USA Today: Possible testing fraud under Rhee in DC (UPDATED)
USA Today recently published an article about their investigation of reported testing irregularities in the DC Public Schools. Independent statisticians and testing experts cited in the investigation find that wrong-to-right-answer erasure patterns in DC schools like the celebrated Crosby S. Noyes Education Campus are too frequent and systematic to have happened by chance. Such patterns have to be four standard deviations above normal to even trigger a flag from the testing companies.
For example, from 2006 onward, between 75% and 80% of Noyes’ classrooms were flagged for unusually high rates of erasures, where wrong answers were changed to correct ones. In one Noyes classroom, seventh graders averaged 12.7 wrong-to-right conversions; the average DC seventh grader had less than one such change. Statisticians say the possibility of erasure rates as high as these happening purely by chance was less likely than “winning the Powerball grand prize.”
I urge everyone to read the full USA Today story as well as the documents from their investigation (I’ll be writing more about that sometime this week). As you do, keep two things in mind:
- Michelle Rhee and others higher-ups in the DC school system have elevated test score Data to an insane degree. They’ve expanded their careers– and ended others’ careers– on the basis of the gains they claim to have made. If all decisions in DCPS are to be guided by “robust data,” shouldn’t there be a robust system in place for ensuring transparency, propriety, and accountability in the administration and scoring of the tests that generate that data? And shouldn’t there be the same kind of accountability for those who might have tampered with such important information as there is for those who fall short by these (questionable?) measures?
- DC’s mayoral control system has essentially eliminated the possibility for independent oversight of district behavior. When these problems first surfaced, investigators noted that they experienced significant push-back from high-level district officials– which is to be expected given how invested they all were in the appearance of immediate, significant success. From the USA Today article:
When D.C. administrators resisted investigating the 2008 scores, there was no counterweight to force the issue. The state board is empowered only to advise OSSE. Mary Lord, a board member with a teenager who attends a D.C. high school, is critical of the decision not to investigate the 2008 scores. “If you are going to add all this weight” to testing, “hanging the principals’ reputations … and the teachers’ pay on it, you have to make sure it is totally accurate,” Lord says.
Board members say that, like parents, they have been kept in the dark about testing irregularities. The state board wasn’t aware, Lord says, of the dispute between the superintendent’s office and Rhee until its members saw reports in TheWashington Post in the fall of 2009. She says she did not see the erasure analysis or the lists of schools flagged by McGraw-Hill until USA TODAY shared its copies.
After Rhee gave bonuses to educators in some schools that posted big gains in test scores in 2007 and 2008, there was little incentive to examine those scores, Lord says. “You’ve handed out these big bonuses. What are you going to do? Take them back?” she says. “It’s a bombshell. It’s embarrassing.”
Those who advocate for mayoral control often say they’re interested in speeding up the pace of change in struggling schools; they find that the democratic process slows that down. But the primary reason for having school boards and other oversight mechanisms is to ensure that everything is happening in a manner that is above-board, and in the students’ and community’s best interests. This is yet another example of why it’s so crucial to have that kind of oversight– and why “reform at 100 miles per hour” isn’t necessarily all it’s cracked up to be.
- “USA Today shows that many of the big test score jumps in DCPS probably are due to fraud” (GF Brandenburg)
- Follow the second comment on this post for links to more contemporary coverage of the would-be scandal, that failed to gain traction in the wider press.
- “USA Today investigates fraud under Michelle Rhee” (RheeFirst.com)
UPDATE: Tavis Smiley interviewed Rhee about the story today. She denies everything, saying that the third-party investigations were thorough and found no evidence of wrong-doing. (I wonder how the investigation would have turned out if people hadn’t been questioned in their bosses’ presence, or if they hadn’t received a share of $1.5 million in bonuses…) I’m looking forward to the rest of their conversation tomorrow night; I’m especially interested to see if/how she tries to explain away the suspicions of parents who claim their children’s academic performance at home didn’t match the scores they got on the tests.