Divorce statistics

Following on last week’s marriage theme: what are the odds of a marriage lasting?

Obviously you can’t know the fate of your specific marriage based on national statistics. But if you hear “half of marriages end in divorce,” that probably affects your thinking about whether it’s a good idea to try at all.

That statistic is only kind of true. 50% is roughly the overall divorce rate in the US, but your personal risk is affected by characteristics like your education level, ethnicity, and whether you already have children. 

This data won’t be super relevant for those who don’t resemble the statistics much – if you’re a 37-year-old with one previous divorce marrying a partner who’s already in a poly family, there hasn’t been a study on that.

And many of the things that affect your chances are characteristics of who you and your partner are, rather than behaviors you can alter. Stories that say “larger weddings are correlated with lower divorce rates, so you should invite more people to your wedding” are bogus. It’s likely that richer people have larger weddings, and also have lower divorce rates for other reasons.

Lastly, none of this tells you how happy the people are. There are smaller studies on marriage satisfaction, etc, but not this kind of large-scale statistics.

Data woes

Most of the US data sources and methods are pretty bad, since the government no longer requires states to report divorce statistics. So a lot of the numbers are cobbled together from various sources, and overall rates are probably underreported.

Obviously it takes at least 10 years to get stats on how likely it is for a marriage to last at least 10 years, and so on. So it’s hard to know how generations differ from each other until time has passed.

For whatever reason, the standard is to use women’s characteristics (age, race, and so on) in stats about the marriage. Same-sex unions are represented very partially or not at all in the statistics.

Some sources count only legal divorces, while others count “dissolution” (divorce or separation). Some studies look only at first marriages (which are less likely to dissolve) and others include remarriages.

I’ve tried to avoid sources that do fancy adjustments to the data, because I don’t understand how reasonable those are.

I’m discussing US data here. Divorce rates are generally lower elsewhere except in Eastern Europe.

Ok, on to the findings!

Timing factors

There are different time-related factors:

• Age of the people at time of marriage 
• Age of the marriage itself (there are times during the marriage when divorce is more likely)
• Cohort (the generation born in the 1950s have a different pattern than those born in the 1970s or 1990s, etc.)

Age at time of marriage

Different sources and ways of doing the stats indicate somewhat different patterns here. Women who marry in their teens have the highest divorce rate. 30 appears to be a bit of a sweet spot, but not by much. (source)

I expect this is mostly about what you are like, not exactly when you get married: marrying the same person at age 24 vs 29 probably doesn’t affect your chances much if at all.  In the words of the sociologist writing the piece linked above, “if the mechanism is selection, then changing your behavior to ride that curve will not work.”

Timing during the marriage

This is all over the map – the periods of higher and lower divorce rates during a marriage varies on the age of the people when they married. More.

Cohort

Marriages that began in the 70s and 80s had a higher divorce rate than more recent marriages. 

“The shifting age pattern of divorce suggests a cohort effect. The same people who had unprecedented divorce incidence in 1980 and 1990 when they were in their 20s and 30s are now in their 40s, 50s, and 60s. The Baby Boom generation was responsible for the extraordinary rise in marital instability after 1970. They are now middle-aged, but their pattern of high marital instability continues.” Source

“The decline in divorce rates among women under age 25 probably reflects increasing selectivity of marriage. Fewer young people are getting married: over 40 % of the population in 2008 had not married by their 30th birthday, marking a fourfold increase since 1980. With the rise of cohabitation, it is likely that many couples who would have been at the highest risk of divorce in the past—for example, those entering unions as teenagers as a result of an unplanned pregnancy, or with low levels of income and education—are forgoing marriage entirely (Cherlin 2004; Smock et al. 2005). As pressures to marry recede, people can be more selective about their partners; thus, it makes sense that marriages may become more stable.” Same source as above

Your parents’ history

“Intergenerational transmission of divorce” is a thing. Unsurprisingly, one study found that “youth from unstable families tended to form unstable unions, whereas youth from stable families tended to form either stable unions or no unions at all in early adulthood. Presumably, many offspring who had not formed unions were biding their time until they had finished their educations, became established occupationally, or found the ‘right’ partner. Postponing first union formation until the late 20s or 30s may be an adaptive strategy for some youth in an era of high union instability.” There are various theories as to why you might see this, including genetic aspects of personality, relationship skills learned from your parents, etc.

Education

More educated people have lower divorce rates (I think this is about education level at the time of the survey, not necessarily at the time of the marriage — again, this is probably about characteristics of you as a person rather than whether you had completed a degree at the time of marriage.) The divorce/separation rate for college-educated women appears to be declining as of the late oughts. (source)

Children

If there’s a child before the marriage, the chance of the marriage dissolving in the first 5 years is roughly double that of marriages without children before marriage. Source

Another source:

Situation	Chance of divorce within 15 years
Child born at least 8 months into marriage	33%
No child	50%
Child born 0-7 months into marriage	52%
Child born before marriage	56%

Race / ethnicity

More data on different Asian ethnicities.

Cohabitation

This was surprising to me, but living together before marriage increases the chance of divorce. This might be changing, depending on which way you squint at the data. In any case, demographers usually think this is because of selection effects – for example, more religious people might be less likely to live together before marriage and also less likely to divorce. It doesn’t seem worth changing your behavior here to try to get on the right side of this one.

Not just marriage

Cohabitation is more common than it was in the past. Some relationships that would have included marriage and then divorce in earlier times now aren’t captured in the marriage statistics. “Cohabitation has moved some of the instability of family life out of the statistical accounting system of marriage and divorce. It has thus likely contributed to the plateau by removing an increasing number of disruptions that, in an earlier time, would have occurred after marriage.” Children’s chance of family instability has risen since the 1980s, largely because of having parents who are cohabiting rather than married and the higher chance of cohabitation arrangements breaking up. (source)

Cancer

One more thing I learned: in 2009 there was a news story that men with cancer are less likely to get divorced and women are much more likely to get divorced, which I found really sad. It’s based on a small study. This review includes much more data and finds that cancer diagnosis is associated with a slightly decreased divorce rate. The exception is cervical cancer (which might be because it directly affects sexual health, or might be because more sexual partners increases risk of human papiloma virus which increases risk of cervical cancer.)