Ed Stetzer, writing at CNN in an opinion piece entitled "No, American Christianity Is Not Dead" on May 16, 2015, in response to a Pew Report showing a surge in people who identify as religiously unaffiliated and a 10% decline in 10 years in the percentage of people who identify as Christian in the United States, makes some valid points, but also provides some shocking (to me) new statistics of his own.
He excludes from his analysis Roman Catholics and historically black Christian denominations, focusing instead on trends in Protestant and non-denominational Christianity.
One of his core points is that the people who no longer identify as Christians were mostly not drawn from the population that regularly attend church. This percentage was 23.2% in 1972 (according to the General Social Survey) and was 19.8% in 2014. Thus, while there was a 3.6 percentage point drop (a 15.5% decline in "market share") over the 42 years that coincide with most of my life, the decline has not been nearly as dramatic as the decline in identification as Christian.
The long term trend, however, has been very different for Evangelicals, and other non-Roman Catholic Christians.
The percentage of people who identify as Evangelical Christians and attended church regularly in 1972 was 7.9% of the people surveyed (46% of the 17.1% of people who identified as Evangelical Christians at the time), and in 2014 it is 12.5% (55% of the 22.7% of people who identified as Evangelical Christians in 2014). Thus, over 42 years the percentage of people identifying as Evangelical Christians, the percentage of people regularly attending Evangelical Christian churches, and the percentage of Evangelical Christians who attend church regularly have all increased.
It isn't all roses for Evangelical Christianity. It peaked in 1994, declined steadily through 2004, bumped up a bit in 2006, and has declined again since then.
The percentage of people who identified as Mainline Christians and attended church regularly in 1972 was 8.6% of the people surveyed (29% of the 28.0% of people who identified as Mainline Christians at the time), and in 2014 it is 3.6% (30% of the 12.2% of people who identified as Mainline Christians in 2014).
This picture is decidedly less rosy. Over 42 years the percentage of people identifying as Mainline Christians has dropped by 15.8 percentage points (a 42% decline in "market share"), the percentage of people regularly attending Mainline Christian Churches has dropped by 5 percentage points (a 42% decline in market share), and the rate of church attendance for people who identify as Mainline Christians has remained constant. The decline in Mainline Christianity has been steady with the exception of a small bump in 1990.
Thus, in 1972, a narrow majority of regularly church attending white Protestants were Mainline Christians (as I was from 1970 to 1992 with the exception of a year in New Zealand where I didn't attend church while I was an exchange student). Regularly church attending Mainline and Evangelical Christians combined were 16.5% of the population.
In 2014, regularly church attending white Protestants combined are still 16.1% of the population, barely changed over 42 years (the decline in the overall church attendance rate probably comes mostly from white Roman Catholics whose trends tend to mirror those of Mainline Protestants, but have been masked by rising numbers of Hispanic and other immigrant Roman Catholic populations in the U.S.).
But, in 2014, 77.6% of regularly church attending white Protestants (more than three out of four and on track to be four out of five soon) are Evangelical Christians.
There is a discrepancy between the GSS figures and the Pew survey results for 2014. GSS says 22.7% identify as Evangelical Christians v. 25.4% for Pew, while GSS says that 12.2% identify as Mainline Christians v. 14.7% for Pew. But, most of these discrepancies come from the fact that Pew is including historically black denominations in its Evangelical and Mainline Christian percentages, while GSS is treating all of the historically black denominations as a separate category (which is probably a more accurate way to describe the reality for most purposes).
Just 1 in 27 Americans identified as Mainline Protestant and regularly attended church in 2014. And, as I've noted previously in other posts, this trend shows no sign of stopping any time soon because the younger you are, the less likely you are to consider yourself a Mainline Protestant, and the regularly church attending population that remains in these churches has been steadily getting older and older. The gradual death of denominations without children and young people is nearly inevitable.
In contrast, 1 in 8 Americans identified as Evangelical Protestant and regularly attended church in 2014.
Meanwhile, 22.8% of Americans in 2014, not quite 1 in 4, but more than 1 in 5, no longer even identify a Christian at all, a number that exceeds the percentage of self-identify as Roman Catholic (20.8% in 2014), whether or not they regularly attend church. Another 5.9% of the population identifies with a non-Christian faith (including Jews, Muslims, Buddhists, Hindus, and Unitarian Universalists).
People who do not identify as Christians now make up 30% of the U.S. population, and 80% of the U.S. population does not attend church on a regular basis. Mainline Christians in predominantly white denominations who do not regularly attend church, who tend to be culturally very similar to the "nones" culturally and in their political beliefs, make up another 8.6% of the population.
About 63% of all people who attend church regularly (including historically black denominations and Roman Catholics) are Evangelical Christians in historically white denominations.