Even more on the widening achievement gap!

Some folks have told me they don’t think the evidence so far of a widening of the gap between the “haves” and the “have-nots” is statistically significant or even real.  I will let the readers make up their own mind, but the evidence so far definitely belies the boasting, everything-is-wonderful, I-fixed-DCPS propaganda of Michelle Rhee.

Here are two tables and graphs that show how the gap is widening when you compare students in DC public schools, all big cities, and the entire nation who are at the 25th percentile with those at the 75th percentile, in the 4th grade, and in the 8th grade.

(That is, we compare the scaled NAEP math scores of students who only score better than 1/4 of their peers, with the same scores students who score better than 3/4 of their peers. Students at the 25th percentile are also said to be at the first quartile, and are not very high achievers. Likewise, students at the 75th percentile are also said to be at the third quartile; they are relatively high achievers.)

First, the fourth graders:

As before, the green line is for DC public schools. The gap is not huge here, but we do see that the gap in DC public schools seems to be getting a bit larger over the past few years.

Now, the gap for 8th graders:

Here, the recent increase in the size of the gap for DCPS between the top quartile and bottom quartile is larger than in the 4th grade, and to my unsophisticated mind, appears more significant, especially since the ones for the rest of the country look pretty stable. And it’s sad that DCPS no longer beats the average for other large cities.

In the 8th grade, in Washington, DC public schools, there are simply not enough white students for NAEP to present any statistics at all, so I don’t have anything to show you, either.

However, we do have data on income levels, as shown by whether the child’s family is deemed to be eligible for the National School Lunch Program or not. If they are eligible, that generally means a lower family income level (especially per person) than if they are NOT eligible.  First, the data for both 4th and 8th grade students across the nation, in large cities, and in DC public schools, taken directly from the NAEP TUDA report. Remember again that in 2009, charter schools are NOT included.

and, now, a table and graph showing the actual gaps in match achievement at the 4th grade level between students eligible for the national school lunch program, and those who are not:

and, for the 8th graders:

Really remarkable, isn’t it? By race, by percentile ranking, and by income level, during the last 2 years, on every single measure, the gap between the “haves” and “have-nots” has dramatically increased. But this same trend is NOT found in the nation as a whole, nor in large cities as a whole. That is, it has happened only in DC public schools, ever since Michelle Rhee (and Adrian Fenty) took over.

How and exactly has this happened????

Published in: on December 17, 2009 at 4:22 pm  Comments (9)  
Tags: , , , , ,

The URI to TrackBack this entry is: https://gfbrandenburg.wordpress.com/2009/12/17/even-more-on-the-widening-achievement-gap/trackback/

RSS feed for comments on this post.

9 CommentsLeave a comment

  1. Hmmm – Any thoughts on this as a former DCPS math teacher who had different ethnic groups in your classroom?

    The supposition in Rhee’s world is that it’s the teacher’s fault. But who’s fault is it when the same teacher has kids at various levels of ability? Again the teacher’s fault for not reaching some kids?

    The previous teachers’ fault for not getting all kids prepared for the next level?

    I wish Rhee and other educational leaders cared enough about kids to actively address what, besides teachers, affects learning. The rate we’re going, I’m afraid she’ll fire a bunch more teachers and still not have addressed the problem and just throw up her hands.

    Maybe that’s the plan. By that time all the vets will be gone, replaced by short-term teachers who don’t help the kids any more, and probably less, than the previous teachers and who aren’t tied into the union.

    So the union is busted, teacher salaries go way down and kids’ achievement problems still aren’t addressed. All that extra money goes into educational think tanks ever more determined to close the achievement gap. They’re staffed by former teachers who got their master’s tuition covered via their alternative certification programs (DCTF, TFA) during their short time as teachers.

    And remember – all this is supposedly in the best interests of the children.


    • As far as the overall gap in math achievement between various groups is concerned, I certainly saw it in my classrooms, and so did my colleagues. When students were called to the stage for being on the honor roll or on various teams, the gap was clear then, too. Part – but only a part – of the problem was that by middle school, a sizeable number (not all) of african-american kids (many of whom appeared to me to be as smart as a proverbial whip) seemed to buy into the myth that working hard, doing homework, studying, and so on, was “white” behavior that wasn’t cool. Some more cheerleading on this topic would undoubtedly help.


      • “white” behavior that wasn’t cool. Some more cheerleading on this topic would undoubtedly help.”

        Right, but that would involve not blaming teachers – a tactic that Rhee seems incapable of.

        And how would this help her break the union? Not at all, so I’m afraid it there will be no public relations campaign around this concept, even though it’s clearly “what’s best for the children.”


  2. Great blog Mr. Brandenburg! I hope you keep posting and disseminate the implications of your data analysis throughout other blogs.

    I have a couple hypotheses as to why the achievement gap widened, although I’m not sure how I’d go about testing any of these them. I haven’t really thought them out yet.

    Maybe due to the influx of DCTF/TFA teachers, DCPS experienced a change in curriculum or instructional style that was more catered toward the social learning styles of the wealthier kids. As a result, they were able to better reap the benefit of those change, but the poorer kids were left behind.

    Maybe due to NCLB’s corrective action/restructuring pressure, DCPS staff and faculty provided more test prep for kids who have a greater chance of passing the DCCAS, and those kids tend to come from wealthier backgrounds.

    Maybe due to the financial crunch that came after the real estate bubble burst, low-income kids are experiencing a higher truancy rate, thus making them miss more instructional days.

    It looks like the change occurred in 2005-2007. Were there any major policy changes then or right before that? Wasn’t the high-stakes DCCAS first implemented in 05-06?


    • It would be interesting to test that theory regarding the TFA/DCTF teachers and easy enough to do so, if Rhee would allow access to the data on achievement from their classrooms. Also would be interesting to know the % of TFA/DCTF’s who are teaching in mixed ethnic/socio-economic classrooms and what effect they are having there.

      I don’t think Rhee wants to know. Or perhaps she already knows and doesn’t want to tell.

      Re Truancy, Whatever the reasons for it, it should be fairly easy to see the changes rates among ethnic groups over the years.


    • Maybe due to NCLB’s corrective action/restructuring pressure, DCPS staff and faculty provided more test prep for kids who have a greater chance of passing the DCCAS

      Mrs. Rhee has herself boosted of doing this, aiming instruction and after school and Saturday sessions at what are known as the “bump kids.”
      They just need a bump to get over the hump.
      I wish I could remember exactly where I saw this, probably the W Post.

      (I also read someone defending this practice as these are the kids who are easier to reach/help, as they need just a little. For those 2 grade levels behind, well someday.)


  3. I have the highest regard for Guy and his teaching ability and mathematics knowledge…he has consistently led the top students in DC to national math competitions…he is part of a generation of math teachers who I deeply admire for their knowledge of mathematics and teaching ability…as a relatively young math teacher, I bemoan the fact that we have not spent more time tapping into their knowledge of mathematics to inspire students…

    We certainly could use them as curriculum consultants in DCPS…it seems a shame to just lose their depth of knowledge in mathematics…one of my biggest issues is what test prep has done to a solid coherent math curriculum…test prep and a strong math curriculum need not be mutually exclusive however they have tended to be…recently I was asked to scrap my current curriculum and teach a unit that merged probability, parallel lines and measurement conversions in the name of test prep???? It should be illegal

    As far as the disparity is concerned, it has to do not so much with race but with “middle class” values…these values precipitate things like reading in the evening and during the summer, participation in enrichment activities, consistent homework, being on time and a host of other activities outside the classroom that are conducive to academic success…I do not blame parents but note that the most successful schools in raising achievement are those that demand not just academic success but these “middle class” values


  4. One potential source of the growing 25-to-75 percentile achievement gap is related to simple selection bias in the statistics. Consider 11 students scoring 0 to 10 on some test. These students average 5 points, with the 25th and 75th percentiles at 2.5 and 7.5, making a 5 point “gap” between achievers and non-achievers.

    Consider a simple example where during the following year the school system attracts 1 achieving students who scores a 9 on the same test(this students, for whatever reason, decided join DCPS rather than attend a private or charter school). Assuming no change in the prior set of 11 students (that is, the schools got no better and no worse), the new set of students now average 5.33–hey, wait, the schools got better?–and the 25th and 75th percentiles now resemble something like 2.75 and 8.25, making the gap now about 5.5–hey, wait, the “gap” got bigger too!

    The same growing gap/improved scores would also occur with lower achieving students leaving DCPS disproportionately for charter and/or private schools. The point is that nothing happened to school quality or nor were students treated differently in this little example and yet both Chancellor and critics have fodder to toss.

    Politics aside, almost certainly DCPS has not improved as much as the Chancellor might espouse, and yet it is equally certain that selection bias also plays a key role in the seemingly dramatic “gap” changes from 2007 to 2009. I would suggest that DCPS neither improves performance as much as they have claimed, nor has DCPS done disservice to “have nots” as alleged in this blog.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: