Whatever resistance state Sen.Bob Oke faced recently as he tried to save his sinking covenant marriage bill—which would allow divorce only in such extreme circumstances as physical abuse or adultery for those who sign a special marriage contract—he found one premise unchallenged. "Certainly the people that suffer most in any kind of divorce are the children," he said in phone interview. "Statistically, we know they're much more likely to get in trouble."
See end of story for related links
Ah yes, statistics—the ammunition that gets rolled out every time one of these increasingly common anti-divorce bills pops up in legislatures across the country. This session of Washington's legislature saw two divorce-reform bills: Oke's, modeled on a recent Louisiana state law, and one sponsored by Sen. Val Stevens that would repeal no-fault divorce for everybody—not simply those who sign a special contract. While Stevens' bill died in committee, Oke's did surprisingly well, passing out of the Law and Justice Committee and on to the Rules Committee, where it got permanently stuck.
Just what are these statistics that Oke is thinking of? The Port Orchard Republican admits to some confusion over the matter. "I thought it was just common knowledge," he says. "Nobody's asked me for that, to be truthful."
Oke is right in that conservatives and liberals alike believe, as Barbara Dafoe Whitehead's famous "Dan Quayle was right" article in The Atlantic Monthly put it, that "the social-science evidence is in." Research statistics, it is often said, show that divorce greatly increases a child's chances of living in poverty, dropping out of school, getting pregnant as teenagers, having emotional problems, abusing alcohol and drugs, and even becoming criminals.
Those in Olympia who would attempt to save the children by making divorce a Herculean task would do well to pay a visit across town to Evergreen State College social historian Stephanie Coontz. A frequent national commentator whose latest book is titled The Way We Really Are: Coming to Terms with America's Changing Families, Coontz has made it a mission of late to counteract what she characterizes as the "real sloppy and suspect overgeneralizations" made by the divorce doomsayers.
"There are times," Coontz concedes, "when divorce clearly does make things worse, by creating financial insecurity, by getting the custodial parent depressed—which interferes with parenting—by forcing the custodial parent to move, or by intensifying parental conflict, which frequently happens."
She believes, however, that the dangers of divorce have been exaggerated. A case in point is California therapist Judith Wallerstein's widely cited study of the children of 60 families who went through a divorce in 1971. Twenty-five years later, Wallerstein reported that many of the children were haunted by emotional problems well into adulthood; a full half in a follow-up study of 26 kids were addicted to drugs or alcohol.
"The problem is, the families are not a representative sample," Coontz comments. She has repeatedly maintained that the families Wallerstein studied were referred to her for therapy because they had problems in the first place. Wallerstein vehemently disputes that notion, saying that she found the families by requesting local divorce lawyers to send clients her way, and offered no therapy in exchange for participation.
Nonetheless, as Coontz points out, it is unwise to make sweeping generalizations from a small and localized sample like Wallerstein's. And Coontz is correct when she asserts that Wallerstein's "estimates of the danger of divorce are way out of proportion to what everybody else gets." Researchers typically find about a 20 percent to 25 percent incidence of behavioral problems among children of single-parent families—far less than Wallerstein's 50 percent but still twice as high as the rate for children of two-parent families.
"On the one hand, you could say that's a doubling of the risk—gee, that sounds pretty substantial," says University of Nebraska sociologist Paul Amato, one of many social scientists who bolster Coontz's argument. "Another way of looking at it is that 80 percent are not exhibiting behavior problems."
Yet, even those numbers deserve closer scrutiny. Politicians and the press who repeat research data like this commonly ignore a fundamental rule of statistics: Correlation does not equal causation. Just as there are factors explaining why African-Americans commit crimes in disproportionate numbers that have nothing to do with their black skin, so there are other dynamics other than the divorce at work in the fate of children of divorced parents. Kids can be traumatized by fighting between parents, for example—a situation that may (or may not) improve with divorce.
Another huge factor is economics—always a handy predictor of how kids turn out. It is true, as the anti-divorcers observe, that divorce often precipitates a devastating income drop for the custodial parent (usually the mom). Whether that is an argument to restrict divorce or to strengthen child support laws is debatable. However, social scientists believe that those who get divorced are more likely to have a low income before the divorce (although plenty of affluent folks also divorce), and that poverty-and-divorce statistics are further skewed because many studies lump together the children of divorced parents with those of unwed single moms, who are even more likely to be poor.
Divorce is also overly blamed by a tendency to cite divorce as the cause of problems that kids started having while their families were still intact. Johns Hopkins sociologist Andrew Cherlin is about to publish a study that attributes approximately half the discrepancy in behavioral problems between the children of one- and two-parent families to pre-divorce factors. "That tells us that there are a lot of troubled families out there whose problems are invisible to us until they divorce," he says.
More dramatically, when the nonprofit Rand Institute on Education and Training carefully analyzed national school data to take into account family income, parents' level of education, the mother's age at the time of giving birth, and other factors, it concluded that "single-parent status by itself was not significant" in affecting school performance.
The absurdity of reducing statistics to crude correlations can be seen even more clearly as it occurs in a related and burgeoning campaign against living together before marriage. The Bellevue-based Washington Family Council, headed by Jack Kemp's son Jeff, recently brought a speaker to town—"Marriage Savers" guru Michael McManus—who warned that couples who live together before marriage have a 50 percent higher chance of ultimately divorcing.
The man upon whose research that statement is based, University of Wisconsin sociologist Larry Bumpass, explains the obvious: "People who hold very conservative views are less likely to enter into a co-habiting relationship" in the first place, and in turn are "less likely to divorce given a certain level of unhappiness."
Social scientists have long distanced themselves from the family values crusaders who skew their work. Despite her gloomy views on divorce, even Judith Wallerstein is opposed to returning to a fault system that would force parents to accuse each other of the worst in order to get a divorce. Like many in her field, she believes that the best thing we could do for kids is to "help the parents more" when a divorce occurs.
Of course, a sensationalistic reading of statistics often suits politicians and the press. But give Sen. Oke credit. By late last week, he had left a message on Stephanie Coontz's voice mail, wanting to talk.
Barbara Dafoe Whitehead's piece in the Atlantic
Sen. Oke's covenant marriage bill
Sen. Stevens' bill repealing no-fault divorce
The Covenant Keepers
The ACLU on covenant marriages