It took the media a while to acknowledge that most of Katrina's victims were black. Apparently, it will take longer to mention that most of the victims were women and children. I noticed three commentators who brought up the delicate subject of the mostly missing males--George Will, Gary Bauer, and Thomas Bray, a columnist for the Detroit News. Will noted that 76 percent of births to Louisiana's African-Americans are to unmarried women, and probably more than 80 percent in New Orleans, since that is the usual estimate in other inner cities. Will wrote: "That translates into a large and constantly renewed cohort of lightly parented adolescent males, and that translates into chaos, in neighborhoods and schools, come rain or come shine."

        A good deal of hard evidence shows that this is so. Two decades of research produced a consensus among social scientists of both left and right that family structure has a serious impact on children, even when controlling for income, race, and other variables. In other words, we are not talking about a problem of race but about a problem of family formation or, rather, the lack of it. The best outcomes for children--whether in academic performance, avoidance of crime and drugs, or financial and economic success--are almost invariably produced by married biological parents. The worst results are by never-married women.

        High crime. In a policy brief released last week, the Washington-based Institute for Marriage and Public Policy, http://www.marriagedebate.com/pdf/imapp.crimefamstructure.pdf, looked at 23 recent studies dealing with family structure and youth crime. In 19 of the 20 studies that found family structure to have an effect, children from nonintact or single-parent families had a higher rate of crime or delinquency. Neighborhoods with lots of out-of-wedlock births have lots of crime. Ominously, one study said that the more single-parent families there were in a neighborhood, the more crime there was among two-parent kids living around them. Again, these studies are controlled for race.

        Among the other findings:
        ·Adolescents in single-parent families were almost twice as likely to have pulled a knife or a gun on someone in the past year. This was after controlling for many demographic variables, including race, gender, age, household income, and educational level of parents.

        · In a large sample of students in 315 classrooms in 11 cities, the "single most important variable" in gang involvement was found to be family structure. In other words, the greater the number of parents at home, the lower the level of gang involvement. A study of American Indian families found that living in a two-parent family reduced gang involvement by more than 50 percent.

        ·Another study concluded that out-of-wedlock childbearing had a large effect on the rate of arrests for murder, an effect that "seems to have gotten stronger over time."

        · "Adolescents in married, two-biological-parent families generally fare better than children in any of the family types examined here," one study reported. The other family types studied were single mother, cohabiting stepfather, and married stepfather families.

        · One study, judged most important by the institute, found that divorce rates had no relationship to violent crime rates but that out-of-wedlock births had a strong relationship to youth crime--nearly 90 percent of the increase in violent crime between 1973 and 1995 was accounted for by the rise in out-of-wedlock births.

        The upshot of these studies is that America is confronted by a form of poverty that money alone can't cure. Many of us think social breakdown is a result of racism and poverty. Yes, they are factors, but study after study shows that alterations in norms and values are at the heart of economic and behavioral troubles. That's why so much research boils down to the old rule: If you want to avoid poverty, finish high school, don't have kids in your teens, and get married.

        But the conventional wisdom is determined to ignore the evidence. It holds that family fragmentation--sorry, diverse family forms--is positive and here to stay. Peggy Drexler, the author of a new book, Raising Boys Without Men, says people who promote intact families are playing a "blame game" against single mothers. She thinks eating dinner regularly with your children is more important than the number or gender of adults in the home. And boys, according to Drexler, have an innate ability to become men, even without a man in the house. (But if boys can raise themselves, why should any father stick around?) The book carries blurbs from various establishment figures. Why not? Her ideas are ordinary ones among our elites.