Testing Strategies Also Helped Boost D.C. Students' Scores
Testing Tactics Helped Fuel D.C. School Gains
Tutoring, Eligibility Review Had Roles
By Bill Turque
Washington Post Staff Writer
Friday, July 17, 2009
When Mayor Adrian M. Fenty announced the continued growth of standardized test scores for District students Monday, he hailed it as "powerful evidence of the incredible work being done by teachers, principals and most importantly our students."
What Fenty did not say was that the two-year improvement in District of Columbia Comprehensive Assessment System results -- including an average of nearly 15 percentage points in the pass rates on elementary reading and math tests -- was also the product of a strategy that relied on improved statistical housekeeping.
These include intensive test preparation targeted to a narrow group of students on the cusp of proficient, or passing, scores, and "cleaning the rosters" of students ineligible to take the tests -- and also likely to pull the numbers down.
Chancellor Michelle A. Rhee described some of these approaches as the pursuit of "low-hanging fruit."
The initiatives are neither novel nor improper. They've been in the toolboxes of urban school leaders since the inception of the No Child Left Behind Act. The law requires schools to show annual progress toward a goal of all students passing reading and math tests by 2014.
Rhee, who says she would like to see the law amended to emphasize year-to-year academic growth, said this week that much of what she had done was a matter of common sense.
"In our first year, we found that certain basic things were not happening," she said. "There were actions we took to ensure we were maximizing our potential to be successful."
The biggest gains were in the elementary grades, where almost half of those tested were deemed proficient: 48.6 percent in math (up from 40.5 percent in 2008) and 49.4 percent in reading (up from 45.6 percent in 2008). In 2007, fewer than a third of elementary students were considered proficient in either subject. At the middle and high school levels, reading proficiency grew from 39 percent to 41 percent; math, from 36 percent to 40 percent.
Some teachers and parents wonder whether the effort and attention devoted to lifting scores helped the children who need the most attention: those years behind grade level in reading and math skills.
One of Rhee's most widely discussed initiatives was "Saturday Scholars," a 13-week program for about 5,000 invited students whose academic records suggested that they were close to scoring at proficiency levels.
Critics call Saturday Scholars and programs like it "educational triage" that focuses disproportionate attention on students who require the least help.
Kerry Sylvia, a social studies teacher at Cardozo High School, said Saturday Scholars was less about serving children and more about making the adults who run the school system look good.
"There are students in my classes who are struggling with basics, and yet we're pouring all of this money into a program not just focused on tests, but on tests for a few students so the scores will look good," said Sylvia, who is a member of Teachers and Parents for Real Education Reform, a group that opposes some of Rhee's initiatives.
School officials said that although students were invited to the Saturday sessions, the classes were open to all and will be a permanent part of the District's academic program, not a short-term fix. There were also other measures aimed at students further behind in basic skills, including help with the written portions of the school system's tests.
Less-visible strategies pursued by Rhee and Fenty (D) involved cleansing and reorganizing the school system's databases. In past years, officials said, high schools did not have a clear idea of which students were eligible to be tested, which resulted in some who were not 10th-graders (the only high school testing year for the District's standardized tests) sitting for the exams. Starting in the 2007-08 school year, Rhee's staff reviewed transcripts and calculated which students should be tested on the basis of credits earned.
Other score-leavening methods included redefining a failing test. During the first two years that the District used the Comprehensive Assessment System exams (before that, it used the Stanford-9), any student who did not take the test counted as having failed. Starting in 2008, students who did not take the test did not count as either a pass or a fail.
Some changes, mandated by the federal government, worked against the goal of higher scores, Rhee said. More than 2,000 students in D.C. public and public charter schools with reading difficulties, who used to have the questions and passages in the English language arts section of the test read aloud to them, must now take the regular written exam.
Michael Casserly, executive director of the Council of Great City Schools, a coalition of the country's 66 largest urban systems, said Rhee's approach has been "right and strategically smart."
Casserly said Rhee met with him shortly after she became chancellor in 2007 and wanted ideas on how to raise scores quickly. Although Rhee was also committed to raising the quality of teaching and learning over the long term, Casserly said she faced a special challenge running public schools in a city that historically has been quick to jettison school leaders. In other words, she needed some early wins.
Finally, Fenty and Rhee have harvested the "fruits" of students and teachers who have simply grown accustomed to the test. The District's CAS, patterned after a similar test used in Massachusetts, was introduced in 2006 by Rhee's predecessor, Clifford Janey. Often, educators find that scores are low in the first year of a test and then rise significantly in the next couple of years. So it was in the District, where the first round of scores was extremely low, with only about a third of all students demonstrating proficiency in reading or math.
Sustaining test score growth will be more challenging for Rhee. Scores in coming years will "absolutely" be based on the District's success in making deeper long-term changes to the quality of classroom learning. Those changes will start showing up next year, she said.
"I'm very excited about next year," Rhee said.
Let's Screw The Most Vulnerable Among Them
Here is some disgusting new evidence that NCLB causes schools to manipulate numbers at the expense of educating children.