Atlanta鈥檚 middle school students were just starting their academic careers when the district began rolling out efforts to improve reading and mathematics instruction. Now, the district鈥檚 8th graders are gaining faster than much of the nation and many other cities in raising achievement in those subjects, according to the urban district results of the .
The 51,000-student district鈥檚 4th graders also turned in impressive gains on the tests, particularly in reading. Such progress has drawn widespread praise for Atlanta even as the district鈥檚 overall scores continued to fall short of national averages and those of several other large-city systems that took part in the study, released earlier this month.
鈥淚鈥檝e been saying since I got here that we would see this improvement move year by year through the grades,鈥� said Beverly L. Hall, the superintendent of the district since 1999. 鈥淚 keep saying we鈥檝e got to fix the pipeline, 鈥� but I still think middle schools are our weakest link that we have not yet fixed.鈥�
On average, scores among students in large cities still lag well behind national results, but achievement rose in 4th and 8th grade math by statistically relevant margins over the past two years and among 4th graders in reading. Eighth grade scores in reading remained flat. The Austin, Texas, and Charlotte-Mecklenburg, N.C., districts, however, performed close to or better than the national averages in both grades and on both tests.
Better Data Usage
The improvements come as 鈥渦rban and other educational systems have better data available to them and more incentives and direction for the improvement of students,鈥� said Heather Zavadsky, the director of policy and communication at the Institute for Public School Initiatives in the University of Texas system.
Ms. Zavadsky has studied the impact of the No Child Left Behind Act on urban districts. 鈥淔rom a systemic view,鈥� she said, 鈥渦rban districts have been taking advantage of this data and direction to better align and organize their systems, and to build overall better systems to inform and improve instruction.鈥�
The scores were released Nov. 15 as part of the Trial Urban District Assessment, a specially collected set of test results from NAEP. Test scores for cities were reported for two previous years, 2003 and 2005.
The study produced information on urban students鈥� performance in two categories: a pool of those who live in 鈥渓arge central cities鈥� with populations of 250,000 or more, and a separate category of students from 11 individual urban districts.
Most city districts saw increases in their 8th grade students鈥� math scores from 2005 to 2007. Reading improvements weren鈥檛 as impressive nor were they universal.
SOURCE: National Center for Education Statistics
, large central cities saw their scores rise from 228 to 230 in 4th grade, and from 265 to 269 in 8th grade鈥攐n a 500-point scale鈥攂oth statistically significant increases. , the group of cities鈥� 4th grade scores increased from 206 to 208, also statistically relevant, but 8th grade reading remained flat at 250.
Individual urban districts also showed the greatest gains in math鈥攁 trend reflected in a number of national tests in recent years. Six of the 11 school systems made headway in 8th grade math from 2005 to 2007, and four improved at the 4th grade level during that time.
Performance in reading in the 11 individual districts was more stagnant鈥攁lso in keeping with recent trends. Four districts improved on their 8th grade scores from 2005. Just two of the 11 districts showed gains for 4th graders in reading over the most recent two years.
Atlanta made the largest gains of any of the 11 districts: It improved its scores in math and reading at both the 4th and 8th grade levels since 2005. The District of Columbia also raised its scores in both grades and subjects.
Atlanta鈥檚 4th graders gained 3 points on the math test since 2005, with 224 points, and 6 points in reading, to reach 207. The district also saw improvements for most subgroups, and had more students demonstrating at least 鈥渂asic鈥� skills than in previous years.
Besides those two districts, Austin, and Charlotte, the others that had their scores reported individually were Boston, Chicago, Cleveland, Houston, Los Angeles, New York, and San Diego.
鈥淭he long-term trends for the cities, in general, are very good,鈥� said Michael Casserly, the executive director of the Council of the Great City Schools, a Washington advocacy group for urban districts. 鈥淸But] we know we鈥檝e got a lot of work to do. We鈥檝e got to accelerate the gains.鈥�
Rise in Exclusion Rates
Mr. Casserly said his organization is conducting a study of the strategies used in urban districts that have posted improvements in recent years. He suggested that the strides made by Atlanta and other districts could be traced to a combination of improved professional development for teachers, stability in leadership positions, and school officials鈥� increased commitment to targeting interventions to students who need the most help.
鈥淲e鈥檝e gotten much better at using our data and at our strategies for turning around lower-performing students,鈥� he said.
The council partnered with the National Assessment Governing Board, or NAGB, the panel that sets policy for NAEP, in developing the urban assessment.
Students who took the urban NAEP were scored on a scale of zero to 500, and they were grouped in one of four performance categories, based on their results, in both reading and math: 鈥渁dvanced,鈥� 鈥減roficient,鈥� 鈥渂asic鈥� and 鈥渂elow basic.鈥�
Mr. Casserly noted that several districts saw decreases in the proportion of students in the below-basic category and increases in those scoring at the basic and proficient levels. He said he was also pleased by the improved performance of minority students in some categories.
The percentage of African-American students scoring at or above basic in 4th grade reading, he pointed out, rose from 35 percent to 41 percent from 2003 to 2007. Reading scores in large cities among Hispanic 4th graders rose from 40 percent to 44 percent over that time.
As has been the case on a number of recent NAEP administrations, many of the most significant gains occurred among struggling students. For instance, the percentage of students in large city schools scoring below basic fell, and the percentage scoring at basic rose in each category except 8th grade reading, which stayed the same.
In 4th grade math, the Atlanta district saw its proportion of students scoring below basic drop from 50 percent to 39 percent from 2003 to 2007. The percentage of students scoring proficient rose from 11 percent to 17 percent over that time.
Several districts鈥攊ncluding the District of Columbia and Cleveland鈥攈ad large increases in the proportion of students designated as having disabilities or as English learners who were excluded from taking the test since 2005.
NAGB Chairman Darvin M. Winick said that because of the variation in exclusion and accommodation policies, 鈥渃aution should be used鈥� in interpreting the new results, as with NAEP state test results. He noted that NAEP policies give states broad discretion in setting such policies, making it difficult to bring more uniformity to those standards.
The governing board has, however, tried to present the different exclusion and accommodation rates more prominently in recent years, he said.
鈥淭he first obligation we have,鈥� Mr. Winick said, 鈥渋s to get all of the information out there.鈥�
