Mark Hoofnagle, a MD/PhD Candidate in the Department of Molecular Physiology and Biological Physics at the University of Virginia, has a very interesting examination of myths and facts about divorce at the denialism blog. Here is the gist of the post:
People believe marriage is declining, divorce is increasing, and in general families are falling to pieces all over. But when you actually look at the data, this picture is false.
When viewed over a 60-year period, the marriage rate, for example, has remained fairly constant, with several long periods of slight ups and downs. The number of marriages per 1,000 people now hovers at 8.5, compared with a 60-year average of 10.1. The variations that did occur tended to come in times of depression and war, when fewer people got hitched because of economic or obvious logistical reasons.
Interestingly, while the 1950s are thought of as boom time for the family, the marriage rates were relatively low through the latter half of the decade and into the early 1960s. (Indeed, today's rate beats that of 1958.) Then in 1968, when hippies were supposedly lovin' the ones they were with, the marriage rate rose and stayed relatively high through 1975. It's only when the numbers are viewed within a narrow 20-year context that marriage looks to be on its deathbed.
And despite the buzz that divorce is at epidemic proportions, the NCHS report shows that the divorce rate has, in fact, been slowly declining since its perilous peak in 1981, when it reached a rate of 5.3 divorces per 1,000 people.
Further, the idea that 50% of marriages end in divorce is just an oft-repeated myth. It's a statistical flub that comes from comparing the number of marriages in a given year to the number of divorces in a given year. However, since the marriages and divorces aren't occurring in the same year, this doesn't give an accurate picture of how many marriages are failing and is notoriously susceptible to population dynamics. Your actual chances of a failed marriage are about one in four, and the 50% figure is considered to be statistical nonsense.
To establish an actual divorce rate requires tracking and analyzing significant samples of actual marriages through decades, which is not an easy task. Recent US scholarship based on such longterm tracking, reported for example in the New York Times on April 19, 2005, has found that about 60% of all marriages that result in divorce do so in the first decade, and more than 80% do so within the first 20 years; that the percentage of all marriages that eventually end in divorce peaked in the United States at about 41% around 1980, and has been slowly declining ever since, standing by 2002 at around 31%,
Divorces reached a peak in the early eighties at a rate of about 40 percent (which was the highest level since the 1950s), and divorce has since entered a 20-year decline. The current rate of divorce is about 30-34% in any given year, and is lower among the college-educated (about 20%), Catholics, Muslims, and atheists. In a comparison of various religious denominations, the group that actually performs the worst are Baptists and Evangelical christians, so I guess it makes sense that they seem to be most concerned with the idea of a divorce epidemic.
Read it all.
I like the hard-nosed approach to the data, and his analysis seems sound to me.
The Associated Press also has a discussion of divorce that includes a few experts opining that that one reason why the divorcerate is going down is that the number of people who are simply cohabitating is going up:
Despite the common notion that America remains plagued by a divorce epidemic, the national per capita divorce rate has declined steadily since its peak in 1981 and is now at its lowest level since 1970. Yet Americans aren't necessarily making better choices about their long-term relationships. Even those who study marriage and work to make it more successful can't decide whether the trend is grounds for celebration or cynicism.
Some experts say relationships are as unstable as ever — and divorces are down primarily because more couples live together without marrying. Other researchers have documented what they call "the divorce divide," contending that divorce rates are indeed falling substantively among college-educated couples but not among less-affluent, less-educated couples. "Families with two earners with good jobs have seen an improvement in their standard of living, which leads to less tension at home and lower probability of divorce," said Andrew Cherlin, a professor of public policy at Johns Hopkins University.
America's divorce rate began climbing in the late 1960s and skyrocketed during the '70s and early '80s, as virtually every state adopted no-fault divorce laws. The rate peaked at 5.3 divorces per 1,000 people in 1981. But since then it's dropped by one-third, to 3.6. That's the lowest rate since 1970.
Read it all. I am sceptical that the divorce rate has a great deal to deal with the number of people choosing cohabitation. I think the more interesting statistic is that divorce rates increase as you go down the socio-economic ladder.