Study Reveals Urban School Progress on State Assessments in Math, Reading;  Focus on Key Subjects Paying Off As Cities
Frequently Outpace State Gains
 
Report Also Reveals Narrowing Gaps In Math And Reading Scores For Blacks, Hispanics But Large Gaps For Low-Income & Special Ed Students,
English Language Learners

WASHINGTON - March 24, 2003 - At a time when urban schools are taking steps to raise standards, bolster course content, and improve instructional practice, many of the nation's largest urban school districts - and the high percentages of minority students they educate - are making notable gains on state assessments in math and reading, according to a report released today by the Council of the Great City Schools.

Data reveal that virtually all districts (90 percent) raised math scores in more than half the grades tested. Nearly half of the districts posted math improvement gains at a faster rate than the state average. Meanwhile, 83 percent of districts increased reading scores in more than half the grades tested with over half (51 percent) doing so at a faster rate than the state. Overall, math scores improved across 87 percent of grades tested and reading scores rose in 72 percent of grades tested. The report also noted that the rates of improvement were higher than they were two years ago.

The report, Beating the Odds III, indicates that large urban districts are narrowing achievement gaps for minority students. In math, the white-black difference in test scores was reduced in two-thirds of 4^th and 8^th grade classes and in seven in ten of the 10^th grade classes. At the same time, the white-Hispanic achievement gap in math was reduced in two-thirds of the 4^th and 10^th grades and nearly four in five of the 8^th grades. In reading, the white-black gap narrowed in four out of five 4^th grade classes, two-thirds of 8^th grade classes, and over half of 10^th grade classes while the white-Hispanic gap shrunk for nearly half of 4^th grades, two-thirds of 8^th grades, and half of 10^th grades.

But while districts are seeing improvements by race, the report reveals troubling gaps in performance for other subgroups. The report notes that gaps in math scores narrowed for only 39 percent of economically disadvantaged students, 29 percent of students with disabilities, and just 26 percent of English Language Learners. Districts fared better in reading, however, with gaps narrowing for 48 percent, 31 percent, and 32 percent of the same groups, respectively.

The report updates three years of achievement data for the 59 big-city districts belonging to the Council, broken down by year, grade level, race and other variables. It is the only national report that has thus far disaggregated data at the district level for all subgroups as required under the No Child Left Behind Act.

"Urban schools are making the kind of progress now being demanded by the No Child Left Behind law. The data suggest that improvement is possible on a relatively large scale - not just school-by-school - and that it can be accomplished under even the most challenging circumstances," says Michael Casserly, the Council's executive director and author of the report.

The report also highlights gains made by districts over time. It reveals that the percentage of districts with all grades improving in math jumped from less than half in 2000 to nearly two-thirds in 2002. The percentage of districts in which all grades have improved in math faster than the state quadrupled from 4 percent to 17 percent over the past two years. Reading gains have been more modest. The report notes that the percentage of districts with all grades improving faster than their state average almost doubled, and the percentage of large city districts with at least half of the grades improving at a faster rate than the state jumped from 34 to 50 percent over two years.

Districts Exceeding State Averages Despite Less Help and Longer Odds

The report notes that most districts highlighted in the report are making progress despite the fact that most contend with inadequate funding for high-need student populations. In CGCS districts, students are almost twice as likely to come from low-income families or to have English as a second language as the average school district. Despite these challenges, the average current per-pupil spending in Great City School districts is only $327 over the national average. Two-fifths of these districts have per-pupil expenditures below statewide averages, including New York City, Philadelphia, New Orleans, Norfolk, and others that educate more than 3.5 million students, about half of the Great City Schools' total.

According to the report, six districts - Albuquerque, Anchorage, Broward County (Ft. Lauderdale),  Hillsborough County (Tampa), Portland and San Francisco - posted math scores that met or exceeded their state averages in all grades. Meanwhile, six cities (Albuquerque, Anchorage, Greenville, Hillsborough County (Tampa), San Diego, and San Francisco) met or exceeded the statewide averages in reading in all grades tested. 

An analysis of successful districts found that many started their reforms at the elementary grade levels instead of trying to fix everything at once, and provided intensive instruction in reading and math to middle and high school students, even if it came at the expense of other subjects. The analysis also showed that successful districts:

¡     Focused on student achievement and specific achievement goals, on a set schedule with defined consequences;

¡    Aligned curricula with state standards and helped translate standards into instructional practice;

¡    Created concrete accountability systems to hold district leadership and building-level staff personally responsible for producing results;

¡    Focused on improving the lowest-performing schools by providing additional resources and training for teacher and administrators; and

¡    Adopted or developed districtwide curricula and instructional approaches rather than allowing each school to devise their own strategies. 

Methodology

The results of individual state assessments were collected by Council staff from a number of sources: state websites, reports, and databases. The report focused on reading (or language arts) and math because not all the states had assessment data on other subjects. Assessment data were then examined to determine the number of years the state had administered the tests to ensure that the report included only results that were comparable from year to year. Data were eliminated if states changed tests or significantly modified their guidelines about which students to test.