The New Math on Graduation Rates
Persistent failure to meet performance targets for graduating more students should carry serious consequences.
Last month, the triumphant chords of "Pomp and Circumstance" echoed in auditoriums across the country. But many who had entered high school four years ago were not joining their friends for the march across the stage. As it turns out, the national graduation rate is not the widely broadcast 85 percent. In our public schools, the correct figure is much closer to 68 percent. My recent Urban Institute research also raised another alarm: Minorities nationwide have little more than a 50-50 chance of earning a diploma. For years, high dropout rates have been treated as an isolated problem plaguing only certain troubled school systems. But when nearly one- third of all students are failing to graduate, it’s time to recheck the arithmetic.
How could we be so far off the mark when it comes to something as crucial and seemingly straightforward as counting high school diplomas? In truth, graduation rates have long been relegated to a dark, dusty corner of the educational statistics enterprise. Despite widespread consensus that completing high school is a social, economic, and educational necessity, little is done to carefully or uniformly measure who gets a diploma. Now that the age of educational accountability has dawned, change is afoot.
We have the No Child Left Behind Act to thank for the unpleasant discovery of low graduation rates. For the first time, this sweeping federal legislation requires states to hold their public high schools accountable for both achievement-test scores and graduation rates. The higher stakes for schools are just starting to illuminate the extent of our high-school- completion problems. But this national crisis has yet to be brought into full daylight.
Too many of us would just assume that the U.S. Department of Education calculates an annual high school graduation rate for the public schools. Not so. That math assignment has largely been a state responsibility, although one often shirked. Prior to the No Child Left Behind law, only a handful of states calculated these rates and held schools accountable for their performance.
The widely quoted 85 percent graduation rate is also misleading. That figure comes from a survey of a post-school-age population, includes private school alumni, lumps regular diplomas together with GEDs, and overlooks people who live in prisons and other institutions. Missing for decades has been an accurate count of graduation rates in our public schools, where the most serious dropout problems exist.
Why be so concerned about graduation rates now? The No Child Left Behind law places unprecedented stakes on student performance. But suppose the law held high schools accountable only for test scores? Beleaguered school officials might feel so pressured to raise test scores that pushing low-performing students out of school would seem like the best way to boost their numbers. Attaching meaningful stakes to graduation rates is intended to counteract this perverse incentive.
Practices exposed during the past year in some school districts illustrate these dangers. More than a dozen Houston schools were accused of reporting dropouts at less than half of the true rate. That city’s schools had been thought to exemplify the "Texas miracle," a shining star in the state accountability system that became a model for the No Child Left Behind Act. Investigations in the New York City public schools—the nation’s largest district, with more than 1 million students—have uncovered what appear to be widespread attempts to push lower-performing students out of regular high schools and into alternative educational programs. Such students would not be counted as dropouts.
The lurking fear is that revelations in Houston and New York City may be just the tip of the iceberg, and that such alarming practices may be business as usual in far more places than we suspect. Unfortunately, fudging the math may remain all too easy. Regulations issued by the Department of Education give states tremendous flexibility over graduation-rate accountability. Consequently, the No Child Left Behind law provides very little protection to lower performers who might be pushed out of school, a disproportionate number of whom are minorities.
States, for example, can select their own approaches to computing graduation rates, even though different calculation methods can produce markedly different results. Most states have chosen methods that depend heavily on dropout numbers. But since dropouts are systematically undercounted, this default strategy paints an overly rosy picture of high school completion.
The reasons that dropouts go uncounted range from deliberate falsification of data to the genuine difficulty in tracking a student who leaves a school. Did she transfer to a different school or district? Has he dropped out? Even well-intentioned attempts to locate a missing student often prove fruitless, especially in communities characterized by instability and poverty. That said, the distinction between a transfer and a dropout makes a tremendous difference when calculating graduation rates.
The double standard being applied to this new accountability—strict for test scores and lenient for graduation rates—raises another major specter. All states must adopt the same lofty goal for tested achievement (100 percent of students earning proficient scores by 2014) and set explicit annual benchmarks for improvement. When it comes to graduation rates, states are allowed to establish their own goals, which range anywhere from 50 percent to 100 percent. Any amount of progress can be considered adequate, even if it is infinitesimal.
More alarming, regulatory double-dealing severely lowers the stakes for the graduation rates of at-risk student groups. By law, failure to make adequate progress in test scores for even one historically low-performing group—a single racial-ethnic category, low-income students, those in special education, or those with limited English proficiency—triggers a series of increasingly aggressive corrective actions. But when it comes to graduation rates, the stakes are penny ante. According to the federal rules, states only have to count the overall graduation rate when meting out sanctions. This allows them to virtually ignore disparities between more and less disadvantaged groups.
Prospects for the future, however, may not be altogether bleak. Despite a lax regulatory stance now, no legal impediments bar the Department of Education from putting graduation rates on a par with test scores in the accountability game. Likewise, states might choose to raise the bar on their own initiative. But that high road is a risky one, since their schools would face federal sanctions should they fail to make the grade.
The No Child Left Behind Act remains a work in progress in many ways, so a midcourse correction is still possible. But what course will lead us to stronger and smarter accountability for graduation rates?
For starters, states should be required to calculate graduation rates using formulas federally approved as valid and reliable. In grammar school, every student learns that math problems need to be double-checked. So the Department of Education should independently analyze graduation rates as a reality check against state-generated results. In December, Secretary of Education Rod Paige took an important first step in this direction by convening an expert panel to investigate these technical issues.
Goals for graduation rates should be attainable, but must also be set high enough to signify a real commitment to progress. Persistent failure to meet performance targets for graduating more students should carry serious consequences, just as it does for test scores. But we cannot allow diplomas to become empty credentials. They must be earned by acquiring the knowledge and skills that will prepare all students for a meaningful life after high school.
If truly no child is to be left behind, we must also show an unwavering commitment to closing the high-school- completion gap. Social justice demands that real stakes be attached to improving graduation rates of minority students and others from disadvantaged groups. This race is not worth winning unless all of our children reach the finish line together.
Finally, our current notion of accountability needs an attitude adjustment to rehabilitate its punitive spirit. It must become a positive force aimed at providing all students with the high-quality learning and supports they need to thrive. Despite challenges, some schools do succeed in serving youths from diverse backgrounds. We should recognize their accomplishments and learn from their examples. Struggling schools need help and hope more than a trip to the woodshed.
Who is affected most by the new math on graduation rates? The nearly 1.3 million current 9th graders who will leave public high school without earning a diploma. Most of these kids are members of racial and ethnic minorities whose opportunities are circumscribed in ways large and small. For them, this is more than just a numbers game. It’s their future at stake.
Christopher B. Swanson is a research associate at the Urban Institute’s Education Policy Center, in Washington. Over the past year, he has published a series of reports on U.S. graduation rates and how they relate to current policy.
Vol. 23, Issue 43, Pages 31, 40Published in Print: July 28, 2004, as The New Math on Graduation Rates