Graduate rates hardly science


Data: Report that finds most states exaggerate high school graduation figures is another example of the `Lake Wobegon effect.'

December 24, 2003|By Mike Bowler | Mike Bowler,SUN STAFF

IN 1987, John J. Cannell, a West Virginia physician, discovered that all of the states were reporting above-average scores on standardized tests. Cannell called this the "Lake Wobegon effect," after Garrison Keillor's mythical town where "the men are good-looking, the women are strong, and all the children are above average."

Cannell's discovery came to mind Monday, when the Education Trust, a Washington-based student advocacy group, said all but three states (Utah, Idaho and Oklahoma) might be guilty of overstating high school graduation rates in their reports to the federal government under the No Child Left Behind Act.

It's the Lake Wobegon effect all over again, and in the case of graduation rates, it's easy as pie to exaggerate. One of education's dirty little secrets is that calculations of these rates (and their obverse, dropout rates) are so inexact as to be all but meaningless. "Differences in the ways states define `graduation rate' not only result in wide variations in the data but, in many cases, significantly understate the problems that schools and students are facing," said the Education Trust report.

Graduation rates are important indicators of how well schools perform. That's why the No Child Left Behind legislation ordered the states to pay attention to dropouts, particularly minorities. States were required to report baseline data by this past Sept. 1 and to make "adequate yearly progress" toward a goal (set by each state) to be reached in 2013-2014.

The problem is that schools can't track individual students through high school. And urban school systems, which have the highest dropout rates, are also the most technologically inept. They lose students all the time.

Maryland has a formula for determining graduation rates. You can read it, or at least try to, on The rate is derived by "dividing the number of high school graduates by the sum of dropouts for grades nine through 12, respectively, in consecutive years, plus the number of high school graduates."

"Huh?" you say. But think about it. This formula divides the number in a graduating class, plus dropouts over the previous four years, into the number in the graduating class. Unlike a formula that simply compares the size of the freshman class to the size of the graduating class four years later, this approach takes into account the fact that some students who leave school aren't dropouts. They transfer to other schools or to private schools. They die. In Baltimore City's case, they simply disappear. Or they're among the hundreds of students retained for two or three years, usually in the ninth grade.

The Web site says Maryland's 2003 graduation rate was 85 percent, Baltimore City's 54 percent and Prince George's County's 90 percent. Using a similar formula, South Dakota reported a 97 percent graduation rate. Excluding dropouts from its calculations altogether, North Carolina reported a 93 percent rate.

But using different data from the federal National Center for Education Statistics, Education Trust put Maryland's rate at 74 percent and North Carolina's at 63 percent. With such wild variation, whom should we believe?

Graduation and dropout statistics are like the lamppost supporting the drunk: They provide plenty of support but not much illumination. And no one pays a lot of attention to them. Thus, Houston bragged about remarkably low dropout rates during and after the reign of Superintendent Rod Paige, who was elevated to U.S. secretary of education largely on the basis of the "Texas miracle." Finally, reporters looked into the miracle, and it lost its shine.

One would expect school officials, who are only human, to use formulas that cast their own situation in the best light. But in the case of graduation rates, are they massaging their numbers to hide dropouts? If so, the Education Trust warned, such action would "represent unprofessional and unethical choices on the part of educators."

Wobegon, indeed.

Fund raising at Towson U. in the name of Johnny U.

Some universities build sports palaces and sell the naming rights (and their souls) to corporations. Towson University took a different path. It named its new stadium after the late football great Johnny Unitas, then set about raising money in Unitas' name. Suggested contributions, all a variation of Unitas' uniform number, range from $19 to $2.19 million.

For $190, they'll put your name on one of the 11,000 seats in Unitas Stadium. So far, 600 have been sold, Jeffrey F. Dudley, director of the campaign, said yesterday.

"Everyone in Baltimore seems to have a Unitas story," he said. "They'll call up and buy a seat just because Johnny took the time to autograph their kid's football 20 years ago, or because they admired his integrity and modesty. Can there be better reasons for giving?"

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.