One of these studies (the Growing Up in Poverty Project, headed up by Professor Bruce Fuller of the University of California at Berkeley) followed 948 mothers with young children entering the welfare system California, Florida and Connecticut for two to four years. In Connecticut, scholars were even able to randomly assign some mothers to a jobs program vs. the traditional welfare check.

What did they find? The news is unambiguously good.

In California and Florida, employment rates for welfare moms climbed from 22 percent to 53 percent. In Connecticut, 69 percent of moms randomly assigned to the reform program were employed three years later, compared to 58 percent of the control group.

Earnings rose. Earnings for workfare mothers in Florida who worked rose from $5.45 an hour to almost $8 an hour. In California and Connecticut, working welfare moms earned more than $9 an hour.

More work at higher pay translated into higher incomes. Overall, the average income of welfare moms in Florida and California rose by $275 a month in just two years. Why? People who work continuously get pay raises. People who stay on welfare do not. In Connecticut, welfare reform moms' income increased $135 a month compared to old-system welfare moms, even after accounting for any losses in aid. Ninety-two percent of welfare reform moms had health insurance for their children (including Medicaid), compared to 81 percent of old-system welfare moms.

Did the children suffer? No. Measures of maternal affection, discipline, depression and child behavior were unaffected by welfare reform. Despite the dire predictions, the results of welfare reform are in: more work, higher earnings, lower dependency, no increase in child problems. Good news, right?

You would never know it, however, from the story by Nina Bernstein splashed on the front page of The New York Times. A Yale University press release reported on this good news as follows: "The slight economic gains felt by millions of single mothers -- who have moved off welfare and into low-wage jobs -- have not discernibly improved the living conditions of families or the daily lives of young children, according to a report to be released in Washington today." The New York Times gloomily reports: "New research findings in two states show that the stricter work requirements of contemporary welfare policy significantly reduce the chances that a single mother will wed."

Should we be concerned? Yes, of course. If there are ways to deliver welfare that are more marriage-friendly, we need to know about it. It is quite possible that, by ignoring marriage altogether, the current welfare-to-work programs inadvertently discourage it. It is also possible that work programs that aid only low-income single mothers, and ignore low-income married fathers, may help destabilize poor families.

But the larger good news, impossible to disguise, is that since welfare reform, the trends in family formation have, for the first time, begun to reverse. Unwed childbearing stopped rising. Marriage rates in low-income parents rose, especially among African-Americans. Previous economic booms produced no such family turnaround. Looking (as this study does) only at single moms of preschool children who enter welfare ignores the biggest beneficiaries of welfare reform: women (and men) who because of welfare reform decided to postpone pregnancy, or to marry, or to avoid welfare altogether.

Can we make welfare more marriage-friendly? Yes. Should we try? Of course. The good news is that by highlighting the alleged bad news about marriage rates, even The New York Times is now conceding that marriage matters.