Christians were shocked when George Barna reported in 1999 that married Christians were not only as likely to divorce as anyone else but actually get divorced more often than the godless hell-bound. Atheists, of course, were delighted and claimed they weren’t the least bit surprised.

These findings confirm what I have been saying these last five years. Atheist ethics are of a higher caliber than religious morals. It stands to reason that our families would be dedicated more to each other than to some invisible monitor in the sky.

– Ron Barrier, national media coordinator, American Atheists

Thus demonstrating once more that an atheist is not primarily defined by any absence of any particular belief but by that sweet and pleasant nature that makes him so universally popular.

However, Barna’s conclusions in his 1999 study were marred by a serious problem. The divorce rate he calculated was not based on the percentage of marriages that had failed within each religious affiliation but, rather, the percentage of divorcees out of the total population of the affiliation. Since one cannot get divorced if one has never been married, the conclusions presented a very misleading picture of the comparative likelihood of divorce for Christians, atheists and everyone else. This flaw may be why the study is no longer available on the Barna site; it is certainly the reason that Barna noted in a later study “One reason why the divorce statistic among non-born-again adults is not higher is that a larger proportion of that group cohabits, effectively side-stepping marriage, and divorce, altogether.”


While the 2001 American Religious Identification Survey has the problem of any survey depending on the ability of the individual to correctly identify his own religion, its questions relating to marital status were asked in a manner devised to elicit more useful information. After demonstrating that its own results regarding marital histories were very similar to those reported by the 2000 U.S. census, ARIS appeared to support Barna’s initial conclusion when it showed that 14 percent of Pentecostals and 12 percent of Baptists were divorced, compared to only 9 percent of those identifying as “no religion”.

But only at first glance, because ARIS also showed that 78 percent of Baptists and Pentecostals were, or had been, married, compared to only 34 percent of atheists. This means that 16.7 percent of Baptist and Pentecostal marriages ended in divorce compared to 26.5 percent of the irreligious marriages. If one takes the varying populations of the different Christian denominations properly into account, the result is that only one in eight of all Christian marriages, 12.5 percent, end in divorce. So it is not only an exaggeration, it is statistically incorrect to assert that Christian marriages are more likely to end in divorce, because atheist marriages are more than twice as likely to fail even though atheists are less than half as likely to get married in the first place.

Perhaps the seven marriages of the atheist champions Bertrand Russell and Richard Dawkins should have been Barna’s first clue that his initial conclusions were awry.

George Barna’s work is both interesting and worthwhile, but it is too often marred by inadequate question structures and other peculiarities. For example, one Barna study reported that 40 percent of Americans described themselves as evangelical Christians, another Barna study declared that evangelicals only represent 9 percent of the American population. While this is evidence that Barna is in the vanguard of statistical science through his demonstration of the unreliability of religious self-identification – it’s both amusing and intellectually appalling to read of evangelical Christians who believe that everyone is going to heaven and atheists who believe in God – but it also renders his work susceptible to abuse by propagandists and misunderstanding by the statistically illiterate.

None of this excuses the frequency of divorce within the Christian body, but I await, with some anticipation, Mr. Barrier’s declaration that atheist ethics are of a lower caliber than religious morals and that atheist families, to the extent that they even exist, are less dedicated to each other.



Related special offers:

“Why Won’t They Listen?”

“God’s Got a Problem”

“Annals of the World”

Note: Read our discussion guidelines before commenting.