Report blasts state over dropouts
Graduation rates inflated, study finds
Nanette Asimov, Chronicle Staff Writer
Thursday, March 24, 2005
In a searing indictment of California's school system, Harvard University researchers say the state graduates only 71 percent of its high school students -- not the 87 percent it claims.
Moreover, the researchers assert, the state is harming students and the public in general by failing to keep students in school and accurately report the numbers.
The Harvard study found that some California schools are simply "dropout factories" and that dropout estimates for nonwhite students are worst of all: Just 50.2 percent of black ninth-grade boys received a diploma four years later.
California's own calculations do not show ethnic or gender breakdowns.
The researchers from Harvard's Civil Rights Project said the state's method of determining its graduation rate is based on a "flawed National Center for Education Statistics formula" that uses unreliable dropout data in the calculation.
"We believe that the most useful and accurate estimates of high school graduation rates currently available are those that are based on the actual enrollment data that each district provides," says the study.
An example is the "Cumulative Promotion Index," developed by the Urban Institute in Washington, D.C., which tracks groups of students from ninth to 12th grade and reports what percent of students finishes. Using that method, the Harvard researchers drew these conclusions about California:
-- Black and Latino students are three times more likely than white students to attend a "dropout factory," a school with graduation rates of 60 percent or less.
-- Only 10 percent of black students and 7 percent of Latinos attend schools that graduate 90 percent of students.
-- Schools with healthy graduation rates usually have few students from poor families: At 80 percent of such schools, fewer than 1 in 5 students is low-income.
"California is doing a miserable job of addressing the graduation rate crisis," said Daniel Losen, a senior researcher with the Civil Rights Project. He added that a school's ability to graduate its students is "the ultimate measure of accountability."
As dropout rates rise, so do crime and poverty, the researchers said. Even relying on California's more conservative estimate of 66,657 dropouts from grades 7 to 12 in the 2002-03 school year, the state will lose $14 billion in wages and have to spend $73 million to support an additional 1,225 inmates, said one of the researchers, Russell Rumberger, an education professor at UC Santa Barbara.
The Harvard researchers began tackling the problem four years ago because "the extremely high dropout rate, particularly for minority students, is a basic civil rights issue," said Harvard education professor Gary Orfield, director of the Civil Rights Project.
"If they don't make it through high school, they don't have a chance," said Orfield, whose group has compiled data nationally and in California, which has the largest state school system. They published a book last year titled "Dropouts in America: Confronting the Graduation Rate Crisis."
"We found there is systematic misreporting of data across the country -- and very little effort to deal with it," he said. "But there are policies that compound the problem, like high-stakes testing."
Surprisingly, state Superintendent of Public Instruction Jack O'Connell agreed with much of the Harvard study.
"Do we have a dropout problem? Yes," he said Wednesday.
"The solution is high school reform -- we need to keep music, art and athletics," he said. "For many students, that's their connection (to school). We also need smaller classes. The governor's proposed budget would be devastating to many of these programs."
O'Connell also agreed that California's method of calculating its graduation rate does indeed inflate the numbers. He blamed the federal government for ordering California to change its method in 2003 to conform with the national No Child Left Behind Education Act. Under the national formula, California's graduation rate soared from 69.6 to 86.9 percent.
Yet federal and state officials have long admitted that both methods are largely guesswork. That's because they rely on flawed estimates of the number of students who quit school and never return.
By contrast, Harvard's numbers are based on enrollment figures, not dropouts. They give California a 71 percent graduation rate -- about 333,800 students in the 2001-02 school year, which is slightly better than the 68 percent that the researchers calculated as the national average.
However, the researchers say that even though their figures are more reliable than California's, the numbers are still only estimates.
All sides agree that the most accurate method would be to assign an identification number to each student. This would allow the state to know whether the student enrolled anywhere else in the state -- say, a school in another district, night school or even a high school equivalency program.
The Harvard study says that "the current state government has refused to fund the measure."
But state Department of Finance officials say the Harvard assertion is wrong. The state has spent $67 million since 1998 to develop such a system, said H.D. Palmer, spokesman for the Department of Finance, and students in 268 of the 1,000 school districts are already involved in a pilot tracking program.
He said that by June, every public school student in the state will have an identification number.
Even so, Palmer said, it will be at least five years before California is ready to track all of its students and get accurate dropout and graduation rates.
Graduation rates inflated, study finds
Nanette Asimov, Chronicle Staff Writer
Thursday, March 24, 2005
In a searing indictment of California's school system, Harvard University researchers say the state graduates only 71 percent of its high school students -- not the 87 percent it claims.
Moreover, the researchers assert, the state is harming students and the public in general by failing to keep students in school and accurately report the numbers.
The Harvard study found that some California schools are simply "dropout factories" and that dropout estimates for nonwhite students are worst of all: Just 50.2 percent of black ninth-grade boys received a diploma four years later.
California's own calculations do not show ethnic or gender breakdowns.
The researchers from Harvard's Civil Rights Project said the state's method of determining its graduation rate is based on a "flawed National Center for Education Statistics formula" that uses unreliable dropout data in the calculation.
"We believe that the most useful and accurate estimates of high school graduation rates currently available are those that are based on the actual enrollment data that each district provides," says the study.
An example is the "Cumulative Promotion Index," developed by the Urban Institute in Washington, D.C., which tracks groups of students from ninth to 12th grade and reports what percent of students finishes. Using that method, the Harvard researchers drew these conclusions about California:
-- Black and Latino students are three times more likely than white students to attend a "dropout factory," a school with graduation rates of 60 percent or less.
-- Only 10 percent of black students and 7 percent of Latinos attend schools that graduate 90 percent of students.
-- Schools with healthy graduation rates usually have few students from poor families: At 80 percent of such schools, fewer than 1 in 5 students is low-income.
"California is doing a miserable job of addressing the graduation rate crisis," said Daniel Losen, a senior researcher with the Civil Rights Project. He added that a school's ability to graduate its students is "the ultimate measure of accountability."
As dropout rates rise, so do crime and poverty, the researchers said. Even relying on California's more conservative estimate of 66,657 dropouts from grades 7 to 12 in the 2002-03 school year, the state will lose $14 billion in wages and have to spend $73 million to support an additional 1,225 inmates, said one of the researchers, Russell Rumberger, an education professor at UC Santa Barbara.
The Harvard researchers began tackling the problem four years ago because "the extremely high dropout rate, particularly for minority students, is a basic civil rights issue," said Harvard education professor Gary Orfield, director of the Civil Rights Project.
"If they don't make it through high school, they don't have a chance," said Orfield, whose group has compiled data nationally and in California, which has the largest state school system. They published a book last year titled "Dropouts in America: Confronting the Graduation Rate Crisis."
"We found there is systematic misreporting of data across the country -- and very little effort to deal with it," he said. "But there are policies that compound the problem, like high-stakes testing."
Surprisingly, state Superintendent of Public Instruction Jack O'Connell agreed with much of the Harvard study.
"Do we have a dropout problem? Yes," he said Wednesday.
"The solution is high school reform -- we need to keep music, art and athletics," he said. "For many students, that's their connection (to school). We also need smaller classes. The governor's proposed budget would be devastating to many of these programs."
O'Connell also agreed that California's method of calculating its graduation rate does indeed inflate the numbers. He blamed the federal government for ordering California to change its method in 2003 to conform with the national No Child Left Behind Education Act. Under the national formula, California's graduation rate soared from 69.6 to 86.9 percent.
Yet federal and state officials have long admitted that both methods are largely guesswork. That's because they rely on flawed estimates of the number of students who quit school and never return.
By contrast, Harvard's numbers are based on enrollment figures, not dropouts. They give California a 71 percent graduation rate -- about 333,800 students in the 2001-02 school year, which is slightly better than the 68 percent that the researchers calculated as the national average.
However, the researchers say that even though their figures are more reliable than California's, the numbers are still only estimates.
All sides agree that the most accurate method would be to assign an identification number to each student. This would allow the state to know whether the student enrolled anywhere else in the state -- say, a school in another district, night school or even a high school equivalency program.
The Harvard study says that "the current state government has refused to fund the measure."
But state Department of Finance officials say the Harvard assertion is wrong. The state has spent $67 million since 1998 to develop such a system, said H.D. Palmer, spokesman for the Department of Finance, and students in 268 of the 1,000 school districts are already involved in a pilot tracking program.
He said that by June, every public school student in the state will have an identification number.
Even so, Palmer said, it will be at least five years before California is ready to track all of its students and get accurate dropout and graduation rates.
