad degrees than Christians of any denomination are, although both groups are dwarfed in the latter category by members of some of America’s non-Christian faiths. Nine percent of Christians have postgrad degrees; 16 percent of atheists and agnostics do; fully 31 percent of Jews do; and fully 48 percent(!) of Hindus do. There may be some margin-of-error effects in those last two numbers given the small subsamples, but if you already hold the stereotype that Hindu and Jewish parents take education more seriously than Americans at large do, there’s nothing here that would convince you otherwise.