(Guest Post by Matthew Ladner)
Look ma! DCPS reform is working great as long as your parents graduated from college. The below charts are district schools only:
That’s a two point decline compared to a 17 point (!) improvement for those scoring at home. What about math?
That is a 10 point improvement for students whose parents did not finish college compared to a 25 point (!) improvement for students whose parents did finish college. You see a similar pattern if you break the numbers down by race/ethnicity. Here is how DCPS looks (absent charter students) compared to the TUDA districts (also district only) for Black students in NAEP 8th grade reading:
Here is how it looks for Hispanic students:
And…here is how it looks for White students:
Jay P. Greene's Blog 
Matt, couldn’t much of the change be attributed to demographic changes in DCPS? Lots of African American and Latino students have switched to public charter schools in DC over the past decade. These are bienniel snapshots of different populations of students.
Strolling around the obviously gentrifying streets of DC certainly inclines me to think that it is playing a role. Kids are moving from DCPS to charters, but it doesn’t seem likely to be driving up the kids with college grad scores does it?
As a native DCer my hometown is the poster child for urban gentrification; my last time home it shocked me seeing streets I knew as the epicenter of depravity now hot spots for the hip. The academic failings of Black students in DCPS is older than Marion Berry’s reign and when the Redskins competed for Super Bowls. Black students still attend Dunbar, Cardozo, McKinley, Eastern, Anacostia, etc., high schools, and the elementary-middle schools that feed them. For decades these students have been failed, in spite of one of the highest ADAs in the nation.
I don’t have time to check all the graphs, but the one for NAEP Grade 8 Math in TUDA districts in 2017 doesn’t match up with one I generated in an analysis of Jefferson County, KY Schools. Here is a link to the paper: http://www.bipps.org/wp-content/uploads/2018/05/JeffCo-2017-NAEP-for-All-Students-Whites-Only-Blacks-Only-R1.pdf.
Also, the NAEP has statistical sampling errors and some of the scores shown are just statistical ties, as you will see when you read my paper.
Still, the main point of the article is sustained for white minus black Grade 8 reading achievement gaps. Per the NAEP Data Explorer, the gap in DC is statistically significantly larger than the gap in any other TUDA district that had scores reported for both groups.
I was using the main NAEP rather than TUDA. I don’t think you can slice out the charter scores in TUDA.
The redesigned NAEP Data Explorer apparently incorporated the TUDA results into a single database with everything else. You can separate charter and non-charter scores by race with this using the cross-tab feature after selecting the districts.
In fact, I used that to assemble these two blogs you might find of interest: http://www.bipps.org/charter-school-shocker-cleveland-charters-much-better-black-students/
and
http://www.bipps.org/new-naep-tells-us-atlantas-charter-schools/
Is there still a separate data explorer for TUDA only?