At this Easter-Passover season, it comes as no surprise that the American people are worried about what they see as the decline in the moral character of the nation. Andrew Kohut, the Pew Research Center’s estimable pollster, reported last week that three-fourths of those he interviewed in March said they believe people are not as honest and moral as they used to be and that young people lack the same sense of right and wrong their counterparts had a half-century ago.

The wish for restoration of traditional values is widespread, spanning religious, racial and ethnic divisions.

So it was a shock to me and, Kohut says, “a real surprise” to him that when this same cross section was asked if they thought the government should develop programs to encourage people to get married and stay married, they gave an emphatic no. By a margin of 79 percent to 18 percent, they said they favored the government staying out of marriage promotion. Even among “highly committed” white evangelicals, Kohut found, the verdict was 60 percent to 35 percent against such programs.

That is odd, because many of the organizations that draw support from these same voters have been leading the cheers for the “marriage initiative” that was a spotlighted part of President Bush’s blueprint for the next phase of welfare reform. In the February announcement of that plan, the White House proposed to set aside up to $300 million for states to use in “innovative approaches to promoting healthy marriage and reducing out-of-wedlock births.” State welfare plans would have to include “explicit descriptions of their family formation and healthy marriage efforts, numerical performance goals and annual reports on achievements.”

The Family Research Council, with roots in the religious community, applauded the Bush initiative and urged that support for marriage promotion be increased to at least $1.5 billion annually over the next five years. “Marriage and stable families,” its statement said, “are the keys to preventing child poverty.”

That may seem obvious, but in fact, there is serious dispute in the policy community about the connection between marriage and the well-being of children.

Some facts are incontestable. In the current issue of The American Prospect, a left-of-center magazine, Theodora Ooms of the Center for Law and Social Policy cites some striking statistics:

“Children living with single mothers are five times as likely to be poor as those in two-parent families. Growing up in a single-parent family also roughly doubles the risk that a child will drop out of school, have difficulty finding a job or become a teen parent. … It’s not just the presence of two adults in the home that helps children, as some argue. Children living with cohabiting partners and in stepfamilies generally do less well than those living with both married biological parents.”

Given these facts, why shouldn’t the government become a marriage broker? The answers come in two parts.

At the practical level, social researchers say, there is little experience or evidence as to what works. And a sizable body of data suggests that the key point is not to get couples to marry but rather to be sure that pregnancy is postponed until after marriage. Unwed mothers have notably poor chances of marrying at all, and the marriages they make are notably less stable. “Ultimately,” says Daniel T. Lichter in a publication of the Democratic Leadership Council, “marriage promotion must begin by discouraging out-of-wedlock childbearing.”

But there is also a values question at the center of this debate. Kohut said he suspects that most of those who said no to government marriage-promotion programs in his survey just thought it was too intrusive a role for even a well-intentioned bureaucracy.

Ooms made the same point in her article. “Most people,” she wrote, “regard decisions to marry, divorce and bear children as intensely private. Any policy proposals that hint at coercing people to marry, reinforcing Victorian conceptions of gender roles, or limiting the right to end bad marriages are viewed as counter to American values of individual autonomy and privacy.”

Kohut’s poll seems to support her view. But Ken Connor, the head of the Family Research Council, challenges that interpretation. “There is a healthy skepticism about government involvement,” he told me, “but the president is really proposing only to make the government an enabler to provide an opportunity for effective programs to be leveraged by government funds. … I don’t think there will be much resistance to it.”

Maybe. But I wouldn’t bet the dowry money on it.

David Broder is a columnist for Washington Post Writers Group.