Less than a month after Baltimore's school system reported )) declining standardized test scores at elementary schools run by Education Alternatives Inc. (EAI), the company presented a much rosier view yesterday and questioned the "integrity" of the city results.
In a move some viewed with suspicion, a panel of eight educators commissioned by the for-profit Minneapolis company disclosed an analysis of EAI-run schools, as well as those in the entire school system, that eliminated the scores of the lowest-scoring students over the past two years.
Citing an inexplicably large proportion of students receiving the lowest possible score, including some who had done well in the previous year, the review panel questioned the validity of results released by EAI's employer, the city school district. "These disparities are highly unusual and totally inconsistent with our experience in public education," said Greta Shepherd, a former school superintendent and New Jersey state education department official who headed the panel. "The scores are clearly not an accurate reflection of student ability."
At a time of growing doubt about the school-privatization experiment here, the company launched its public relations counteroffensive, beginning with the release of the test scores analysis yesterday morning at Harlem Park Middle School.
The second phase in EAI's publicity blitz came late yesterday afternoon at Harlem Park, where the company held a most unusual annual meeting. Little business was transacted beyond the nomination of a director -- EAI chief executive John T. Golle, who was unopposed. Revenues, earnings and cash flow were hardly mentioned.
Instead, the annual meeting, EAI's first ever in a school, was acombination of talent show and revival, complete with "amens."
The kids were wonderful.
A dapper Tavon Campbell, 8, played Alex Trebek in Harlem Park Community School's version of the "Jeopardy" game show. "The number of computers in each classroom," he said. "What is four?" came the response from a beaming contestant.
Tears ran down the cheeks of some at the meeting as shareholders listened to Tavonne Hasty, an 11-year-old sixth grader, sing a gripping version of the Gospel song "I'm Still Holding On," backed by his father, Melvin Muldrow, on keyboard.
Schools Superintendent Walter G. Amprey, who has been repeatedly criticized as becoming too close to EAI, professed his total commitment to the school system's "partnership" with EAI, which began running nine schools in 1992 and has since taken on more limited roles at three others.
And he boasted about a "healthy deception" that he said he has used to bring the program along so far.
"If we were to stop our relationship with EAI because the forces that are against us were to insist that we stop, I confess to you my deception: We have already changed because of what has happened in the short time we have been together," Dr. Amprey said.
"As John Golle, my good friend, would say, the genie is out of the bottle. This is proliferating nationwide," Dr. Amprey said.
The evaluation panel analyzed 1993 and 1994 results for the Comprehensive Test of Basic Skills (CTBS). Scores are measured in normal-curve equivalents (NCEs) on a scale of 1 to 100 with 50 being the national average.
The panel threw out all scores of "1," citing questionable validity.
Citywide, the number of "1" scores increased to 5,738, from 4,181, Dr. Shepherd said, and many of those students had received good scores the previous year. The proportion of students receiving the lowest score was higher at EAI-run schools than in the district as a whole, the panel said. Also, only the scores of students who had been in the schools at least one full year were used for comparison, because many students in the city switch schools at least once a year.
The CTBS is not a formal part of the Maryland state testing program, though the state monitors results in the 24 school districts and publishes the results for grades three, five and eight. Ronald A. Peiffer, a spokesman for the state Department of Education, said yesterday that "there is no precedent for scores on the CTBS to be thrown out after the fact." Some students, such as those with disabilities, are excluded from taking the test, but that happens before it is given, he said.
According to the panel's analysis, average overall math scores rose at seven of the eight EAI-run "Tesseract" elementary schools, six of which exceeded the districtwide gain. In reading comprehension, four EAI-run schools gained in reading, and five of the privatized schools exceeded the districtwide average gain.
But the panel included no scores from spring 1992, before EAI began running the Tesseract schools, making assessments of the experiment's progress during its first two years impossible -- least by employing the panel's methods.