D.C. Mayor Adrian M. Fenty’s choice to reform public schools has been unable to provide proof of the remarkable student improvement she achieved during her brief teaching stint in Baltimore.
“We were told that these kids came in on this level and they were leaving on average at this level,” said D.C. schools chancellor-nominee Michelle A. Rhee, who has noted a dramatic improvement in student test scores in her resume.
“I didn’t think to ask back then for solid documentation or proof or any of those things,” she said. “As a new teacher, I didn’t think those things were particularly relevant.”
In the 1993-94 school year, when she taught second-graders at the inner-city school, those students had scored at the 13th percentile on standardized tests.
By the end of the 1994-95 year, after Mrs. Rhee had taught the same students as third-graders, 90 percent of them scored at the 90th percentile, according to her resume.
Mrs. Rhee said the test results were achieved on the Comprehensive Test of Basic Skills (CTBS).
Her biographical information on the mayor’s office Web site (http://dc.gov/mayor/news/) and on the Web site of her former nonprofit — the New Teacher Project (www.tntp.org) — both say such “outstanding success” in the classroom earned Mrs. Rhee national media acclaim.
But education experts note that most low-income schools have a high student-turnover rate and Mrs. Rhee taught her students as part of a team. Tying the percentile jump specifically to her is extremely hard to do, they said.
“Although there were some significant gains for third-grade Title 1 students in reading [during Mrs. Rhee’s tenure], there is nothing that would establish a sufficient evaluation link between that particular population of students and any particular individual staff member,” said Ben Feldman, who is in charge of testing for Baltimore schools. “You couldn’t go there.”
In addition, establishing a precise link between student achievement and Mrs. Rhee’s performance in the Baltimore school system is difficult in part because of dated information systems and antiquated storage.
Mr. Feldman said retrieving data from a decade ago is hard because his office changed its information storage systems for the year 2000.
Still, the normal curve equivalent score (which is similar to a percentile) on the CTBS for Harlem Park second-graders was 27 in reading and 43 in math in the 1993-94 school year, according to a 1995 report published by the University of Maryland-Baltimore County.
The report also shows that third-graders at the school for two years achieved a score of 45 in reading and 51 in math in 1994-95. The report does not break down scores by specific class and excludes some students from the totals, including those who received special-education services.
Those scores show significant gains at Harlem Park, but the question remains whether they support the remarkable gains highlighted by Mrs. Rhee and her backers.