New National Assessment of Educational Progress (NAEP) results from the 2009 Trial Urban District Assessment in Mathematics are out. For the first time, the Jefferson County Public School System in Louisville participated in this special administration of the NAEP.
Unfortunately, while Louisville had strong demographic advantages compared to other participating school systems, its math performance was no better than middle of the pack, at best. For example, Jefferson County’s eighth grade whites only outscored students in three of the other 17 participating school systems, while the eighth grade blacks in Jefferson County did only slightly better, besting blacks in five other participating cities.
And, poverty was no excuse.
A map summarizing the NAEP Trial Urban District Assessment results shows that Louisville’s scores in both fourth and eighth grade were no better than average scores in the nation’s largest city school systems.
The same summary page shows that for some reported categories, Louisville students had statistically significantly lower scores than the nation’s largest cities.
In fourth grade, whites and students eligible for the federal school lunch program (often used as a measure of poverty) scored worse than their peers in the nation’s large city school systems.
In eighth grade, both whites and blacks as well as students in the lunch program scored below the large city averages.
I’m still looking at the new NAEP reports, but it is clear that Louisville had some significant advantages compared to the other cities in the new urban district score release. I obtained the following numbers using the Main NAEP Data Explorer.
Sixty percent of Louisville’s fourth grade students were eligible for the federal lunch program – the second lowest eligibility rate among the 18 large city school districts that participated. Only Charlotte had a lower rate. Thirteen of the eighteen participating systems had a lunch eligibility rate at least 10 points higher than Louisville’s. In other words, Louisville is comparatively wealthy compared to the other participants.
Louisville’s eighth grade lunch eligibility rate situation was virtually identical. Only Charlotte had less poverty, and thirteen of the eighteen participating systems had a lunch eligibility rate at least 10 points higher than Louisville’s eighth graders.
Louisville also should also have enjoyed a strong advantage because of the relatively high percentage of whites that attend its schools.
Whites score higher than other racial groups. No other participating district exceeded Louisville’s 53 percent white rate in fourth grade. In fact, the next closest district, Charlotte, only had 36 percent whites in its testing sample. Things were a bit more spread out in eighth grade where Louisville again had the leading percentage of whites in the testing mix – 55 percent. Charlotte was again second, with an even lower 32 percent whites in its testing sample.
These numbers are not surprising given the Jefferson County School System’s large geographic size that encompasses large suburban areas along with inner city areas. In contrast, most large city school systems do not contain significant upscale suburbs-like areas.
Based on its advantages in both poverty rates and its racial mix, I expected Louisville to walk away with the Trial Urban District Assessment. It didn’t.
I’ll have more once I get a chance to review the extensive reports available.
Tuesday, December 8, 2009
Federal testing says Louisville’s schools no better than in nation’s largest cities
Subscribe to:
Post Comments (Atom)
2 comments:
Black students in Jefferson County scored higher in 8th grade mathemtaics than their peers in Detroit, the District of Columbia, and Milwaukee. They scored lower in 8th grade mathematics than Black students in Austin, Boston, Charlotte, Houston, Miami-Dade, New York City, and San Diego. There was no significant difference beween Jefferson County Black students in 8th grade mathematics and their peers in Atlanta, Baltimore City, Chicago, Cleveland, Fresno, Los Angeles and Philadelphia.
White students in Jefferson County scored higher in 8th grade mathematics thay their peers in Cleveland and Milwaukee. They scored lower in 8th grade mathematics than White students in Austin, Boston, Charlotte, Houston, Miami-Dade, New York City, and San Diego. There was no signifance difference between Jefferson County White students in 8th grade mathematics and their peers in Chicago, Fresno, Los Angeles, and Philadelphia. The samples of White students in Atlanta, Baltimore City, Detroit, and the District of Columbia were of insufficient size to provide reliable scores.
Anonymous December 8, 2009 3:27 PM ran the statistical significance tests for the new TUDA data, which helps point out a notable limitation in the NAEP, which is a sampled assessment.
When the NAEP samples get smaller, it takes a bigger score difference to show a statistically significant difference in the scores.
The sample sizes for the TUDA were only about half the size of the samples usually taken for a full state, so the scores required to detect a real difference in performance are notably larger than for the state assessments.
That makes the NAEP a somewhat blunt measuring instrument when you try to compare scores across different school systems and states.
Anyway, as Anonymous shows, we can only say with confidence that Jefferson County's whites outscored just two of the other participating school systems.
The district's blacks only outscored three other districts, one being the notoriously under-performing Washington, DC school system.
As I said, Jefferson County's performance is middling or below, at best, even though its racial mix and relatively low poverty rate should have given Kentucky's largest school system a real advantage in comparison to the other school systems in the TUDA.
Post a Comment