A few weeks ago I asked why modern Christians are so delighted with current divorce rates. As I explained in the post, Shaunti Feldhahn has a new book* and multiple articles proclaiming the good news about our current no fault divorce/sexual morality free form of marriage. I previously avoided delving into her questionable statistics to focus on what I see as a defense of our new form of marriage. In this post I’ll review some of her bold claims and the problematic statistics she offers as evidence to support those claims.
Before I dive into her claims and statistics, it is important to note that statistics around marriage and divorce can be difficult to pin down. In some cases this is because the data doesn’t exist, either because no one has collected the data or we are talking about future events. But even when the data is available and free from controversy it is easy to become confused as to what the data means. When looking at marriage and divorce data you need to always be clear about what question you are trying to answer, and what any given statistic actually tells you. For example, it has been widely reported that we are currently experiencing an “explosion” of grey divorce. There is a kernel of truth here, as divorce rates per 1,000 married women have risen in the older age brackets over the last 20 years, while divorce rates for younger age brackets have declined some. However, this has been widely misreported as couples experiencing an increase in divorce rates around retirement age. This is simply untrue, as divorce rates decline dramatically as the wife ages. Likewise, I’ve previously explained the problem with the way the marriage rate per 1,000 unmarried women is often interpreted.
The answer to these challenges is to be very careful in what data you use and what conclusions you draw from it. This is unfortunately where Feldhahn goes terribly wrong right out of the gate, with the very title of her Catalyst article: Everything We Think We Know About Marriage and Divorce is Wrong. She reinforces this in bold and all caps at the beginning of the article:
I ALSO HAD NO IDEA THAT EVERY ONE OF THE STATISTICS I WAS QUOTING – STATISTICS THAT FIT BOTH WITH CONVENTIONAL WISDOM AND WHAT I SAW REPORTED IN THE MEDIA – WERE NOWHERE CLOSE TO TRUE!
Feldhahn isn’t trying to explain some of the finer points on divorce data, she is claiming the data commonly used is nowhere close to true. This is a bold claim, and proving it would require bold evidence. What she offers instead is more confusion.
Do 40-50% of marriages really end in divorce?
To answer this question in a meaningful way we need to be specific about which marriages we are discussing. Divorce rates vary widely depending on the demographic you are looking at as well as the time frame in question. Also, if we are talking about lifetime divorce rates for a cohort which is still alive, all we can do is create our best model to guess at what the cohort’s lifetime divorce rate will ultimately be.
Feldham makes the extraordinary claim that divorce rates have never come close to the 40-50% statistic often quoted:
Now, expert demographers continue to project that 40-50% of couples will get divorced – but it is important to remember that those are projections. And I’m skeptical because the actual numbers have never come close, and divorce rates continue to drop, not rise! Even among the highest-risk age group –baby boomers—seven in ten are still married to their first spouse. Most of them have had 30 years’ worth of chances to get divorced…and they are still together.
However, her claim about baby boomer divorce rates is misleading at best, or perhaps outright untrue. According to the Bureau of Labor Statistics, 45% of the youngest (ever married) boomers have experienced one or more divorces:
Eighty-seven percent of baby boomers born in the years 1957–1964 had married at least once by the time they reached age 46. Of those who had married, 45 percent had experienced at least one divorce.
Note that above she claimed actual divorce rates have never come close to the 40-50% range, and offered the boomers as proof of this. Yet according to the BLS, the younger half of the boomers already have a 45% divorce rate, exactly in the middle of the very range she is claiming we have never come close to. Note also that this is the floor of this cohort’s divorce rates, as more divorces will occur until they have all passed away.
Feldhahn offers her own estimate of the lifetime divorce rate:
No-one knows what the average first-marriage divorce rate actually is, but based on the rate of widowhood and other factors, we can estimate it is probably closer to 20-25%. For all marriages (including second marriages, and so on), it is in the 31-35% range, depending on the study.
As she herself points out, the kind of lifetime divorce rates she is describing are projections based on the best guess of the person making the prediction. There is no hard data on what the divorce rates will turn out to be (in retrospect) 20 years from today. All we have are educated guesses based on the past, and these are highly dependent on the credibility of the person making the guess. Nothing that I’ve seen of Feldhahn’s handling of the data gives me any reason to believe that she is better at modeling this than the demographers she is claiming to debunk.
Torturing the Barna data until it confesses.
Feldhahn also explains that the Barna data has been terribly misunderstood. I don’t doubt that, given the nature of the statistics involved. However, she goes a step further and re runs the Barna data to see what divorce rates look like for Christians who regularly attend church:
The Barna Group studies were focusing specifically on the divorce rates of those with Christian and non-Christian belief systems and didn’t take worship attendance into account. So I partnered with the Barna Group and we re-ran the numbers: and if the person was in church last week, their divorce rate dropped by 27%. And that is one of the smallest drops found in recent studies: overall, regular church attendance lowers the divorce rate anywhere from 25-50%, depending on the study you look at.
This is very problematic, because she is using current (at the time of the survey) church attendance to explain previous divorce. It could well be that going to church regularly leads to lower divorce rates. However, it could just as easily be that getting divorced tends to cause people to not attend church. The mechanism for this latter possibility could be that the person feels too ashamed of their decision to divorce to continue attending. It could also be that the person who was divorced against their will left the church in outrage when their church failed to stand by biblical marriage.
But even if we could determine that people who go to church today have a lower risk of experiencing divorce in the future, there is still the problem of telling how much of this is due the impact of going to church verses a correlation with something else. For example, divorce rates vary dramatically based on education levels:
Is there something being taught in college or something about campus life which helps women honor their marriage vows? Almost certainly not, because if so then we would expect women with some college to do better than those who never went at all. And why do those who didn’t graduate high school have lower divorce rates than those who graduated or received their GED? The simplest answer is that when we look at educational attainment it very often tends to tell us more about the person themselves rather than what they learned on campus**.
Regarding college attendance vs graduation, the key factor would seem to be the person’s tendency to see a long term project through. This is relevant to Feldhahn’s analysis of the Barna data because going to church every week is also very likely a measure of follow through. The message about sexual morality at the church could be no better than the message young people are learning at college, and we would still expect to find that regular church attendance is strongly correlated with lower levels of divorce.
Feldhahn makes sweeping claims about divorce rate statistics without offering compelling evidence to back them up. Her claim that we have never seen a 40-50% divorce rate is simply untrue, and several other statistics she offers are highly misleading at best. However, even with the glaring problems with the statistics she presents, the far bigger issue is the desire to put a happy face on our new sexual-morality-free view of marriage. In this new view marriage isn’t about making and keeping a lifetime vow, it is about couples therapy. Lifetime marriage is no longer seen as the moral place for romantic love and sex, but instead romantic love is seen as the moral place for sex and marriage. Nearly all Christians have adopted the same view as the rest of the culture, where the focus is now to make the couple (mostly the wife) happy enough in their marriage that they won’t choose to divorce. This new view of marriage is front and center in Feldhahn’s conclusion of the Catalyst article (emphasis mine):
Those of us who work with marriages may secretly wonder whether there is reason for our ministry, if the news about the divorce rate is better than we think. And the answer is a resounding yes. Because I have seen in the research what every marriage counselor knows intimately: divorce isn’t the greatest threat to marriage. Discouragement is.
*I have not read the book, so this post is focused on the statistics she presents in the articles. However, the articles are part of the promotion of her book, and far more people will be influenced by the statistics she presents in her articles than will read her book.
**In the case of divorce rates for those who didn’t graduate high school being lower than those who did graduate, the unexpected result here appears to be explained by different divorce patterns among first generation immigrants. When foreign born women are excluded, this paper notes that high school graduates divorce at lower rates than those who didn’t complete high school:
When analyses are limited to native-born women, the relationship is no longer curvilinear; the highest educated women have the lowest rates of first divorce (14.5) compared to those with less than a high school education (20.9), high school diploma/GED (18.0) and some college (24.2) per 1,000.
See Also: Nowhere close to true.