Skip Navigation

Mark Schneider
Commissioner, National Center for Education Statistics

National Assessment of Educational Progress
NAEP 2005 Mathematics and Reading Trial Urban District (TUDA) Results

December 1, 2005

Good morning. My name is Mark Schneider. I am the Commissioner of the National Center for Education Statistics. I am here today to release findings from the National Assessment of Educational Progress 2005 Trial Urban District Assessment for both reading and mathematics - what we call the NAEP TUDA assessments.

Background of the Trial Urban District Assessment (TUDA)
The TUDA assessments are the newest part of NAEP. As the name implies, these are trial assessments, and they have not yet been made a permanent part of NAEP. NAGB and NCES worked with the Council of Great City Schools to create the first TUDA assessment in 2002, in reading and writing. The urban school districts included in these assessments volunteered to participate.

A primary purpose of NAEP is to provide a common yardstick for comparing results among jurisdictions. While the states and some school districts use their own tests to measure their students' performance, these tests differ from each other and it is difficult to compare their results. The NAEP assessments given at the national, state, and urban district levels are the same in each subject, at each grade tested.

In 2002, five districts participated-Atlanta, Chicago, Houston, Los Angeles, and New York City. For comparison purposes, we also include results for the District of Columbia, which participates in the NAEP state assessments. In 2003, TUDA assessments were given in reading and mathematics in nine districts: in addition to the 2002 participants, Boston, Charlotte-Mecklenberg, Cleveland, and San Diego also participated. In 2005, Austin was added, and the assessments were given in reading and mathematics. The sizes of the districts' student populations range from about 50,000 in Atlanta to about one million in New York City.

Overview of the 2005 Trial Urban District Assessments
We administered TUDA as part of the regular NAEP assessment, with a range of 1,100 to 2,400 students per district per grade being sampled in each subject. We report NAEP results in terms of both scale scores and percent at or above achievement levels. Scale scores for both the fourth and eighth grades are presented on a 0 to 500 point scale. These scores measure what students know and can do. Results are also reported according to three achievement levels established by the National Assessment Governing Board. These levels set standards for what students should know and be able to do. The three achievement levels are defined as Basic, Proficient, and Advanced.

We compare the performance in the 11 districts with the average scores and achievement level percentages for fourth- and eighth-graders in public schools nationally and for those attending public schools in what we call "large central cities." NCES defines "large central city" as a central city with a population at or above 250,000. Examples of large central cities are Lexington, Kentucky and Philadelphia, Pennsylvania.

The urban school district students differ in demographic composition from the nation as a whole. For example, in the reading assessment at grade 4, the public school populations of the 11 TUDA districts were 60 percent or more nonwhite. In comparison, the nation's public school population was about 43 percent nonwhite.

Urban districts vary widely in their percentage of students with special needs. Thus, caution should be used in interpreting results and making comparisons. In reading at grade 4, for example, the percentage of all students who were identified as either students with disabilities or English language learners ranged from 11 percent in Atlanta to 59 percent in Los Angeles, and the national percentage for public schools was 23 percent.

Some of these students were assessed with accommodations, such as being tested in small group sessions or being given extra time, and some could not participate and were excluded. The percentage of all fourth-graders who were assessed in reading with accommodations ranged from 2 percent in Houston to 16 percent in New York City, with a national public school percentage of 7 percent.

The percentage of all fourth-graders who were excluded from the reading assessment ranged from 4 percent in Atlanta and Charlotte to 23 percent in Houston, with a national percentage of 7 percent. Exclusion rates were generally lower in the mathematics assessments because of the availability of additional accommodations such as spanish translations and having test questions read aloud.

Overview of Results
I will first present an overview of results for each grade, over time and in 2005.

  • Overall Findings-Grade 4
    In mathematics, both average scale scores and percentages at or above Basic increased for fourth-graders in 8 of the 10 districts that participated in both 2003 and 2005. Six of these districts also experienced increases at or above Proficient. For reading, from 2003 to 2005 one district increased in percentage at or above Proficient. In 2005, in mathematics, five districts scored above the large central city average, and five were below. In reading, four were above and five were below the large central city average score. In comparison to the nation, most, but not all, trial urban districts had lower average scores in mathematics and reading.
  • Overall Findings-Grade 8
    At the eighth grade, the results were as follows:

    In mathematics, performance improved from 2003 to 2005 in four districts. In reading, one district improved its average score. In 2005, in mathematics, four districts were above the large central city average score, and five were below. In reading, as in math, four were above the large central city average score, and five were below. Compared to the nation, nine districts scored lower in mathematics; these same districts also scored lower in reading. Detailed results are presented in the next section of the presentation.

Mathematics
I will now present a closer look at the fourth- and eighth-grade results in mathematics, and then go on to reading.

It should be kept in mind that NAEP assessments are based on samples and with samples there is a margin of error associated with each score. When we make comparisons between scores, or compare the percentages of students at the various achievement levels, we test the differences to see if they are larger than the margin of error involved-differences that we call statistically significant. These differences are indicated by asterisks or arrows in the tables and charts in the reports.

  • Mathematics Score Results, Grade 4
    Between 2003 and 2005, mathematics scores for public school fourth-graders increased in 8 of the 10 districts that participated in both years. These districts were Atlanta, Boston, Cleveland, the District of Columbia, Houston, Los Angeles, New York City, and San Diego. These same eight districts experienced increases in the percentage of students at or above Basic, and six of them also had increases in the percentage at or above Proficient. Scores in two districts-Charlotte and Chicago-did not show a significant difference. Austin, as noted previously, did not participate in 2003.

    In the report, results for the TUDA districts are compared to those for the nation and large central cities. Because of certain changes that were made in the identification of the schools included within specific large central city areas, we cannot make comparisons over time using large central city results.

    In 2005, two districts had average scores above the nation-Austin and Charlotte. The remaining nine districts scored below the nation. In comparison to the large central cities, five districts-Austin, Charlotte, Houston, New York City, and San Diego-scored above the large central city average. Six districts-the same five plus Boston-had higher percentages at or above Basic. Three districts-Austin, Charlotte, and San Diego-also had higher percentages at or above Proficient. Five districts-Atlanta, Chicago, Cleveland, the District of Columbia, and Los Angeles-scored below large central cities, and the same five had lower percentages of students who performed at or above Basic and Proficient.
  • Score Changes by Race/Ethnicity, Grade 4: 2003-2005
    Examining scores by race/ethnicity over time, average scores did not decline for any group in any district. White students' scores increased in three districts-Boston, Houston, and San Diego; scores for Black students increased in four districts-Atlanta, Boston, Cleveland and the District of Columbia; Hispanic performance increased in 6 districts-Boston, the District of Columbia, Houston, Los Angeles, New York City, and San Diego; and performance of Asian/Pacific Islander students improved in 2 districts-Boston and San Diego. No trend comparisons could be made for some race/ethnic groups in some districts, due to lack of reportable data.
  • White-Black Score Gap, Grade 4: 2005
    I will now compare the 2005 average scale scores of White and Black fourth-graders in mathematics.

    In many urban districts, the size of the White student population is quite small, which leads to large margins of error. Thus, while there have been improvements in minority performance in several districts, most gaps between White and minority students have not narrowed sufficiently over time to exceed the margin of error. In 2005, at grade 4, White students had higher mathematics scores than Black students in every TUDA district. The gap nationally was 26 points, and for large central cities it was 31 points. Gaps for the 11 districts ranged from 18 to 58 points.

    Note that the size of the gap does not necessarily reflect the level of performance. For example, the disparity in Cleveland is smaller than in the District of Columbia, but White students performed higher in the District of Columbia than in Cleveland.

  • White-Hispanic Score Gap, Grade 4
    White students have a higher average score than Hispanic students in each of the 9 districts with a sufficient Hispanic sample to produce reliable scores. The gap nationally was 21 points, and for large central cities it was 24 points. Gaps for these nine districts ranged from 18 to 51 points.
  • Mathematics Score Results, Grade 8
    Between 2003 and 2005, mathematics scores for public school eighth-graders increased in Boston, Houston, Los Angeles, and San Diego. These districts also experienced increases at or above Basic and Proficient. None of the other districts that participated in both years experienced significant changes.
  • Score Comparisons to the Nation, Grade 8: 2005
    The same two districts as at grade 4, Austin and Charlotte, had average scores above the national average. The remaining nine districts scored below the nation.
  • Score and Achievement-Level Comparisons to Large Central Cities, Grade 8: 2005
    Four districts-Austin, Boston, Charlotte, and San Diego-had average scores above that for large central cities; five districts-the same four plus Houston-had higher percentages of students who performed at or above Basic; and three districts-Austin, Boston, and Charlotte-had higher percentages at or above Proficient. In five districts-Atlanta, Chicago, Cleveland, the District of Columbia, and Los Angeles-the average score was below that for large central cities; the percentages of students who performed at or above Basic was lower than that for large central cities in the same five districts; and the percentage at or above Proficient was lower in six districts-the same five districts plus Houston.
  • Score Comparisons to Large Central City for Students Eligible for Free/Reduced Price School Lunch, Grade 8: 2005
    The following results compare the performance of eighth-graders eligible for free or reduced price school lunches in each district to the average score for such students in large central cities. For analytic purposes, we use eligibility for the National School Lunch Program as a proxy for poverty.

    Eligible students in five districts-Austin, Boston, Charlotte, Houston, and New York City, had an average score higher than that for large central cities. Eligible students in four districts-Atlanta, Cleveland, the District of Columbia, and Los Angeles-had a lower score than their counterparts in large central cities.

Reading
The fourth- and eighth-grade results in reading are as follows.

  • Score Results, Grade 4: 2002-2005
    For reading, there are results for three assessments, although not all districts participated in each year. At fourth grade, scores in Atlanta and New York increased from 2002 to 2005. From 2003 to 2005, there were no significant changes in any district. The nation as a whole did score higher in 2005 than in 2003.
  • Score Comparisons to the Nation, Grade 4: 2005
    The score in Charlotte was above the national average. Nine districts scored below the nation. And Austin's score did not differ significantly from the nation.
  • Score and Achievement-Level Comparisons to Large Central City, Grade 4: 2005
    Compared to students in large central cities, those in four districts-Austin, Charlotte, Houston, and New York City-had higher average reading scores; three districts-Austin, Charlotte, and New York City-had higher percentages of students at or above Basic; and two districts-Austin and Charlotte-also had higher percentages at or above Proficient.

    Five districts scored lower than large central cities-Atlanta, Chicago, Cleveland, the District of Columbia, and Los Angeles; the same five districts had lower percentages of students at or above Basic; and five districts-Boston, Chicago, Cleveland, the District of Columbia, and Los Angeles-had lower percentages at or above Proficient.
  • Score Results by Race/Ethnicity, Grade 4: 2005
    The average reading scores for White students ranged from 209 to 253 on the scale of 0 to 500. Note that within each group, a number of these urban districts performed the same as or higher than that group's averages for the nation and large central cities, and that these patterns were repeated across subjects and grades. For example, in reading in grade 4, White students in five districts performed higher than the large central city average for White students, those in five districts performed on a par with the large central city average, and one district scored below.

    Black students' performance, on average, ranged from 187 to 207 in the TUDA districts. Black students in four districts performed higher than the average for Black students in large central cities, those in four districts performed on a par with the large central city average, and those in three districts performed below.

    The range of Hispanic average scores is similar to that for Black students-190 to 209. Here again, Hispanic students in three districts performed above the large central city average for Hispanic students, those in six districts performed on a par with the large central cities average, and one performed below.

    Also note that the gaps in performance between White and Black students, and between White and Hispanic students, were substantial, and these differences have not lessened over the period of the TUDA assessments.
  • Score Results, Grade 8
    In reading at grade 8, the average for Atlanta increased between 2002 and 2005. The average score for Los Angeles increased from 2003 to 2005. There were no decreases among the districts.
  • Score Comparisons to the Nation, Grade 8: 2005
    Compared to the nation, nine of the eleven districts scored below the nation in 2005. The average scores in Austin and Charlotte were not significantly different from the nation.
  • Score and Achievement-Level Comparisons to Large Central City, Grade 8: 2005
    Four districts-Austin, Boston, Charlotte, and San Diego-had average scores above those of the large central cities; Charlotte had a higher percentage at or above Basic; and three districts-Austin, Boston, and Charlotte-had higher percentages at or above Proficient.

    Five districts-Atlanta, Cleveland, the District of Columbia, Houston, and Los Angeles-scored below the large central cities; the percentage at or above Basic was lower in four districts-Atlanta, Cleveland, the District of Columbia, and Los Angeles; and the percentage at or above Proficient was lower in five districts-the same four plus Houston.
  • White-Black Score Gap, Grade 8
    Turning to the performance of the racial/ethnic groups in reading at grade 8, White students, on average, had higher scores than Black students in every district except Cleveland and Atlanta. The disparity in reading scores between White and Black students nationally was 27 points in 2005, and for large central cities it was 30 points. Gaps for the 11 districts ranged from 19 to 66 points. Between 2003 and 2005, the mathematics gap between Black and White students at grade 8 widened in Houston.
  • White-Hispanic Score Gap, Grade 8
    The score difference between White and Hispanic eighth-graders in reading was 24 points nationally in 2005, and for large central cities it was 26 points. Gaps for the 10 districts with a sufficient Hispanic population to produce meaningful scores ranged from 7 to 53 points. The size of the difference narrowed between 2003 and 2005 in Los Angeles.

For More Information
That completes my overview of results from the 2005 TUDA mathematics and reading assessments. There is much more information in the Report Cards themselves. In addition, the NAEP initial release website-http://www.nationsreportcard.gov-contains extensive information on the performance of students in each district, access to released assessment questions, and use of the NAEP Data Explorer, our new online data analysis tool.

Powerpoint presentation for The Nation's Report Card - Trial Urban District Assessment Mathematics and Reading 2005 (PPT 3,392KB)

In closing, I would like to thank the schools and their students who participated in these assessments. I would also like to thank my staff and the contractors who made these assessments a success.

Thank you.