From one point of view, the U.S. Census Bureau offers a glowing report on America's goal of becoming an educated nation.
A record 81% of the population has completed high school or its equivalent, and the percentage appears to be headed still higher.
But another number tells of trouble on the horizon. Among the most recent crop of young adults, those now 18 to 24 years old, only 75% have finished high school, suggesting a substantial dropout rate.
Therein lies a mystery: Why is the national graduation rate climbing while so many school-age children are dropping out?
The question evokes an array of answers from demographers, statisticians and educators. Some think immigration of less-educated young people is skewing the numbers. Others point to "redshirt" students who spend an extra year in high school, then graduate at age 18 or 19. They also cite "second-chance" dropouts who earn degrees in their later years by passing an exam.
That the Census Bureau seems to be pointing in both positive and negative directions is not just a matter of academic curiosity.
Critics contend that the government is inflating the nation's educational achievement by counting those who pass equivalency exams as essentially the same as those who graduate from high school. They say equivalency certificates are pseudo diplomas with little value.
But the one factor that probably contributes most to the conundrum stirs hardly any mention--possibly because it is a little morbid.
Demographically speaking, the graduation rate is almost guaranteed to climb year after year as long as people born before 1940 continue dying.
The oldest generations have the lowest rate of high school completion in the nation. In 1940, only a fourth of the adult population had at least 12 years of schooling. Since then, the rate has climbed steadily by an average of almost 10 percentage points a decade, according to the Census Bureau's Current Population Survey, which takes an annual look at educational attainment.
For many years it was the young who drove that machine-like pace; they completed high school in ever-increasing numbers. But that changed in the mid-1970s, when the graduation rate for ages 25 to 29 hit 84%. After that, the gains slowed dramatically. The rate had inched up only to 88% in 2000.
But just when youth faltered, America's oldest and least-educated generations began dying.
"The completion rate of your total population [age] 25 and over will continue to increase, as those coming in have higher levels than those going out," said Jennifer Day, chief of the Census Bureau's education and social stratification branch.
Day estimated that the trend could hold for another census or two. At that point, the rates for those coming in and those going out would converge, with both groups' graduation rates reaching about 90%.
Significant as that achievement might be, it would fall two decades past the 2000 target year set in 1990 by former President George Bush's panel on national goals.
And that's assuming the incoming groups don't backslide. Based on the numbers in the recently released Census 2000 Supplementary Survey, that expectation looks a little shaky. That survey shows that, if trends among the youngest group of high school dropouts hold steady, national rates in the next two decades could get stuck where they were in the 1970s.
Some demographers are unfazed by the poor numbers for the 18- to 24-year-olds, even though those rates are the most current. For one thing, 1.1 million 18-year-olds are still in high school, said Vance Grant, a statistician at the U.S. Department of Education. Assuming most of them obtain diplomas, the completion rate won't look quite so bad.
New immigrants with little or no schooling are also part of the picture. Some of them will earn high school diplomas through adult school or obtain equivalency degrees.
For those reasons, the Census Bureau monitors a later age group, 25 to 29, as the bellwether for American educational attainment. By that measure, the goal set by former President Bush's committee is in range.
"We have close to 90% of students graduating from high school or having an equivalent by the time they are 25," Day said.
But there are plenty of skeptics who dismiss that number as practically meaningless.
For one thing, the Census Bureau's survey does not include the incarcerated. By ignoring the nation's 2 million inmates, the survey omits a group that is disproportionately made up of dropouts, said Gary Orfield, professor of education and social policy at Harvard University.
Still more error is presumed to creep into the study because no attempt is made to verify the survey answers.
"The big question is whether people are fibbing or not," said Alan Bonsteel, a physician and education activist who has done research on California's dropout rate.
All these squishy variables only serve to aggravate a more serious complaint about how the Census Bureau--and the nation--defines a basic education.
Detractors believe that by lumping the high school diploma into the same category with alternatives such as "general education" degrees, the government lends support to the idea that the easier path is just as good.
"They're doing a disservice by aggregating high school dropouts and high school graduates and calling them high school equivalency," said Thomas Mortenson, senior scholar at the Center for the Study of Opportunity in Higher Education in Washington, D.C.
Mortenson, who publishes a newsletter that analyzes educational data, contends that the "actual" high school graduation rate--minus equivalency degrees--has declined steadily in all but a few states. One exception is California, where the rate declined in the 1980s but turned the corner in 1994. Now at 68%, the rate is the highest it has been since the early 1980s, according to his calculations.
"I don't think we're anywhere near a 90% graduation rate in this country," Mortenson said. "I think it's probably 60% to 70%. Politicians can be smug that we're making some progress and things look pretty good. I don't think things look pretty good at all."
One problem obscured by the numbers, Mortenson says, is the economic cost paid by the third of general education degree holders who don't get the certificate until after age 25.
"If you wait till [your] late 20s, that means you have worked 10 years without a degree," Mortenson said. "That means you've been washing dishes or cleaning out [rooms] in a hotel. The job market teaches them the lesson: You have to have education or training to qualify for better-paying jobs."
Many educators regard the census survey, even with its limitations, as the most accurate measure of how many students are dropping out and from what groups.
The alternative to the census is to rely on school administrators' reports, which are widely perceived as unreliable.
"If they're allowed to define their own dropout rate, it looks like almost nobody is dropping out," Harvard's Orfield said.