Superintendent David W. Hornbeck of Philadelphia is hailing a rise in city test scores as “the first fruits” of his embattled reform program--an interpretation disputed by the city’s combative teachers’ union.
At a news conference attended by U.S. Secretary of Education Richard W. Riley this month, Mr. Hornbeck announced an across-the-board increase in scores last spring on a standards-based achievement test introduced last year amid loud complaints from city teachers. He also highlighted a drop in the sizable number of students who were missing in action when the tests were first given--a problem that prompted union leaders to question the validity of the year-to-year comparison.
“These are the first fruits of a three-year struggle to become the first district where fundamental reform results in systemwide improvement in student achievement,” Mr. Hornbeck said in announcing the results. “This is a remarkable accomplishment for our students and our teachers.”
While calling the results positive, union leaders questioned whether the district’s reform efforts deserved the credit.
“These kids did better because they were prepared,” said Ted Kirsch, the president of the Philadelphia Federation of Teachers. “I don’t think it’s simply, ‘Oh, my program works.’”
Breakdown Shows Gains
The Stanford Achievement Test-9 assessed reading, mathematics, and science in grades 4, 8, and 11. Based on their scores, students were grouped into four performance levels: below basic, basic, proficient, and advanced.
At every level and grade, the proportion of students scoring at the basic level or above increased, when viewed as a percentage of the total number of students who should have been tested, officials in the 215,000-student district reported.
If only students who took the test are counted--a method that Mr. Hornbeck called invalid--the results were more mixed. Viewed in that way, for example, 4th grade gains were even greater, while 11th grade scores in reading and math were slightly lower.
Regardless of the reporting method, scores were generally strongest in the 4th grade and weakest in the 11th. The poorest showing was in 11th grade science, where only 8.7 percent of students showed at least basic understanding, the district reported. The best was in 8th grade reading, with 54.2 percent of scores in at least the basic range.
Among 4th graders, the proportion of students scoring at a basic level or above rose from 43.7 percent to 50.9 percent in reading, from 38.3 percent to 44.3 percent in math, and from 39.9 percent to 48.8 percent in science.
In the 8th grade, those scores climbed from 48.6 percent to 54.2 percent in reading, from 21.1 percent to 24.2 percent in math, and from 23 percent to 28.7 percent in science. And among 11th graders, the percentage of students scoring at least at a basic level rose from 26.2 percent to 34.8 percent in reading, from 12.1 percent to 14.8 percent in math, and from 5.2 percent to 8.7 percent in science.
No-Shows Stir Debate
Since it started last year, Philadelphia’s standards-based testing program has drawn fire from the PFT. Union leaders argue that the test’s first administration should not be used for comparisons, in part because so many students were not tested and because schools were not fully aware of the exams’ importance. (“Phila. Assailed for Large Numbers Of 11th Grade ‘No Shows’ on Tests,” Feb. 19, 1997.)
“The first test from our point of view was totally invalid,” Mr. Kirsch said last week.
Despite such criticisms, Mr. Hornbeck is using the first-year results as a benchmark. Starting next year, schools will be rewarded or penalized based in large measure on whether their test performance has improved since 1996, in addition to other factors such as promotion and graduation rates and attendance.
Mr. Hornbeck not only disagrees that the high number of untested students compromised the first-year results, but he objects strongly to calculating the passing rate based only on those tested.
“If we shifted now, and we went to that way of doing it next year,” he argued, “even the 4th grades would have an incentive to say to certain kids, ‘Why don’t you not come to school today?’”
In both years, the ranks of untested students were greatest in the 11th grade and smallest in the 4th.
Besides the strength of the 4th grade scores, Mr. Hornbeck said he was most pleased by the 16 percent increase in the number of students tested and by the fact that about 40 percent of the extra students were those with disabilities or limited English-language skills.
Such children were traditionally considered “throwaways” to be excluded from testing because of their potential to drag down scores, the superintendent said. “The important thing is they are on the radar screen.”