Divorce Rate Horseshit
Jill posted this, about this, which is a great way to get annoyed. It’s so annoying that I don’t even have the patience to take on the whole thing. However, the author — who, with a Ph.D., I think we have a right to expect better from — quotes an old, inaccurate fakey bullshit pop-culture statistic-like-factoid about marriage that makes me cringe. I feel like I should respond to that.
Dr. Walsh says, “the rest are vulnerable to a 50% divorce rate.” There’s that bullshit factoid. As geek friends of mine say, “it’s so bad, it’s not even wrong.” That is, it doesn’t even represent something for which a real number can simply be plugged in; it’s not a real stat, it’s just pop culture.
How many of the people who get married now will get divorced? That’s what the number purports to be. What’s the answer? We don’t know! We can’t, the answer is in the future! We know, for people who got married, how many have gotten divorced. All attempts to speak for marriages that start today, then, are projections.
There was a spike in divorce rates, up to a peak in the late 1970s, as economic changes, legal reforms, a bunch of bad marriages caused by societal constraints in a repressive era, and yes, the effects of the Second Wave arguing for actual autonomy for women, took effect. The pig in the python peaked around the time anyone gave a shit about Bo Derek or Dudley Moore. The per-capita rate, and the rate per married person, have been on their way down pretty steadily since (the latter more than the former because marriage is declining slightly, too.) The chart looks like this.
Projecting forward is highly inexact. Divorce rate changes wildly with things like age at first marriage, income and education, so one has to estimate these factors as a proportion of the population of newly married people to even try a projection. Nobody really knows, for example, what would happen to the rate if we stopped reserving marriage for people whose partners are opposite sex (whatever that is; don’t get me started on the silliness of the states deciding which people folks can marry if they have a trans history). And when factoring demographic changes, which way do the causation arrows go? If women who marry for the first time after age 30 divorce less (and they do), does an increase in that group as a proportion of marriages mean fewer divorces, or does it mean more bad marriages in this demographic and result in an increase in that group’s rate?
All that said, apparently, the current loosey-goosey frequently revised statistical model-based guess is something like 41%. Which is, to my mind, a long way from fifty. I get that from these folks, who say:
PROJECTION/PREDICTION. This is the Census Bureau’s often-cited “50%” rate, the proportion of marriages taking place right now that will eventually divorce, which has since been revised downward to roughly 43% by the National Center for Health Statistics but was moved back up to around 50% by the Census Bureau in 2002, with even more ifs ands and buts than usual. Most recently, according to the New York Times, it has been revised downward to just over 40%.
But, really, that’s a model-based wild-ass guess.
About those demographics: social position matters a lot. People love to break it down by religious denomination, but really I doubt that has a lot of effect. Progressives and anti-religious atheists like to note that Protestant evangelicals have higher divorce rates than other folks, but it’s not clear to me that this difference remains if controlled for income, education and age at first marriage.
I read a really good rundown of all this in the NY Times in 2005, and it has been posted on that divorce reform cite. The Times article says:
Researchers say that the small drop in the overall divorce rate is caused by a steep decline in the rate among college graduates. As a result, a “divorce divide” has opened up between those with and without college degrees, said Dr. Steven P. Martin, an assistant professor of sociology at the University of Maryland.
“Families with highly educated mothers and families with less educated mothers are clearly moving in opposite directions,” Dr. Martin wrote in a paper that has not yet been published but has been presented and widely discussed at scientific meetings.
As the overall divorce rates shot up from the early 1960’s through the late 1970’s, Dr. Martin found, the divorce rate for women with college degrees and those without moved in lockstep, with graduates consistently having about one-third to one-fourth the divorce rate of nongraduates.
But since 1980, the two groups have taken diverging paths. Women without undergraduate degrees have remained at about the same rate, their risk of divorce or separation within the first 10 years of marriage hovering at around 35 percent. But for college graduates, the divorce rate in the first 10 years of marriage has plummeted to just over 16 percent of those married between 1990 and 1994 from 27 percent of those married between 1975 and 1979.
See also this sex/race/education breakdown, thought he race breakdown is a severely deficient black/white breakdown.
There’s a metric fuckload of information at Centers for Disease Control, which I’m sure someone has time to read …. but Table 3 at p. 5 of the Results tab, and Figure 1 on the next page which is a colorful chart that I’m too inept to paste in here, show the steady decrease in divorces as women’s age at marriage increases.
Women who marry later, women who are more educated, women with more family income … I’m sensing a theme. Women who have more bargaining power, and who choose to marry, find marriages they stay in at higher rates. The next time the shrill antifeminist whiners call for the whaaaaaambulance, claiming that feminism is an ideology of victimhood, bear this in mind. The opposite is true, which is why they try so hard to lie about us. The next time the mainstream media turns on the “you’ll never find a man” scare tactics, remember that empowerment leads to better matches.