A study commissioned by the Virginia Department of Education has found that high school graduation rates held steady for white students in 2004 -- the first year in which Standards of Learning exams were required for a diploma -- but dropped significantly for black and Hispanic students.
According to researchers at Virginia Commonwealth University, the on-time graduation rate for white students last year was 77.4 percent, compared with 61.3 percent for black students and 66.5 percent for Hispanics. Those figures represented a drop of 4.9 percentage points from 2003 for black students and of 11.6 percentage points for Hispanics.
The researchers concluded that "while there may be some negative consequences for graduation associated with high-stakes testing for specific groups of students, there is little evidence that there have been dire consequences for most students."
They also estimated that the percentage of students who dropped out of school during the four years of high school, 10 percent, was lower for the Class of 2004 than for past classes.
Starting in 2004, seniors had to pass SOL exams in reading and writing, as well as four other subjects of their choice, or tests from a list of accepted alternatives to receive a standard diploma. The department contracted with the Commonwealth Educational Policy Institute at VCU, paying $48,000 to study how students met the requirements.
Charles B. Pyle, spokesman for the Virginia Education Department, said officials were not surprised by the results for minority students. But he said they are determined to give them more help rather than back away from the tests.
"It clearly shows that the increasing focus of this agency on closing the achievement gap is vital," he said. "That's where the battle is today in Virginia's standards and accountability program."
Among other conclusions, the researchers found that, overall, 5.6 percent of last year's graduates received a "modified standard" or "special" diploma, categories reserved for special education students, compared with 3.2 percent the year before. It was more pronounced for black and Hispanic students: The percentage of black graduates earning modified or special diplomas rose from 5.5 percent in 2003 to 10.1 percent in 2004, and among Hispanic students it rose from 2.9 to 4.9 percent.
Department officials had contested several major components of the report, rejecting two previous drafts. They accepted the final version, submitted April 8, but attached a letter from State Superintendent of Public Instruction Jo Lynne DeMary explaining some areas of continued concern.
They had objected strenuously to a finding that the graduation rate for students who participated in the department's Project Graduation, which provides extra tutoring for struggling students, was only 58 percent. The finding was the result of a survey of school systems that DeMary said vastly overestimated the number of participating students.
In fact, she wrote, 75.3 percent of the 2,893 students who participated in the program graduated. The issue was particularly sensitive because the department and Gov. Mark R. Warner (D) had touted the program's support for students as standards were raised.
She also wrote that the report omitted context about the different kinds of diplomas, including the fact that more students with disabilities, including black and Hispanic students, completed high school in 2004 than in previous years. That suggests that the modified standard diploma, introduced in 2000, has provided such students an incentive to finish school, she said.
VCU researchers studied graduation rates by examining the size of each freshman class compared with the number of diplomas handed out four years later, a method called for in the federal No Child Left Behind law. Several researchers contend that the technique skews results because many students are held back after completing the ninth grade, meaning that the class is abnormally large.
Jay P. Greene, who has studied graduation rates for the Manhattan Institute, said he has found that exit exams in other states have had little impact on graduation rates when they are calculated differently. He said it is common to find an inflated ninth-grade class in the years after tests are introduced because teachers hold students back so they can master basic skills before facing the exams.
Walt Heinecke, an associate professor of research, statistics and evaluation at the University of Virginia's Curry School of Education, contested that claim. He said other studies have found that exit exams encourage students to drop out. Heinecke studied drafts of the VCU report and faulted it for not examining whether students who drop out leave school because of the SOL tests.
In addition, he said the study could have examined why rates vary widely in different parts of the state. Fairfax County had an on-time graduation rate of 86.4 percent last year, while in Norfolk the figure was 39.3 percent.
"If it's important for the state to find out whether these exams are having an effect on graduation rates, particularly for minority students, then it's important to spend some more money on this," he said.
James McMillan, lead researcher for the study, said asking those questions would "probably be very appropriate" but said this study was not intended to ask them and called for continued research.
Pyle said that the department plans to continue studying graduation rates but that there is no plan to contract with an outside group next year.
没有评论:
发表评论