Church Attendance Used to Drive Up Trust: It Doesn’t Anymore
Unsplash image by Nolan Kent
(ANALYSIS) A couple months ago, I wrote the following:
My bias is that I believe that religion is a net positive for a functioning society. I’ve been clear about that for quite a while now. But there’s a huge caveat when it comes to that assertion. Religion is a net positive only when people actively engage in all the aspects of religious life — including regular corporate worship.
I think that if you look at the preponderance of the evidence in social science, it’s hard to argue that regular congregational attendance is a net negative for individuals and society. There are tons of examples of how it seems to improve mental health, generate feelings of tolerance and provide opportunities for people to volunteer in their communities.
There’s another area in which religious attendance has been demonstrated to generate positive outcomes: interpersonal trust. This type of trust is pretty simple, really. It’s the idea that we can have some assurances that random strangers are not out to get us. Here’s how scholars have defined it:
The confidence in another person (or between two persons) and a willingness to be vulnerable to him or her (or to each other).
I don’t think it takes a huge leap in logic to say that a reasonable baseline of interpersonal trust is an essential part of any functioning society. Without it, we have “The Purge.” Everyone is always on high alert, thinking that someone is just looking for a chance to take advantage of them.
So, what’s the state of interpersonal trust in the United States over the last five decades, and what role does religion play?
Here’s how the General Social Survey poses the question: “Generally speaking, would you say that most people can be trusted or that you can't be too careful in dealing with people?”
They also let them respond with, “It depends.”
This is the high-level view of interpersonal trust over the last five decades.
In the early 1970s, a bare majority of Americans (51%) said that they “can’t be too careful” when it comes to trusting other people. About 45% of folks said that others can be trusted. The trend lines only diverge from there.
By 2000, 60% of respondents said that others can’t be trusted, compared to 35% who believed that they could. In the most recent survey, collected in 2022, 65% of people chose the distrusted option, 26% said other people could be trusted and 9% said that it depends on the circumstances. Obviously, interpersonal trust is in a noticeable decline.
Let’s get a bit more granular now. I split the sample into Democrats and Republicans and calculated the same trend lines between 1972 and 2022.
Democrats in the 1970s were significantly less trusting of others compared to Republicans during the same time period. About 55% of Democrats said that you couldn’t be too careful in trusting others. That sentiment was expressed by 42% of Republicans. But, generally speaking, the average Democrat today is slightly less trusting now than five decades ago. The percentage who said “could be trusted” has dropped by about ten points (41% to 31%).
Republicans saw a more dramatic shift. In the early waves of the GSS, the share who had a trusting view was higher than those who were skeptical of their fellow man. Those lines crossed in the early 1980s. Now, 65% of Republicans don’t trust other people, while just 28% of them say that others can be trusted. For Democrats, those percentages are 57% and 31%.
Interpersonal trust is one of those areas in which a ton of factors can be important, though. Partisanship is one of them, but maybe the most predictive variable is level of educational attainment. You have to consider just how much you have to trust other people to earn a college degree. Your roommates, your classmates, your professors, support staff, and so on. On a college campus, you are always around other people, and oftentimes you have to work in groups to complete assignments.
Here’s how educational level impacts interpersonal trust:
These results can’t be any clearer. Educated people express much higher levels of interpersonal trust than those with lowers levels of education. In surveys conducted in the 1970s, 60% of those with a high school diploma or less said that folks can’t be trusted. Among those with a graduate degree, only 28% gave the skeptical response.
That same pattern exists in every single decade in this data. If anything, the relationship between education and trust has only gotten more severe. The gap in the 1970s was 32 percentage points on the distrustful answer. In the 1990s, it had risen to 35 percentage points. In the most recent data collected, it was even larger, at 39 percentage points.
The main culprit for that growing divide is that those with low levels of education have grown more distrustful: 60% in the 1970s, up to 77% in the 2010s. I think this should be ringing an alarm bell for American democracy. There are lots of folks out there with low levels of education who are *deeply* distrustful of their fellow man.
But this Substack is called “Graphs about Religion,” so let’s get to that religion part, because I think that plays a really big role in interpersonal trust. Let’s keep it simple at this point — this is the share of each attendance level that says that other people generally can be trusted.
Note how consistent those trend lines are for the first four decades of the GSS. They tend to tilt upwards — not dramatically so, but it’s definitely a positive association between religious attendance and interpersonal trust. One thing that I see consistently here is that it’s the “almost weekly” and “weekly” categories where trust tends to be at its highest (those are second and third bars from the right in each graph). Trust does seem to be dip a bit among those who are going to church services multiple times per week, though.
But then things change in the data collected in 2010-2018. That trend line, which has been consistently positive for four decades now, turns downward ever so slightly. If you look at these results from that 10-year time period, it’s clear that at a minimum, there’s no more positive association between religious attendance and trust. If anything, it may be a slightly negative relationship now.
Let’s not get ahead of ourselves, though. This may be a situation where it’s there in the bivariate, but the result shakes out when you test the relationship in a regression analysis. So, let’s go ahead and do that now. Here are the controls: gender, race, partisanship, age, education and income. All the basic demographics.
And the finding holds. That’s really interesting. I want to point out a couple things, though. The first is that trust really took a dive in the 1990s. It dropped about eight percentage points in that decade from the prior one. That’s a good little breadcrumb for future analysis. It’s also important to note that the upward slope of the line really slowed down in the 1990s data, too. It’s not quite flat, but close to no result. In 2000, the line did become clearly positive again, though. But that’s likely because the low attenders were a lot less trusting.
But that line for the 2010s decade is just pointing downward, no doubt about it. Religious attendance used to clearly drive up interpersonal trust, by about 5% from the bottom of the attendance spectrum to the top. Now, overall trust seems to decline just a bit (2-3 percentage points).
Obviously, the question is why the relationship has flipped. I don’t know if I have an easy answer to that one, but let me hypothesize for a bit. I think this has a lot to do with the work on social capital, specifically the discussion of bridging and bonding. There’s a great discussion of those concepts here.
Simply stated, church used to be a great way to interact with folks who were different than you. They voted for a different candidate, came from a different economic background — folks with doctorates sitting next to folks with a high school diploma. That offers a tremendous amount of opportunities to learn about other people, build bridges, generate social capital and all kinds of good things.
Now, houses of worship have become monocultures. That’s exactly the point I make here.
Churches, synagogues and mosques are full of educated, middle class folks who did everything “right.” They got married and had kids and lived a classic American life. There isn’t a whole lot of mixing anymore on a Sunday morning. Most White churches now are overwhelmingly Republican. There’s no chance to get to know a liberal when you are a conservative. Now, we are stuck in our enclaves of worship and peer out with suspicion on people who are different than us.
So, interpersonal trust is dropping because we are not seeing real life experiences of “other people” being good, generous citizens. Instead, we just see them being portrayed in the media as being immoral, evil and looking to take advantage of anyone who is different than them. It’s a sad state of affairs, not just for American religion but also American democracy.
This piece was originally published in Ryan Burge’s “Graphs About Religion” Substack.
Ryan Burge is an assistant professor of political science at Eastern Illinois University, a pastor in the American Baptist Church and the co-founder and frequent contributor to Religion in Public, a forum for scholars of religion and politics to make their work accessible to a more general audience. His research focuses on the intersection of religiosity and political behavior, especially in the U.S. Follow him on Twitter at @ryanburge.