This commentary was written by an employee of the Néw York City Department of Education who specializes in data analysis. He/she requires anonymity.

An opinion piece in the New York Daily News by Robert Pondiscio wondered, “Is Eva Moskowitz the Michael Jordan of education reform, or is she Mark McGwire?” To give some context- New York State recently released the results of the 2013-14 grades 3-8 exams in English and in Math. Success Academy schools, of which Eva Moskowitz is the $475,244 a year CEO, did relatively well on these exams.

How well? Well, that depends on how you break down the numbers. Looking at the mean scale scores of general education students at K-8 schools in New York City- Success Academy schools average out at #79 in English and #12 in Math. This way of measuring Success Academy is most transparent- it includes the data on all their schools and, given their very low number of the highest-need special education students, most compares like to like. It also avoids the number tricks Success Academy has played in the past, such as focusing on the test results of one of the exams for one of their grades at one of their schools.

OK, so Success Academy may not be doing quite as well as they claim, but there is no doubt that their results are pretty good. To return to the question asked by the Daily News and echoed by many others, “how does Success Academy get their results?”

Well, the same data set that provided the test results suggests an answer. The only Success Academy school that has fully grown to grades 3-8 tested 116 3rd graders and only 32 8th graders. Three other Success Academy schools have grown to 6th grade. One tested 121 3rd graders and only 55 6th graders, another 106 3rd graders and only 68 6th graders, and the last 83 3rd graders and only 54 6th graders. Of course, this data set represents a snapshot from a single year.

Longitudinal analyses have found extremely high rates of attrition within student cohorts and students with disabilities and English Language Learners are over-represented among the students who disappear from Success Academy rosters.

Eva is no Michael Jordan. Her numbers are gimmicks, obtained by removing low scoring students from her schools. The high-scoring students remain and the low-scoring students are gone, along with their potentially disruptive effects on classrooms and the school as a whole. Eva is more like Lance Armstrong. They both win through artificial means. Lance through blood transfusions and EPO. Eva through attrition of students and obsessive test prep. According to the Daily News, while “suppressing the truth” Lance engaged in “an endless behind-the-scenes campaign to bully and intimidate people into silence. Some of it bordered on gangsterism.” Eva employs similar tactics to bully employees at the New York City Department of Education and to take space from special needs students to expand her schools.

There is another aspect of this whole sorry story that makes the Lance/Eva analogy so apt. In Lance’s case the world cycling governing body was complicit, or at the very least turned a blind eye, to the cheating. In Eva’s case education reformers, op ed writers, and think tanks refuse to acknowledge that throwing out students with low test scores and ending up with high test scores as a result is not a model for bettering education. They blindly insist that charter school chains that disappear low scoring students from their rosters show public schools how it can be done.

Every city and state seems to have one of these cheating charter chains- from Achievement First in Providence (over 50% attrition) and Achievement First in New Haven (over 50% attrition), to Uncommon Schools in Newark (38% attrition), to BASIS in Arizona (over 60% attrition), to McKeel Schools in Florida (“McKeel Charter School System has no control over which students are admitted to its three schools, its superintendent said, but it does control who gets to stay”), to the Noble Network in Chicago (up to 36% attrition), to Harmony Charters in Texas (40% attrition), to DSST in Denver (38% attrition), to KIPP in San Francisco (up to 45% attrition) and KIPP in Tennessee (18% attrition in a single year!).[1] What happens to all the children left behind, or should we say “kicked to the side of the road?” by these “high performing” charter chains?

No sport can be built on a foundation of rampant cheating. No education system can be built on school models that are based on number games.

[1] Refer to the linked studies for the particulars on each charter chain, as the grade-levels and time-span differ in each analysis.