Bad Signs

Each Tuesday, in its “Health and Fitness” section, the New York Times features a round-up of current health-related research findings called “Vital Signs,” and therein hangs a tale. “Good News for the Children of Divorce” announced a “Vital Signs” headline in August of this year: “The divorce rate for adult children of divorced parents has declined over the past two decades, a new study has found.” Wow! Such a finding clearly suggests that, for children, divorce is becoming less harmful. Specifically, a lower divorce rate for the adult children of divorce would indicate a welcome reduction in the intergenerational transmission of divorce. 

The only problem is, the story is bogus. As anyone familiar with this issue knows, the divorce rate for the adult children of divorce has not declined. It has increased significantly. In the early 1970s, about 35 percent of ever-married adult children of divorce were themselves divorced. By the early 1990s, the number had increased to 45 percent. Yet during this same period, the divorce rate for everyone else increased even more. In the early 1970s, about 18 percent of ever-married adults raised in intact homes had themselves divorced. Twenty years later, the number had jumped to 35 percent. 

The study cited in “Vital Signs,” which was conducted by Nicholas Wolfinger of the University of Utah, does show that the gap separating the two groups has narrowed. But this development has nothing to do with declining divorce rates. On the contrary, the convergence cited in Wolfinger’s study  is entirely the result of a remarkable increase in divorce-proneness in recent decades of U.S. adults who were raised in intact families. This is good news? 

Actually, it’s terrible news - not only because divorce has increased, but also because the convergence described by Wolfinger suggests that, in a high-divorce society, everyone’s marriage is made weaker. Today, divorce and its consequences are everywhere, affecting even those who were raised in intact families. One result is a general, across-the-board weakening of the ideal of marital permanence. As Norval Glenn of the University of Texas presciently suggested more than a decade ago, in such a society we would expect to see a gradual convergence of divorce rates, at increasingly higher levels, among various subpopulations. And that is exactly what we are seeing, even as we are reassured by the New York Times that the entire phenomenon amounts to “Good News for the Children of Divorce.”

“For Single Parents, One Less Worry” announced another “Vital Signs” headline in October of this year: A new study finds that “in and of itself, single parenthood seems to have no effect on how a child does in school.” Wow! Such a finding is significant because it directly challenges the work of many leading researchers, most of whom have consistently found that growing up with a single parent is a significant risk factor regarding the child’s educational achievement.

The problem is, the story is worthless. The researcher being cited, Henry N. Ricciuti from Cornell University, examined an extremely limited range of outcomes - vocabulary and math test results and mother’s reports of behavioral problems - affecting the very young children (age 6-7) of mostly young, lower-income, poorly educated mothers. These are severe, important restrictions. The age limitation is especially problematic, since many of the ill effects of growing up in a father-absent home only show symptoms (especially symptoms clear enough to be detected by crude social science instruments) after the child’s seventh or eighth birthday. Even if the study were competent, it could provide no basis for the global assertion that “single parenthood seems to have no effect on how a child does in school.”

But the study is not competent. Ricciuti crumbs his entire inquiry by playing fast and loose with the definition of “single parent.” If the unmarried mother has a live-in boyfriend at the time the questionnaire was administered, Ricciuti counts it as a two-parent home. That’s quite a trick. Since many of these young mothers have boyfriends, and since many of the problems associated with unwed parenthood are made worse, not better, by the presence of boyfriends in the home, Ricciuti can, in one fell swoop, transfer much of the bad stuff that would have been in column A over to column B.

That’s quite enough to invalidate his study, but Ricciuti is also a devoted practitioner of what I described in an earlier letter (Spring 1998) as the “per se” argument. For Ricciuti, the issue is never single parenthood; it’s always single parenthood “per se.” Here is how he puts it: The results [of this study] do suggest that in the presence of maternal and household and family characteristics favorable to young children’s development, single parenthood in and of itself need not represent a risk factor that by definition implies negative developmental consequences for children.” That sentence deserves a prize. Let’s translate it. If the mother has adequate financial resources, and if the mother has time to spend with her child, and if the mother has other family and community resources that are “favorable” to her child, then the mere fact that the child’s father is absent from the home does not necessarily, in every case, by definition, harm the child. In short, if we pretend that the child has all the advantages that typically accrue from having a father in the home, then it doesn’t matter (necessarily, in all cases) whether or not the father is in the home. Well, what a relief. Or as the New York Times puts it: “For Single Parents, One Less Worry.”

Articles like these have consequences. Our friends over at the Council on Contemporary Families, whose main scholarly mission is defending the divorce revolution, will probably be citing this stuff for years. Divorce is becoming less harmful for children. Single parenthood has no effect on how children do in school. Don’t you remember? It was reported in the New York Times.

Norval Glenn (University of Texas) replies: “The Ricciuti study is in some respects even worse than you say it is. You are of course correct in pointing out that live-in boyfriends shouldn’t be considered to be ‘parents.’ Furthermore, the study should have distinguished between step-families and families including both biological parents, because there is evidence that the effects of the two kinds of families on some child outcomes are typically different. Even more important, as Ricciuti admits, just looking at family type at age six or seven, as was done in this study, is not adequate. It is family history that is important. Consider that a child who lived in a single-parent family for six years, but whose mother recently married, is classified by Ricciuti as living in a two-parent home, while a child who lived for six years with both biological parents who recently divorced is classified as living in a single-parent home. These deficiencies in the definition and measurement of the independent variable make the findings of the study essentially meaningless.” 

Sources: “Good News for the Children of Divorce,” New York Times, August 17, 1999. Data on divorce rates of ever-married children of divorce and ever-married children of intact homes from Norval Glenn, University of Texas. Nicholas H. Wolfinger, “Trends in the Intergenerational Transmission of Divorce, Demography 36, no. 3 (August 1999). Wolfinger, “Coupling and Uncoupling: Changing Marital Patterns and the Intergenerational Transmission of Divorce,” paper presented at the annual meeting of the American Sociological Association (Chicago, August 1999). Norval D. Glenn and Kathryn B. Kramer, “The Marriages and Divorces of the Children of Divorce,” Journal of Marriage and the Family 49 (November 1987). “For Single Parents, One Less Worry,” New York Times, October 5, 1999. Henry N. Ricciuti, “Single Parenthood and School Readiness in White, Black, and Hispanic 6- and 7-Year-Olds,” Journal of Family Psychology 13, no. 3 (September 1999).

First published Fall 1999.