Dayton Daily News education reporter Scott Elliott writes about schools, kids, teaching and learning.
Home > Blogs > Get on the Bus > Archives > 2003 > September > 15 > Entry
Error bedevils charter schools
DAYTON - Some charter schools will have to live with errors on their state report cards because they failed to recognize the need to add together results from fourth-grade test administrations in October and March last year.
The Ohio Department of Education will allow no report card changes because officials say schools had ample opportunity to correct their scores. Schools had 15 opportunities from the time they received their scores in June until the Aug. 1 deadline to request changes, said Jill Dannemiller, associate director for the office of accountability.
“They have plenty of time over the summer to review and correct the data,” she said.
But charter school operators say their smaller operations cannot afford professional testing coordinators that many school districts employ, leaving principals and others to tackle the often complex mathematics of reporting proficiency test results to the state.
“Public schools have had many more years of working with the state like this,” said Don Vermillion, interim superintendent of Alliance Community Schools, which operates the Dayton Academy and Dayton View Academy. “Charter schools are still figuring it out.”
Charter schools are free privately run but publicly funded schools.
A Dayton Daily News review of Dayton charter school data found mostly small discrepancies for at least seven of the 12 local charter schools that had fourth-grade students last year. In none of the cases does the data problems appear the change a school’s report card rating. All Montgomery County public school district data appeared accurate, the Daily News review showed.
The problem with fourth-grade scores relates to Ohio’s plan, which was later dropped, to start a “fourth-grade guarantee” requiring fourth-graders to pass the reading proficiency test or be held back. In preparation for the guarantee, which was to go into effect in the 2001-02 school year, Ohio began offering fourth-graders more chances to take the fourth-grade test. But in June 2001, the guarantee was softened so it would not require those who fail to repeat the grade.
Unlike the sixth-grade exam, which is administered once in March, the fourth-grade test now is given in October and March each year. Last year, the company that compiles test results for Ohio reported both the March-only figure and cumulative results for school districts and charter schools.
But Dannemiller said the cumulative figures were often inaccurate, because the set of students who took the test was often slightly different from October to March - because of students who changed schools or were absent on test day.
So the testing company no longer compiles cumulative data, instead leaving it up to the schools themselves. Under Ohio’s reporting process, the company provides test data to the schools, which then report their scores to the state. At fourth grade, Ohio now expects schools themselves to create cumulative data.
That works fine for most school districts, which have administrators, or in some cases testing coordinators with expertise in statistics, to crunch the numbers. But at charter schools, the task often falls to principals or office staff members.
And the calculations are not always easy. Each student’s October and March scores must be matched to determine who passed each portion of the test on at least one of their two chances.
Some charter schools apparently reported just their March results or just their October results by mistake, and as a result got no credit for some students who passed.
At the Dayton Academy, for example, the school’s state report card shows 19.5 percent passed citizenship, but data from the March administration shows 27 percent passed in March alone. The 19.5 percent may be an October-only figure.
The Dayton Academy appears to have conflicting data for its results on all five parts of the fourth-grade exam. The scores were too low by 3.5 percentage points in math, 7 points in reading, 3 points in writing and 4 points in science, Vermillion said.
Mike McCormick, superintendent of the three Richard Allen academies, said his schools merely sent the state data for October and March as is, without creating cumulative results.
“It was news to our folks that you could change the numbers,” he said. “That’s an incredible way for the state to do business. No one should be able to manipulate the data of test scores.”
But Dannemiller said only the schools know the names of each student - not just identification numbers - and therefore are best able to correlate results from the two tests. The state offers considerable training for schools to learn the data reporting system, she said.
The affected schools worry that the report card, which is mailed to parents and available on a state Web site, won’t reflect how well the students really performed and that the school’s credibility could be questioned when it sends out its own data.
“We’re in the process of putting together information to send to parents, and those who fund our schools, to give them the actual results,” Vermillion said. “The report card is going to stay the way it is.”
Permalink | | Categories: Charter Schools and School Choice, My Favorite DDN Stories
