On Sunday, the New York Times published an op-ed critical of school choice policies in Washington, DC. As Matt Yglesias very fairly pointed out, the author, Natalie Hopkinson, failed to cite student achievement data to back her claim that residential segregation and the expansion of the charter school sector have left many DC families with only “mediocre” public school options. Thankfully, on Wednesday the federal government dumped a mass of new NAEP test score data from the nation’s largest cities, including DC. NAEP, the National Assessment of Educational Progress, is the gold standard in education research: the only exam overseen by the federal government and adminstered to a random selection of schoolchildren annually, with no rewards or punishments attached to corrupt the scores.

I spent this morning diving into the new NAEP numbers in an effort to paint a more complete picture of student achievement in DC over the past decade, which included the three tumultous years of Michelle Rhee’s chancellorship. While white, black and Hispanic children all made modest test score gains in DC since 2003, the Rhee agenda has not significantly narrowed achievement gaps between the various demographic groups, nor has it brought disadvantaged DC youth up to the national average scores for peers of their same race and class in other cities.

I’m going to focus this post on fourth grade math, since it seems to be the subject and grade level most suspectible to reform efforts. In DC since 2003, the black/white score gap remained constant, the poor/non-poor gap grew, and the Hispanic/white gap closed slightly.

Achievement gaps would be less disturbing in and of themselves if overall achievement levels were moderate or high. But what we continue to see in DC is that white students score well above both national and urban district averages for their race; black, Hispanic and poor children score well below national averages for their races and classes. This makes DC the city in the nation with the largest black-white student achievement gap.

If you are a white or middle-class family living in Washington, your child will likely attend a socioeconomically segregated neighborhood school or a higher-quality magnet, and will outperform her peers in suburban public schools. If you are a poor parent of color, on the other hand, your child will do worse in the DC public schools than he likely would have done in other urban or suburban districts. The NAEP data does not include charter school students, but the Washington Post reports that black and Hispanic children in DC charters score better on standardized tests than their traditional-school peers. The bottom line, however, is that there are far too few seats in high-quality charter schools to serve every disadvantaged child in Washington, and those left behind at neighborhood public schools continue to be shortchanged. Indeed, the district’s own data show experienced and highly-rated teachers more clustered than ever in affluent neighborhoods and schools.

In short, Michelle Rhee and her successor, Kaya Henderson, have presided over a landscape of modest raw test score gains. Meanwhile, the expansion of school choice in DC encouraged more white and middle-class families to send their children to public schools, and provided an escape route to some poor children who would otherwise have attended failing neighborhood schools. What Rhee and Henderson have not done is shrink ahievement gaps or guarantee to DC low-income parents that their children will receive at least an adequate, “average” education, no matter what school they attend.