Networks and Religion: Changing Our Religion

In the opening chapter of my forthcoming book ("Networks and Religion: Ties that Bind, Loose, Build-up, and Tear Down"), I explore the "surprising" persistence of religion in spite of predictions in recent centuries by philosophers (e.g., Voltaire, Karl Marx), social scientists (e.g., Sigmund Freud, Anthony Wallace), and other intellectuals (e.g., Christopher Hitchens, Richard Dawkins, and Sam Harris) that religion's death was all but inevitable and soon to be realized.

These predictions appear to be no more than wishful thinking, however. As I've noted previously, the world is more religious than ever ("Is The World More Religious Than Ever?") and will probably continue to be ("Trends in World Religions: More, not Less"). Moreover, although there has been a recent decline in religious affiliation in the U.S., church membership rates are still higher today than they were in 1776 ("Religion's Surprising (at least to some) Persistence"). And finally, even among those who no longer claim a religious affiliation, they are often still quite religious. For example, of the 20% of Americans who in 2012 claimed no religious affiliation, 18% considered themselves religious, 30% had had a religious or mystical experience, 33% said they believed that religion is somewhat or very important, 37% considered themselves to be spiritual but not religious, 41% prayed weekly or more, and 68% said they believed in God. European "nones" display similar patterns (although you'll have to pick my book up to see how).

What I find interesting, and which I spend less time on in my book, is an observation made by Rodney Stark and Bill Bainbridge back in the 1980s ("The Future of Religion: Secularization, Revival, and Cult Formation") that when people leave a religion, they seldom stop believing. Instead, they usually trade in one belief for another. This led to their prediction that where traditional forms of religion are weak, less traditional forms of belief and practice will flourish (or at least be more prevalent). One of Stark's students, Chris Bader later teamed up with Stark in order to test this hypothesis by comparing the rate of astrologers (per 10,000) by state with state church membership rates, and as expected, they found a strong, negative correlation between the two. That is, where church membership rates were high, the rate of astrologers was low, but where church membership rates were low, the rate of astrologers was high.

I decided to update their study, using search data from Google Trends, which has been used by other social scientists to explore a number of social phenomena (e.g., see "Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are"). The data are indexed, which is preferable to raw counts, because it makes them comparable. Here, I compared (from 1/1/2013 to 12/31/2017) the proportion of searches (by state) for "Churches" with "Astrologers," "New Age," "Yoga," and "Zen" and found strong negative correlations between all four (see the note on correlations below):

"Churches" & "Astrologers" = -0.670
"Churches" & "New Age" = -0.213
"Churches" & "Yoga" = -0.711
"Churches" & "Zen" = -0.684

As with all tests of hypotheses, this doesn't "prove" that Stark and Bainbridge are right. However, the results are consistent with their prediction and highly suggestive. I'll stop here for now, but on a closing note, these results are also consistent with the sociologist Christian Smith's claim that human beings are "moral, believing animals" whose primary drive in life is the quest for meaning, and one that we often satisfy by joining (traditional and nontraditional) communities of faith. In other words, people may leave a faith tradition (e.g., one in which they were raised), but they typically will search for some other community that satisfies their quest for meaning. I'll return to Smith's argument in a later post ("Moral, Believing, and Storytelling Animals").

Note: Correlation coefficients range from -1.00 to 1.00. A correlation of 1.00 indicates perfect positive correlation, while a correlation of -1.00 indicates perfect negative correlation.