In the comments of a recent post, a discussion arose about Christian divorce rates. One of the commentors insisted that Christian divorce rates were no different than anyone else’s divorce rates. I pointed to the recent, extensive polling data developed by Gallup and Baylor University on the issue, as presented by one of the leading sociologists in the United States, Rodney Stark, in the book What Americans Really Believe.
R. Stark does not provide data on divorce rates by belief, but he informs as to divorce rates correlated with attendance at religious services. “The average person is 50 percent less likely to be divorced or separated if he or she attends religious services at least twice a month.” Stark, Ch. 23, What Americans Really Believe. On the other hand, “[t]he divorce rate among those who never attend religious services is close to double that of weekly church-goers.” Id.
I have heard others refer to a study by The Barna Group showing that "born again" Christians have the same divorce rate as the rest of the country. Here is how one Christian news source describes Barna's study:
The Barna Group found in its latest study that born again Christians who are not evangelical were indistinguishable from the national average on the matter of divorce with 33 percent having married and divorced at least once. Among all born again Christians, which includes evangelicals, the divorce figure is 32 percent, which is statistically identical to the 33 percent figure among non-born again adults, the research group noted.
There are a number of reasons not to take this conclusion at face value. Even Barna's numbers show that "evangalicals" had a divorce rate of 26%, lower than the national average. Moreover, according to Barna, Catholics and social conservatives also have lower rates of divorce, at 28%. There is also data supported in other studies showing that those with higher income and college degrees are less likely to get divorces than those with lower income and no college degrees. Further, "born again" and evangelical Christians were significantly more likely to get married than those with no religion, who are more likely to live together outside of marriage.
In this post, I want to focus on two significant issues I have with how Barna's numbers have been used.
First, Barna compared the divorce rates of Christians with the divorce rate of other Christians. There are many observant mainline denominational Protestants and Catholics who may not embrace the "born again" language Barna used to distinguish "born again" Christians from others. Remember that Stark compared those attending "religious services" with those who do not.
Sociologist and blogger Bradley Wright provides some very helpful insights into Barna's numbers and uses other polling data (GSS and Midlife in the U.S. Study) to add further corrective. Using MIDUS data:
We find the following divorce rates by religious group:
1) Christians reporting a born-again experience: 36%
2) Christians not reporting a born-again experience: 34%
3) Members of other religions: 37%
4) Individuals with no religious beliefs: 52%
As B. Wright explains, what Barna did is compare No. 1 against Nos. 2, 3, and 4. Given the preponderance of Christians in the United States, this means that we have the divorce rates of one set of Christians being matched against the divorce rate of another set of Christians. This tells us little about Christian divorce rates overall. So, what happens if you compare all Christians against the rest of the U.S. population? The Christian divorce rate is 35% and the non-Christian divorce rate is 45%. That is a significant 10 point shift. Accordingly, Christians do not have the same rate of divorce as non-Christians.
Second, comparing R. Stark's conclusions to Barna's conclusions is comparing apples to oranges because Stark measures attendance at religious services rather than merely stated belief. R. Stark is thus using indicia of religious belief other than self reporting, is testing those who are more exposed to Christian teaching and doctrine, and is likely accounting for fervency of belief. As R. Stark points out, this results in an even greater shift in divorce rates, with regular attenders having half the divorce rate as non regular attenders.
B. Wright, using GSS data, reinforces this point. Here is the GSS data (2000-2004) comparing divorce rates by frequency of church attendance:
49% Never attend church
46% Less than once a year
46% About once or twice a year
42% Several times a year
42% About once a month
41% Two or three times a month
31% Nearly every week
27% Every week
28% Several times a week
Although a little older, the GSS data reinforces the Gallup/Baylor data reported by R. Stark. People that go to church more frequently have a much lower rate of divorce. The breaking point appears to be going "nearly every week" or more, with a ten point jump to the next lower level of attending, "two or three times a month."
B. Wright also provides useful GSS data about denominational and racial differences correlated with divorce rates.
Here are the divorce rates among ever-married respondents in the General Social Survey (GSS, 2000-2004)—one of the best known sources of sociological data. “Frequent” is attending church about once a week or more.
58%, non-frequent Black Protestants
54%, non-frequent Evangelicals
51%, no religion (e.g., atheists & agnostics)
48%, ALL NON-CHRISTIANS
48%, non-frequent, other religions
47%, frequent Black Protestants
42%, non-frequent, mainline Protestants
41%, ALL CHRISTIANS
41%, non-frequent Catholics
38%, frequent other religions
34%, frequent Evangelicals
32%, ALL FREQUENT CHRISTIANS
32%, frequent mainline Protestants
23%, frequent Catholics
Once again we see that the key to a low divorce rate is frequency of church attendance. Unfortunately, there is a high rate of divorce among Black Protestants even if they frequently attend church (though the rate is 10 points lower than for non-frequent Black Protestants). Among likely candidates for explaining this result is higher rates of poverty, a history being discriminated against, and perhaps ill-conceived government assistant programs. What is interesting is that even factoring in a large subgroup with elevated levels of divorce, frequent church attending Christians have one of the lowest divorce rates in the United States. Indeed, their divorce rate is lower than those claiming "No religion" or who infrequently attend religious services.
Christians in the United States, whether describing themselves as "born again" or not, have significantly lower divorce rates than non-Christians in the United States. Those Christians who attend church frequently have even lower divorces rates; among the lowest in the United States.
(Special thanks to B. Wright for his work on this, especially the analysis of Barna's numbers. Check his blog out.).