A bleaker view of dropout problem; 17 districts’ ratings would be lower if federal definition used

By Joshua Benton
Staff Writer

Page 1A

Seventeen area school districts would have had their state ratings lowered last year if the word “dropout” meant the same to Texas officials as it does to the federal government, according to a new data analysis.

For the first time, the state has recalculated the dropout rates of all Texas school districts using the federal definition of a dropout, which some critics say gives a more accurate picture of how many students actually quit school.

The result: The federal data say more than 9,000 students dropped out of North Texas high schools in 1999-2000. The state had reported only about 3,000.

Dropout rates are used to calculate school and district ratings. Test scores are also used in the calculations. Had the federal definition of dropout been used in the state’s accountability formula, seven area districts would have been rated “academically unacceptable” – the state’s lowest possible rating – including the area’s three largest districts: Dallas, Fort Worth and Arlington.

“The federal definition is the one that’s most universally used,” said Jay Smink, executive director of the National Dropout Prevention Center at Clemson University. “It’s not perfect, but it’s better than what Texas uses.”

Districts generally strive for higher ratings to improve the perception of their schools. But when a district is deemed “academically unacceptable,” it receives an official visit from the state and must create a plan to improve its performance.

The question of how to count dropouts has long dogged Texas and other states. The state’s preferred calculation is called an annual event dropout rate: It attempts to count the number of students in grades seven through 12 who drop out in a given school year.

By that count, the annual dropout rate in Texas was just 1.3 percent in 1999-2000.

But by other measures, the state underestimates the true size of its dropout problem. A study done last year for The Dallas Morning News by the education research group Just for the Kids estimated that about 20 percent of Texas high school students drop out before graduating.

Last year, Texas enrolled 364,000 high school freshmen, but only 225,000 seniors.

“I do not believe the way the state calculates it now is accurate,” said Jack Singley, superintendent in Irving, which reported a dropout rate of 0.6 percent to the state in 1999-2000. “I just don’t believe our dropout rate was 0.6 percent. I firmly believe we have a higher dropout rate than that. I believe that in my heart.”

Different method

The federal government uses a different method of calculating dropout rates. First, it looks at only grades nine to 12, excluding from the count a large number of middle school students, who rarely drop out.

Second, if a student drops out of a high school but says he will pursue a General Educational Development certificate, Texas officials don’t count him as a dropout – even if he never actually gets a GED. Federal officials count him as a dropout unless he actually earns the certificate.

As a result of these and other differences, the federal government argues Texas’ dropout rate was 5 percent in 1999-2000, not 1.3 percent. Irving’s annual dropout rate increased from 0.6 percent to 4.7 percent when using the federal definition.

“We have all the data, but if you use different definitions, you end up with different results,” said Criss Cloudt, the state’s associate commissioner for accountability reporting and research.

Arguing over statistics may seem unimportant, but dropout rates are crucial to school districts because they are a key element of the state’s rating system.

For the year in question, districts had to keep their annual dropout rate below 1 percent to earn an “exemplary” rating, the state’s highest. To be “recognized,” the state’s second-highest rating, the rate had to be below 3 percent. Finally, districts had to keep their dropout rate below 5.5 percent to avoid being rated “academically unacceptable.”

Lower assessments

Because dropout data collection lags behind other statistics the state gathers, the 1999-2000 dropout rates were used to determine school ratings last year, in 2001.

Of 53 North Texas school districts, 17 would have had their state ratings drop if their dropouts had been counted the federal way instead of the state way. Seven districts would have dropped all the way to academically unacceptable: Arlington, Azle, Castleberry, Dallas, Fort Worth, Terrell and Wilmer-Hutchins. As it was, only one district in the state received the academically unacceptable rating in 2001.

Nine other districts would have dropped from recognized to acceptable: Birdville, Carrollton-Farmers Branch, Ennis, Everman, Forney, Irving, Mabank, Mesquite and White Settlement.

Another district, Highland Park, would have dropped from exemplary to recognized, but that was because of a data error in that district’s dropout reporting, not because of an actual increase in the number of dropouts.

“I’ve never felt good about that state number,” said Dr. Singley, the Irving superintendent. “I don’t think it’s accurate.”

2 ways, 2 opinions

Neither method of counting dropouts has universal support. Critics such as Dr. Smink say the state system created artificially low rates. But Dr. Cloudt and others say the state’s more restrictive definition encourages schools to help more at-risk students by enrolling them in GED programs or re-enrolling dropouts back in school.

“We worry about the state definition because, once you start talking about anything else, there’s just no end to it,” said Arlington Superintendent Mac Bernd.

“There are a number of different organizations that calculate it in different ways. We follow what the state law and state regulations tell us.”

In future years, the state will report both its traditional annual dropout rate and a rate based on the federal definition. But it won’t use either when it rates school districts in the future, Dr. Cloudt said.

The state’s new accountability system, which will debut in 2004, will use an altogether different calculation called a completion rate, which tracks a group of students over four years of high school to see how many graduate.

“I think a completion rate is much closer to the public’s understanding of the goal of high school,” Dr. Cloudt said. “That’s to complete four years of education and receive a credential.”