How Americans became convinced divorce is bad for kids’ mental health. – Slate

Welcome to State of Mind, a new section from Slate and Arizona State University dedicated to exploring mental health. Follow us on Twitter.

Many Americans think it’s a shame that only 70 percent of children in the U.S. live in one home with two parents and under 50 percent live with two biological, married parents. They see divorce and single parenthood as normal yet still unfortunate, since accepted wisdom says children suffer long-term, irreparable harm when their parents live separately. On social media, that looks like everything from an Isla Fisher meme with the quote “You can’t underestimate how traumatic divorce is for the children,” to Catholic influencer Leila Miller writing “Divorce is simply a bloodless form of child sacrifice.”

But a careful review of academic research—and the historical and cultural context in which it was conducted—calls this shared understanding into question. Most of the problems associated with being a child of divorce are instead related to sexism, racism, homophobia, shoddy recordkeeping, and insufficient government support.

Divorce was rare before the American Revolution. For those who were not enslaved, about one-third of children lived with a stepparent or single parent. But that owed to mortality rates; marriages just weren’t often dissolved. When they were, rights to the children belonged to their father. In the business section, next to ads for horse rentals, colonial newspapers published “runaway wife” reports. Illustrating just how commercial an undertaking marriage was in early white America, these husbands didn’t plead for their erstwhile partners’ return; they just declared themselves off the hook for debts the women incurred. But after America divorced King George III, attitudes toward personal independence and contractual obedience shifted. By the 19th century, suffragists argued that not only would an expanded right to exit marriage permit women to escape bad ones, but the threat of separation would allow them to renegotiate the terms of good ones. Men also increasingly felt entitled to love matches and fulfillment. Around the time frontier mentality took hold, runaway husband ads became more common than runaway wife ones. The “tender years” doctrine assumed children needed to be with their mothers, reversing the custodial default. By 1970, California had authorized no-fault divorce, and most states followed suit within a decade.

This liberalization of divorce during the second half of the 20th century occurred against a backdrop of increasing female employment and independence. In the attendant cultural battle, some people directly defended patriarchal gender roles ordained by nature and God. But many instead made children their rallying call. The 1965 Moynihan Report—a document written by Assistant Secretary of Labor Daniel Patrick Moynihan and his staff to convince the federal government to bolster civil rights legislation with policies addressing economic inequality—decried the “crumbling” of family structure in Black communities. Divorce rates, which were even across race until 1960, had indeed risen more steeply and earlier among Black women. But the report’s emphasis on family structure gave conservatives an excuse to ignore the structural forces behind that stat and insist on self-help rather than actual help. On the campaign trail in 1976, Ronald Reagan held up the straw man of the “welfare queen,” a Black single mother gaming the system. This was the sexist, racist narrative of the declining family that the conservative “family values” movement of the 1990s and 2000s carried forward. Pro-marriage PR campaigns—run by organizations like the Institute for American Values and the Promise Keepers—were also deeply homophobic. They promoted heterosexual marriage with funding from conservative groups like the American Enterprise Institute and the Heritage Foundation. (This connection continues: Leila Miller wrote an April 2022 article titled “The Ten Sins of Gay Surrogacy.”)

These groups and other members of the “marriage movement” wielded as fact the findings of Judith S. Wallerstein, who began following 131 children from 60 divorced families in Marin County, California, in 1971. She checked in with this same group of kids every five years for 25 years. More frequently, she checked in with the public, spreading her dire conclusions about divorce’s long-term harm to children via the Today show, Good Morning America, The Oprah Winfrey Show, and more. Another psychologist, E. Mavis Hetherington, also published findings throughout the ’80s and ’90s. Her studies, which included 1,400 families, showed that the vast majority of kids of divorce became well-adjusted adults. And in 1994, The Good Divorce made a big splash. In that book, Constance Ahrons argued that kids thrive when divorced parents behave like well-functioning married couples.

Despite Wallerstein’s small sample size, lack of a control group, and selection effect (parents signed their kids up after seeking therapy), her “the children suffer” narrative won out. The media referenced Wallerstein somewhere between 6 and 14 times more often than Hetherington, even though fellow social scientists were more likely to cite the latter.

Her arguments appealed to the general public in part because there were incentives for some people to cast children as victims. Fathers’ rights groups, such as the National Fatherhood Initiative, spread horror stories of ex-wives who alienated dads from their children or fabricated tales of abuse as vengeance (“Uncle Dad” and “Medea” stories, respectively). Mothers embroiled in court cases told of midlife-crisis abandonment, deadbeat dads, and children who were, indeed, harmed by their parent. These worst-case scenarios entered the American ethos in films. Kramer vs. Kramer (1979) encouraged audiences to empathize with the plight of devoted fathers over selfish working mothers, and The War of the Roses (1989) portrayed destructive and protracted conflict as inevitable. Fathers’ rights advocacy made an impact off the screen, too, with custody law shifting again. This time, state statutes increasingly established a preference for frequent and continuing contact with both parents.

The “children will suffer” beat, produced by these different drums, had incredible staying power. An analysis of popular press articles published from 1968 to 2005 revealed an increasingly negative take on divorce. And the federal government began spending hundreds of millions on the Healthy Marriage Initiative in 2001.

But scientists immediately questioned Wallerstein’s study and other early ones that looked at kids who didn’t live with two married parents—and they found a slew of problems.

In a 2003 paper, Paul Amato of Penn State created an index of overall psychological well-being and concluded that the overlap between adults with and without divorced parents was 90 percent, meaning only about 10 percent of those with divorced parents had more mental health issues and reported lower life satisfaction and happiness than those with married ones. Almost half, 42 percent, exceeded the average well-being score of the married parents sample.

For comparison, this effect size is less than one-third that of parenting classes on child behavior. Statistics on teen dating and popularity also lend perspective. Partnering in adolescence tracks with poorer academics; having high status as a teen has been linked to lower-quality relationships; and both are tied to more substance use. But these outcomes don’t incite parental anxiety or shame the way such marginally increased risk does in the context of divorce. No one tells parents they must keep their kids single in the same way Isla Fisher told the Evening Standard, “I’d urge parents to strongly consider working things out. I’d work things out and I’d definitely stay put.” Or the way Miller wrote to adults considering divorce, “I pray that you will slam on the brakes and consider any other course than the one you are on.”

There’s also a time dimension in the research. Divorce is usually accompanied by disruption to routines and emotional stress. “I would never want to minimize how difficult it is for children to go through that transition,” Ahrons said in an interview with us before her 2021 death, “but it doesn’t mean that they are damaged.” Grief and anxiety fade. Behavior tends to return to normal within two years. A study of 23-year-olds found that 11 percent of those whose parents had divorced suffered from mental health problems, compared with 8 percent of those with married parents. Yet humans underestimate resilience thanks to a bias dubbed “immune neglect.” In reality, researchers have concluded that most children touched by divorce are ultimately well-adjusted, indistinguishable from peers raised in nuclear families. It’s hard to overestimate the impact this information could have on a person hoping to divorce—or ruminating on how their breakup has affected their children. And women initiate more divorces than men do.

It’s also possible even the small negative effect that’s seen along with divorce isn’t caused by divorce. Texan kids who live in homes with pools go to college at higher rates, but it’s not because swimming makes them smarter; their parents’ wealth independently drives both phenomena. Similarly, children whose parents divorce are more likely to have lived in a high-conflict home prior to the breakup. If a child ends up struggling with, say, substance use, the stress of that conflict would be behind it, not the divorce. In support of this theory, a 1999 study found that children whose high-conflict parents don’t split experience even greater behavioral problems, and a 2004 paper showed positive effects of ending high-conflict marriages. A 2014 study found no effect of family structure on psychopathology after controlling for conflict.

There’s more to a critical look at causality. For example, kids whose parents don’t live together are more likely to experience poverty, both because those with less education are less likely to marry and because splitting the income that once supported one household means a lower budget for both new ones. For those kids, researchers theorized, divorce isn’t driving bad outcomes; poverty is. Indeed, a 2015 study found a higher likelihood of behavior problems when post-divorce household income was lower (although a critical look at this metric is in order since Black and brown children tend to be written up more often due to bias). This result confirmed a 2013 finding out of the Netherlands that negative effects are reduced when kids have more educated mothers, and that children who were poor prior to divorce suffered more. In one of the most impressive studies of child development ever, the Australian Temperament Project, researchers were surprised to discover no significant differences between 17- and 18-year-olds with married and divorced parents in behavioral or emotional adjustment, academic outcomes, or social competence. They pointed to Australia’s social security benefits and enforcement of child support.

Then there’s genetics to consider. Divorce is more common among people who display aggression, mental health problems, and antisocial behavior. Their children, as genetic heirs, are more likely to score lower on psychosocial measures regardless of parental marital status. Professor Brian D’Onofrio has attempted to unravel this knot with studies on first-cousin children of twins, with mixed results. Substance use and early cohabitation appear to be more closely related to a genetic legacy that contributed to divorce rather than to the divorce itself, but the same isn’t true of other outcomes. And D’Onofrio acknowledges that his team did not completely control for either post-separation income change or pre-separation conflict.

Doing that is harder than it sounds, but many modern studies have found ways to take kids’ pre-divorce levels of things like depression, low self-esteem, and lackluster school performance as a starting point. A 2001 study found that such preexisting struggles almost entirely accounted for post-divorce differences. Other studies using a child or sibling “fixed-effects model” found mixed results (2005), no effect on achievement scores (2005), no effect on premarital births (2005), no effect on educational attainment (2006), and no effect on emotional well-being (2007). A 2019 study, after adjusting for pre-divorce factors, found that divorce moves kids from an 81 percent likelihood of graduating high school to a 78 percent likelihood. In other words, as studies got more sophisticated, already small differences got even smaller, and sometimes disappeared entirely.

Some researchers point to one more factor: lingering stigma, including terms like “broken home” and picture books and TV shows that overrepresent nuclear families. The message that divorce is rare or bad can create a self-fulfilling prophecy of distress. Authors of a 2019 study suggested that divorce isn’t as big a deal to kids who know many nonnuclear families. Whether children experience “an acute sense of deprivation” could be key to adjustment, they said. Authors of a 2020 study would seem to agree.

Inevitably, some children will fare better than others in the wake of a parental split. Individual kids can wind up traumatized. Many experience subclinical effects like painful memories, and many display above-average competence and resilience (think, smooth seas and skilled sailors). But we don’t have enough data to understand why divorce is such a mixed bag at both the population and individual levels. Under these circumstances, as Amato wrote in 2010, “researchers should focus … more attention on the factors that produce variability.”

For one thing, the vast bulk of the research omits an important variable: parenting quality. Research suggests high warmth and support, even from just one parent, is what children need to succeed.

Some academics have tried to apply what we know about the corrosive effects of conflict and uplifting impact of high-quality parenting (a so-called “counter-ACE”) in the context of divorce. Remember Ahrons’ book? Amato’s 2011 paper “Reconsidering the ‘Good Divorce’” tried to test her hypothesis by assigning children to cooperative co-parenting, parallel parenting, and single parenting groups. Though the researchers found that youth in the “good divorce” cluster exhibited fewer behavioral problems, that wasn’t the case for other bad outcomes. This study has been cited again and again as evidence that Ahrons was wrong and Wallerstein right. But its authors acknowledged they might not have been able to accurately identify “good divorces,” their sample was relatively small, parents today may behave differently than those in the ’90s (from whence the data came), and findings were affected by the time that elapsed between each divorce and the study. Not even they thought their work provided a definitive answer.

And significantly more definitive answers may now be impossible. In 1988, the Centers for Disease Control and Prevention stopped gathering complete data on children affected by divorce, and in 1996, the National Center for Health Statistics stopped supporting reporting of divorce statistics.

“How many American children are the focus of family court proceedings each year?” wrote professor Bob Emery in Family Court Review. “I have no idea. … We also don’t know, for example, how many children in the United States are living in joint physical custody.” The 2020 census didn’t allow families to select a 50-50 share, instead requiring parents to say where their children “usually live or stay.” Wisconsin keeps thorough records and a group of academics has tried to use them as national tea leaves, but no one can confidently say how many kids experience co-parenting, nor what high-quality co-parenting produces. There’s little financial support for that type of research, Ahrons told us: “The money comes if you are looking for problems.”

Some have joined the call for more research, yet efforts to identify best practices for kids aren’t just hamstrung by limited data and low funding. These days, Emery said, divorce research is considered outmoded by psychologists: been there, done that. You may have noticed that we haven’t been able to cite any research to support the idea that the majority of us still mistakenly believe divorce commonly does irreparable harm to kids. That’s because the question isn’t asked by nonpartisan national surveys anymore. And yet parent after parent tells us they think Wallerstein is right. In other words, while the culture warriors and academy have now largely moved on from divorce, parental apprehension and guilt persists. And it isn’t helped any by a cultural trend toward intensive mothering and Instagram’s #tradwife.

This scientific evidence accumulated over the last three decades with no single study monumental enough to garner significant press coverage. That has meant divergent narratives could coexist, and the one influenced by sexism, racism, homophobia, and other types of fear prevailed. Wallerstein’s take has proved indelible even though she was proved wrong. As a result of the way the Christian right was able to frame—and effectively close—the policy debate, national solutions have focused on individuals’ decisions and bolstering the institution of marriage: Choose the right spouse. Go to couples therapy. All but ignored is the government’s opportunity and obligation to families. And that disproportionately affects women, Black families, and lower-income kids and caregivers.

If national policy were based on research rather than zealotry, we would invest in children’s well-being by promoting research-backed strategies for parenting and co-parenting, supporting caregivers financially and systemically, underwriting research on nonnuclear best practices, and dispelling myths about the “brokenness” of families and children that can become feedback loops. Other ideas supported by the data include subsidized therapy and legal reforms that make the process less onerous for everyone.

“I’m still singing the old song,” Ahrons said. “Divorced families can be healthy after divorce. There is still an enormous amount of resistance to that.” She said mechanisms for collaborative divorce are a start, and she found optimism in some parents’ behavior as well. But still, “that baseline assumption about divorce and the results of divorce is still being perpetuated, and if not perpetuated, at least not disrupted.”

State of Mind is a partnership of Slate and Arizona State University that offers a practical look at our mental health system—and how to make it better.

Favorite