John Leo

Last week's 2000 census report on families also generated some odd and misleading journalism. By framing statistics in terms of the total number of households, it produced off-kilter headlines like the one in The New York Times: "For First Time, Nuclear Families Drop Below 25 Percent of Households." Another newspaper headline said: "Traditional Families Fading as Single-Person Households Gain." The not-so-subliminal message here was that traditional families are dwindling away and perhaps disappearing (though 62 percent of all kids are still in them). This was an impression generated by the Census Bureau's use of total households as a yardstick.

Even if the number of nuclear families were rising, they would likely account for a shrinking percentage of households. Americans live longer and marry later, so they live alone more in youth and old age. More households are created by longevity and affluence. But this has nothing to do with the status of nuclear families.

The use of household stats to make nuclear families seem anachronistic and irrelevant is an old story in the 30-year war over the family. A decade ago, people who wanted to downgrade traditional homes were saying that "Ozzie and Harriet" families were only 10 percent of U.S. households. One trick to get the percentage down was to count the family as nontraditional if mom had any job at all in the workforce, even just a couple of hours a

week. Empty-nesters and newlyweds were nontraditional, too.

Ex-Congresswoman Patricia Schroeder forced the number down to 7 percent by defining "Ozzie and Harriet" families as those with exactly two children -- families with one or three kids were nontraditional. Blankenhorn charges that last week's census report is just a "slightly more sophisticated version" of Patricia Schroeder's cooked numbers.

Is the Census Bureau playing a Schroeder-like political game with family numbers? Could be. Congress should ask. The bureau certainly played politics on racial and ethnic numbers. Treating Hispanics as if they were a nonwhite race made it seem as if white Americans were well on their way to becoming a minority in America. But half of Hispanics consider themselves white. Harvard sociologist Orlando Patterson accused the bureau of "quietly abetting the process of demoting and removing white Hispanics" from the white majority to make it seem smaller. Now the bureau is arranging numbers that make the percentage of traditional families seem smaller.

We need to know who makes these decisions at the Census Bureau and why. We also should know why the bureau keeps compiling data in a way that obscures family trends. What is the trend line for the percentage of children living with married parents? Thanks to Blankenhorn's private arithmetic, we know this crucial number for 1991-1996, but not after. Maybe Congress can get the Bureau interested in giving us the facts we need and staying out of politics.

John Leo

John Leo is editor of and a former contributing editor at U.S. News and World Report.

Be the first to read John Leo's column. Sign up today and receive delivered each morning to your inbox.