NEW YORK — The commencement speeches are done, the confetti has been swept up, the decorations stored away. But even as graduation 2003 fades into memory, an unanswered question continues to trouble the US education community: When it comes to high school graduation rates, should the nation be celebrating, or looking more deeply into some worrisome trends?
A recent study by the Child Trends DataBank in Washington offers interesting figures on the nation's high school dropout rates. Overall, US high school dropout rates have fallen, the study indicates.
In 2001, 11 percent of people aged 16 to 24 had dropped out of high school, compared with 15 percent in 1972. Graduation rates compiled by the federal government also show high school graduation has been steadily on the rise for decades. But broken down by race, the Child Trends figures tell a more nuanced story.
The numbers demonstrate that the dropout rate for black students has been cut almost in half over the past 20 years, having fallen to 11 percent in 2001, compared with 21 percent in 1972. But at the same time, the study also shows a stubborn trend for Hispanic student dropout rates. Although down from a peak of 34 percent in 1972, this rate remains extremely high, at 27 percent, in 2001.
On the surface, it would appear that black students are connecting to high school more firmly than in the past, while Hispanic students, for whatever reason, are not.
But it's essential to look more deeply into such numbers, experts say. Ironically, the figures on black students actually conceal a troubling deficit, some insist, while the news about Hispanic students is surprisingly positive.
"There absolutely was a fall in the high school dropout rate" of black students, confirms Bruce Western, professor of sociology at Princeton University in New Jersey. That's the good news. But the bad news follows quickly behind. "About half the fall in the dropout rate is due to the rise in imprisonment of young black males," he says.
Prison inmates generally are excluded in both graduation and dropout counts. At any point in time that practice can invalidate the accuracy of both figures, but the distortion is particularly dramatic at this point in time.
Incarceration rates for black men aged 22 to 30 who haven't finished high school leaped to 40 percent in 1999 from only 14 percent in 1980 - a stunning acceleration unparalleled in US history, Professor Western says.
That doesn't mean that there has been no improvement in dropout rates for black students. "From the mid-1980s to present it's reasonably clear that the dropout rate has declined among blacks," says Robert Hauser, professor of sociology at the University of Wisconsin in Madison.
Professor Hauser uses his own numbers, which show dropout rates for black students declining from a high of 20 percent in the mid-1980s to about 15 percent in the early 1990s, and remaining more or less flat ever since.
But that decline, he says, has little to do with anything schools have or have not done and much more to do with changes within black families.
"There have been increases in parental education and decreases in the number of siblings," he says. "In the mid-1970s and late 1980s there was a big change in the number of children, while at the same time the education level of the parents just kind of marches up."
In the early 1980s black parents had completed an average of 11 years of education, while by the late 1990s that figure was up to about 12 years.
That's incontestably good news for the black community, sociologists agree, but it balances very uncomfortably against the shocking rise in incarceration rates.
Meanwhile, the news about dropout rates may be far better for the Hispanic community than it appears on the surface.
While many studies confirm that the number of Hispanic students who drop out remains high, most fail to take into account the record-high numbers of Spanish-speaking immigrants pouring into the US over the past decade or so.
"You have to make a distinction between Hispanic youth born in the US and spending their whole careers in US schools versus those who came later in life," says Richard Fry, senior research associate for the Pew Hispanic Center in Washington.
The dropout rate for Hispanic youths born in the US has actually shown a small but significant decline over the 1990s, Dr. Fry says, from 15 percent to 14 percent. "The news is not astoundingly positive, but it's positive," he says. Considering the challenges Hispanic students face in the US, "it's encouraging that dropout rates have fallen a little bit."
But overall dropout rates remain a problem that simply doesn't get the attention it deserves, says Pedro Noguera, professor of sociology at the Harvard Graduate School of Education in Cambridge, Mass.
Hispanic students obviously face a particular academic challenge if English is not their first language - particularly, points out Professor Noguera, those who may not be literate in Spanish, either. But, in an era in US public schools that tends to focus on high-stakes testing and that continues to place higher demands on students, "it's an issue that's not even on the radar for a lot of policy-makers," Noguera says.
But at the same time, he adds, sometimes focusing on dropout rates for minority students minimizes the ongoing problems faced by poor white children as well.
"We tend to generalize about groups, and there's a tendency to render poor white kids invisible," he says.
There is also a tendency to overlook the huge number of young men - particularly young black men - in prison without high school diplomas, Western says.
What many Americans forget when reading headlines about soaring prison populations is that the vast majority of these prisoners will eventually leave jail and reenter society - without the benefit of enough education to find even an entry-level job.
Ironically, today's more crowded prisons offer inmates fewer educational opportunities than in past decades. "Resources for prison programming, including educational programs, have all contracted as the prison population has gotten larger and larger," Western says.
The best time to focus on breaking the link between low levels of educational attainment and crime is long before either high school graduation or prison, most educators say.
That's why Baldwin-Wallace College in Berea, Ohio (about 15 miles from downtown Cleveland), decided to tackle this particular problem in its own backyard.
According to city figures, about 50 percent of students in Cleveland do not make it from ninth to 10th grade. And the group most clearly at risk, says Denise Reading, vice president of student affairs at the college, is young urban males.
It's hard to look at Cleveland schools, she says, and not ask the question: "Where are our young men, particularly our young men of color?"
This summer, Baldwin-Wallace launched the Barbara Byrd-Bennetts Scholars Program. The program will bring 34 Cleveland eighth-graders on campus for two weeks this summer to sharpen English and math skills.
Throughout the school year, the boys will continue to receive mentoring and tutorial support in addition to participating in on-campus residential programs over the next four years - a level of involvement that will hopefully support them all the way up to high school graduation.
"We know how to tackle" the problem of dropouts, Dr. Reading says. "The question is, Are we willing to do what we know will work?"
One of the hardest things about tracking high school graduation rates is determining which numbers to use.
Statistics often vary dramatically. For instance, numbers compiled by the federal government show the national high school graduation rate for 2000 to be 86.5 percent, while Jay Greene, senior fellow at the Manhattan Institute in New York, pegs the same figure at 69 percent.
The government includes in its graduation rate students who attain a high school equivalency diploma, while Mr. Greene does not. At the same time, Greene counts students who land in prison as nongraduates, while the government simply doesn't count them at all.
Dozens of other factors may skew numbers as well. If the dropout rate is calculated by comparing the senior class with the junior class, that figure fails to take into account all students who dropped out between ninth and eleventh grades. A comparison with ninth-graders might show a markedly different number.
Also, when calculating national figures, Census Bureau data are often used to determine how many 18-year-olds graduate. But those data may underrepresent the poorest populations - exactly those most likely to fail to graduate.
It's hard to agree on standards to be used in compiling such figures. The danger, however, is that groups determined to do so can use these numbers to make widely divergent points.