Michelle Rhee said on the recent Frontline PBS special that there were ‘dozens and dozens’ of DC public schools that supposedly made steady progress on the DC-CAS over the past four or five years.
Jay Mathews only found 13 schools which did what Rhee claimed: Brent, Eaton, Murch, Oyster-Adams, Payne, Plummer, Prospect, Ross, Thomson, Tubman, Hart, McKinley Tech, Sousa.
A commentator by the name of ‘LetsBeReal’ pointed out that a large fraction of those schools in fact were populated mainly by relatively affluent white students or were schools with selective admissions: Brent, Murch, Eaton, Oyster-Adams, Ross, and McKinley Tech.
I looked at the grade-by-grade proficiency ratings at the remaining schools and found a LOT of very suspicious rises and falls in proficiency rates for same-cohort groups from year to year, in all but one of those schools: Hart.
Here is what I found:
At Payne, the cohort that was in the 4th grade in 2012 went from 38% proficient in reading in 2011 to 55% proficient the next year, a 17-percentage-point rise, which means (to me) that it should be flagged. Either the teacher last year was doing something so wonderful that entire books should be written on how to replicate those feats, or there was cheating. Same group went from 14% proficient in math in 2011 to 60% proficient last year. Unbelievable, frankly.
The cohort that was in the 5th grade at Payne in 2012 had DC-CAS reading proficiency rates since 3rd grade of, respectively, 17%, 23%, and 46%. In math, the scores for that same cohort, by year were 23%, 35%, and 50%. Either amazingly good or brazen cheating, one or the other: in any case, it needs to be checked out.
The Payne cohort that was in the 5th grade in 2011 had DC-CAS reading proficiency rates of starting in the 3rd grade of 52%, 31% and 50%. In math, 58%, 34%,, and 31%: very suspicious as well.
Still at Payne, the cohort that was in the 5th grade in 2010 had DC-CAS proficiency rates in reading of 41%, 32%, and 63%, which is again unbelievable. The math scores were a lot steadier: 24%, 32%, and 33%.
At Plummer, I also found a lot of suspicious rises and falls. Cohort in 4th grade in 2012 in reading went from 31% in the 3rd grade to 68% proficient the next year. In math, the same group went from 38% to 81%. The cohort that was in the 5th grade in 2012 went from 40% to 18% to 46% proficient in reading over three years, which is unbelievable. In math their scores were much more believable: 36%, 32%, and 40%.
Still at Plummer, the cohort that was in the fifth grade in 2011 had reading scores that went like this over their three years there: 44%, 33%, and 36%. In math, their scores were 60%, 18%, and 29%. The cohort that was in the fifth grade in 2010 had reading scores of 47%, 22%, and 30%. In math, they were 37%, 13%, and 33%.
Prospect LC is a special education school, where teachers are apparently able to rewrite the DC-C AS to fit their students’ needs and abilities. (At Sharpe Health school, where students are often unable to walk, feed themselves or clean themselves, between 95% and 100% of the students are supposedly “proficient” or “advanced”, but that doesn’t mean the same thing as it does in other schools.) In any case, at Prospect, I found one cohort (the group that was in 7th grade last year) whose proficiency ratings went from 25% to 0%, then 0% again, followed by two years of 5% — in reading. In math, that same cohort went from 13% to 31% to 0% to 24%, hardly reassuring. The cohort one year older had reading pass rates of 0%, 0%, 0%, 6%, and 0%. In math, their pass rates were 0%, 0%, 11%, 29%, and 25%. (I’m not making this up, as Dave Barry used to say.) And the cohort that reached the 8th grade in 2011 had pass rates in reading of 0%, 0%, 0%, and 17%.. In math, the same group had pass rates of 0%, 0%, 17%, and 42%.
At Tubman, to summarize, I found nine cases where proficiency rates jumped or fell by more than ten percentage points from one year to the next from 2008 through 2012.
At Sousa, I “only” found seven such cases.
At Thompson, I found eight suspicious rises and falls.
At Hart, I only found one suspicious rise, but if Rhee thinks that going from about 12% proficient overall to about 25% proficient is wonderful progress, then I don’t know quite what to say, given that Rhee herself bragged — falsely — on her resume that she brought an entire class or two of students in Baltimore from below the 13th percentile to above the 90th percentile, using methods that she has never shared publicly.
Both Erich Martel and I feel that a ten-percentage-point rise or fall raises a red flag. Just possibly, such a rise would demonstrate tremendous teaching. However, from our own experience, it’s much more likely the result of cheating. In any case, it needs to be checked out – but not by “See No Evil” Inspector General Charles Willoughby.
In any case, if these are the only schools which Jay Mathews found that had sustained gains, and if these schools fall into two groups: schools with mostly non-poor students and relatively large fractions of white students on the one hand; and schools with many very many suspicious rises and falls in cohort proficiency rates – with the single exception of Hart, then I think we can say pretty clearly that Michelle Rhee and Kaya Henderson have a pretty clear legacy:
Abetted by fraud and deception.
You can find the school-by-school, grade-by-grade proficiency ratings at http://nclb.osse.dc.gov/index.asp for 2008 through 2011. Unfortunately, OSSE still has not released the grade-by-grade scores for 2012, but I was leaked a spreadsheet containing that data. If you would like to see it for yourself, I have posted it on Google Drive, here:
I will shortly post tables containing the exact numbers so you can see what I’m talking about.