More on the D.C. Achievement Gap and Michelle Rhee’s Legacy

In response to my Nation piece on achievement gaps in Washington, D.C. district public schools, commenter E.B. wondered how things would look different if we measured student proficiency instead of raw NAEP scores. This is a great question, since proficiency–defined as "solid academic performance"–is the standard to which we should hold most children.

As you can see from the chart I've whipped up below, things still look pretty abysmal when we measure proficiency instead of raw achievement. While D.C. public school students from every demographic group made modest gains over the past four years, just a small minority of black (12 percent), Hispanic (22 percent), and poor children (11 percent) in Washington perform at grade level in math. I've included scores from Charlotte, North Carolina as a comparison, since Charlotte typically outperforms other urban districts.

When we compare Charlotte to D.C., we see that demographics play an important, but ultimately limited, role in a child's academic performance. Poor, black, and Hispanic students do better in Charlotte than they do in D.C. There are many reasons why this is so, starting with "peer effects:" The Charlotte district is more diverse than DCPS, with a greater percentage of white, Asian, and middle-class students, as well as individual schools and classrooms that are more socioeconomically-integrated. (To review why integrated schools are often better schools, click here and here.) There are also an almost-unlimited number of curricular, pedagogical, and human resources practices that could be responsible for one district, like Charlotte, outperforming another, like Washington, D.C.

I want to be clear–especially in response to Alexander Russo–that I'm not attempting to hold Michelle Rhee responsible for the existence of these achievement gaps, which far predate her term and are partly attributable to demographic realities out of her control. What I do want to do is call attention to the continued underperformance of disadvantaged D.C. kids compared to their peers in other cities. The Rhee agenda was multi-faceted. It included elements I support, such as bureaucratic streamlining and the recruitment of more college-educated families into the public school system, and elements about which I am skeptical, such as the tying of teacher evaluation and pay to student standardized test scores. The Rhee years also conincided, as Alexander notes, with an increase in charter school enrollment, and there is some limited evidence that D.C. charters may be outperforming the city's traditional schools.

The takeaway, I think, is that Rhee pursued a number of reforms, but there is no evidence that her most controversial, anti-union moves are responsible for the limited growth we've seen–or that teacher-evaluation reforms alone can, over time, move many more poor, black, and Hispanic D.C. chidren to academic proficiency. Indeed, the D.C. public schools are still highly segregated by race and class, with quality teachers more clustered than ever in whiter, wealthier schools. These trends are negatively correlated with high academic achievement for disadvantaged kids.

6 thoughts on “More on the D.C. Achievement Gap and Michelle Rhee’s Legacy

  1. Jersey Jazzman

    Dana, I like the work you’re doing here very much, but I have to disagree with something: I don’t believe it is realistic to hold ALL children to a “proficient” standard on the NAEP. “Proficient” has a specific contextual meaning on the NAEP:

    From Diane Ravitch’s Death and Life of the Great American School System:

    The term “proficiency” – which is the goal of the law – is not the same as “minimal literacy.” The term “proficiency” has been used since the early 1990s by the federal testing program, the National Assessment of Education Progress, where it connotes a very high level of academic achievement. (p. 102)

    link to

    That said, I think your analysis is still solid. And if we shouldn’t blame Rhee for this, she sure as hell shouldn’t take credit for it either:

    link to

  2. D

    Thanks for this additional analysis, Dana. I’m appealing to others in the media, including Alexander Russo, to do the same. I find it startling that after the media hoopla crediting Rhee’s bold reforms with every score increase while she was in DC (despite greater gains before she came and evidence of widespread cheating during her tenure), there are reservations about crediting her and her former deputy and successor Kaya Henderson with the obvious dearth of positive results almost five years into their huge reform efforts.

    Also, rather than just commenting on the many factors that could have improved achievement in Charlotte, it seems that education analysts and reformers who truly care about students would be obliged to want to take a closer look at factors such as teacher characteristics – which Rhee and Henderson emphasized so much in DC. For instance, compare teacher experience and educational levels, type and perceived quality of teacher evaluation (tied to student scores?) teacher turnover, percentage fired vs. quitting or retiring, teacher satisfaction, the number and quality of PD opportunities. Beyond studying teachers, analysts could study central office operations – staff experience level and salaries, use of consultants, programs implemented, etc.

    The current caution about criticizing and analyzing school reform is in stark contrast to the license with which many of these bold but untested reforms were promoted just a few years ago. If school reform is really about improving student achievement, the focus should shift from praising or protecting the reformers to figuring out how to help the children.

  3. Dana Goldstein

    Thank you JJ and DD for the smart comments. For more detailed information on teacher characteristics in DC, check out this post. There has been high turnover, and only about half the people who leave have poor evaluation scores. In other words, a large number of effective teachers are leaving the district.

    link to

  4. D

    Dana — DD or D is actually efavorite – don’t know why my yahoo ID is not showing up properly.

    I’ve seen your churn and burn post — very nice. Now if you can get similar teacher info on Charlotte and compare, it could provide really useful insights. If Charlotte’s teacher data is similar to DC’s, it will not be helpful or suggest that teachers don’t matter much – that something else matters more. If Charlotte has much higher teacher retention, experience, etc, then it suggests (but doesn’t prove) that DCs pattern is not conducive to higher student achievement.

    But you can bet that if DC’s scores had significantly improved, Rhee’s slash-and-burn approach would have gotten credit for it. It seems like the press, along with the reform establishment, is more ready to praise a hero than it is to assess the facts.

    Hopefully, that is finally changing. The fact that Duncan is not out trying to make lemonade out of lemons and the press is largely quiet about the NAEP results are good signs. The next steps are reality-based assessment and criticism.



Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>