On New York State Tests, A Growing Achievement Gap Between White/Asian and Black/Hispanic New York City Students
Here's why: The achievement gap in New York City has increased in the last five years, and the decreases in the achievement gap in grade 8 ELA have come at the expense of white and Asian students. Coupled with my analyses of NAEP achievement gaps - which also showed no progress and in some cases growing gaps - these findings are quite troubling.
Over the period 2003 to 2008, New York state tested only 4th and 8th graders continuously. Here's what I found:
* In 4th grade ELA, the black-white, Hispanic-white, and black-Asian gaps have all grown. The black-white gap has increased by 13%, the Hispanic-white gap by 6%, and the black-Asian gap by 2%. The Hispanic-Asian gap has narrowed by a measly 4%.
* In 8th grade ELA, gaps have decreased - but only because the average scale scores of Asian and white students fell between 2003 and 2008. In contrast to Bloomberg's claims of a 50% reduction in some cases, the black-white gap has only decreased by 12%, the Hispanic-white gap by 3%, the black-Asian gap by 21%, and the Hispanic-Asian gap by 13%.
* In 4th grade math, black-white, Hispanic-white, black-Asian, and Hispanic-Asian gaps have all grown. The black-white gap has increased by 7%, the Hispanic-white gap by 5%, the black-Asian gap by 15%, and the Hispanic-Asian gap by 13%.
* In 8th grade math, the black-white, black-Asian, and Hispanic-Asian gaps have all grown. The black-white gap has increased by 4%, the black-Asian gap by 28%, and the Hispanic-Asian gap by 21%. Only the white-Hispanic gap has decreased by 6%.
It's hard to imagine how the NYC Department of Education will finesse these results. Rest assured that they will try. So sit back, grab some popcorn, and get ready for some serious spin. As I suggested below, an apt slogan might be, "The New York City Department of Education: It depends on what the meaning of the word 'is' is."
You can find the scale scores, and the race-specific standard deviations, here. Analysis-wise, here's what I did: I calculated the point in the white distribution that the average black student scored: i.e. (White average score for 2003 - Black Average score for 2003)/White standard deviation. I performed the same calculation for all of the gaps above for both 2003 and 2008. If you have questions, as always, email me at eduwonkette (at) gmail (dot) com.