The latest Pew Forum results (from the Pew Research Center's second U.S. Religious Landscape Study, a follow-up to its first comprehensive study of religion in America, conducted in 2007) shows a growing number of irreligious Americans, with good reason to think this is a part of a trend that will continue, as the youngest cohort is moving away from religion even more rapidly than the whole population is. Over a third of the youngest slice of adults now declare no religious affiliation. This is heartening for those of us who are convinced that America — and the world — will be better off when we stop expecting a non-existent "Power" to step in and help us solve problems that only we (human beings), and we alone, can solve. The increase from 16.1 percent to 22.8 percent of "unaffiliated" adults — about 56 million Americans — appears to be both dramatic and in our favor, but there are reasons to be cautious in interpreting such numbers regarding religion.
There has long been a serious stigma regarding atheists or other irreligious identification in this nation. How much of this has been caused by malicious, intentional misrepresentation of a causal connection between religiosity and ethical character, and how much is based instead on simple ignorance of what it means to be irreligious is hard to know. Whatever the causes, the stigma is real, persistent though declining, and inaccurate. Its persistence is most easily seen in repeated surveys asking voters whether they would vote for someone of their own party who agreed with them on most issues but who was (black or a woman or an atheist, etc.) Opinions of atheists do seem to have improved by this criterion, but atheists still score quite low on such measures. For more details on such studies, see NPR's "Would Voters Entrust the White House to an Atheist?".
The stigma against atheism or the automatic presumption that being religious is good has long proved to be a problem for social science researchers. An example from over twenty years ago showed this clearly in one Ohio county. Researchers there asked people whether they had attended religious services in the most recent period being measured, and 36 percent reported that they had. But researchers wisely attempted to verify that data by visiting every church, temple, and mosque in the county and counting actual attendees — and only 20 percent of the population had been present. (See "Church Attendance" for much more.) Whether respondents were lying to themselves or to the researchers, there was a clear bias in favor of appearing to be more religious than they actually were.
A possibly unanswerable question is whether recent changes in religious identification among American adults is more affected by actual changes or just by a greater social acceptability of the irreligiosity already there. Either way, a pro-religiosity bias seems to be fading, especially among the youngest adults in the Pew 2014 study. That may be more important, and more positive, than even the changing reported identification percentages. It also seems quite likely that some of the anti-irreligion bias remains and therefore it may be that there are even more "unaffiliated" than the Pew results report. According to the latest Pew results, it is startling to learn that there are now more American adults who are "unaffiliated" than there are Catholics. Or mainstream Protestants. The already unreasonable and dated practice of politicians to pander to religious voters may be on its way out for demographic and political reasons. A new group, AtheistVoter,) is lobbying hard to help make that happen.
While only a little over 7 percent of adult Americans labeled themselves as "atheists" or "agnostics" in the Pew study, that approaches double the 4 percent who chose such labels in 2007. And the proportion of the unaffiliated who used those labels rose from 24.8 percent in 2007 to 31.1 percent in 2014.
Many religious advocates, especially among the most conservative perspective, decry the declining identification with religion as responsible for an alleged moral decline in America. But such declarations are at odds with the facts, as crime rates, expanding civil rights, and other indications of ethical improvement show. Discussions and interpretations abound. See, for example, here and here.
It seems a safe bet that religious affiliation will continue to fade in the U.S., that labels such as atheist, humanist, and agnostic will become more prevalent and more socially acceptable, and that we will continue to be a better, more ethical, responsible, freer nation as a result.