The New York Times Magazine has a long article about Eva Moskowitz and her chain of charter schools in New York City. The charter chain was originally called Harlem Success Academy, but Moskowitz dropped the word “Harlem” when she decided to open new schools in gentrifying neighborhoods and wanted to attract white and middle-class families.

I spent a lot of time on the phone with the author, Daniel Bergner. When he asked why I was critical of Moskowitz, I said that what she does to get high test scores is not a model for public education or even for other charters. The high scores of her students is due to intensive test prep and attrition. She gets her initial group of students by holding a lottery, which in itself is a selection process because the least functional families don’t apply. She enrolls small proportions of students with disabilities and English language learners as compared to the neighborhood public school. And as time goes by, many students leave.

The only Success Academy school that has fully grown to grades 3-8 tested 116 3rd graders but only 32 8th graders. Three other Success Academy schools have grown to 6th grade. One tested 121 3rd graders but only 55 6th graders, another 106 3rd graders but only 68 6th graders, and the last 83 3rd graders but only 54 6th graders. Why the shrinking student body? When students left the school, they were not replaced by other incoming students. When the eighth grade students who scored well on the state test took the admissions test for the specialized high schools like Stuyvesant and Bronx Science, not one of them passed the test.

I also told Bergner that Success Academy charters have among the highest rates of teacher turnover every year, which would not happen if teachers enjoyed the work. Helen Zelon wrote in “City Limits”: “In Harlem Success Academies 1-4, the only schools for which the state posted turnover data, more than half of all teachers left the schools ahead of the 2013-14 school year. In one school, three out of four teachers departed.” I also told Bergner about a website called Glass Door, where many former teachers at SA charters expressed their candid views about an “oppressive” work climate at the school. As more of these negative reviews were posted, a new crop of favorable reviews were added, echoing the chain’s happy talk but not shedding light on why teachers don’t last long there.

Bergner argued every issue with me. He reiterated Success Academy’s talking points. He said that public schools lose as many students every year as SA charters; I replied that public schools don’t close their enrollment to new students. Again, defending SA, he said that closing new enrollments made sense because Moskowitz was “trying to build a culture,” and the culture would be disrupted by accepting new students after a certain grade. I responded that public schools might want to “build a culture” too, but they are not allowed to refuse new students who want to enroll in fourth grade or fifth grade or sixth grade or even in the middle of the year.

He did not think it mattered that none of her successful eighth grade students was able to pass the test for the specialized high schools, and he didn’t mention it in the article. Nor was he interested in teacher turnover or anything else that might reflect negatively on SA charters.

Subsequently I heard from his editor, who called to check the accuracy of the quotes by me. I had to change some of the language he attributed to me; for example, he quoted me defending “large government-run institutions,” when what I said was “public schools.” He was using SA’s framing of my views. I asked whether Bergner had included my main point about attrition, and the editor said no. I explained it to her and sent her supporting documentation.

This is the paragraph that appeared in Bergner’s article, which understates the significance of selective attrition while not mentioning SA’s policy of not accepting new students after a certain grade:

“On the topic of scores, the U.F.T. and Ravitch insist that Moskowitz’s numbers don’t hold up under scrutiny. Success Academy (like all charters), they say, possesses a demographic advantage over regular public schools, by serving somewhat fewer students with special needs, by teaching fewer students from the city’s most severely dysfunctional families and by using suspensions to push out underperforming students (an accusation that Success Academy vehemently denies). These are a few of the myriad factors that Mulgrew and Ravitch stress. But even taking these differences into account probably doesn’t come close to explaining away Success Academy’s results.”

This minimizes the stark differences in demographics when comparing her schools to neighborhood public schools. The Success Academy charters in Harlem have half as many English language learners as the Harlem public schools. The Harlem Success Academy 4 school, which has 500 students, has zero students with the highest special needs as compared to an average of 14.1% in Harlem public schools. This disparity is not accurately described as “somewhat fewer.” It is a very large disparity. Attrition rates are high, which would not be happening if the school was meeting the needs of students. As I wrote earlier this year:

“Moskowitz said [on the Morning Joe show on MSNBC], referring to the students in her schools, “we’ve had these children since kindergarten.” But she forgot to mention all the students who have left the school since kindergarten. Or the fact that Harlem Success Academy 4 suspends students at a rate 300 percent higher than the average in the district. Last year’s seventh grade class at Harlem Success Academy 1 had a 52.1 percent attrition rate since 2006-07. That’s more than half of the kindergarten students gone before they even graduate from middle school. Last year’s sixth grade class had a 45.2 percent attrition rate since 2006-07. That’s almost half of the kindergarten class gone and two more years left in middle school. In just four years Harlem Success Academy 4 has lost over 21 percent of its students. The pattern of students leaving is not random. Students with low test scores, English Language Learners, and special education students are most likely to disappear from the school’s roster. Large numbers of students disappear beginning in 3rd grade, but not in the earlier grades. No natural pattern of student mobility can explain the sudden disappearance of students at the grade when state testing just happens to begin.”

I have no personal grudge against Eva Moskowitz. On the few occasions when we have appeared together, we have had very cordial conversation. What I deeply oppose–and this is what I stressed to Bergner and he deliberately ignored–is that Success Academy is not a model for public education. No one expects that Bronx Science is a model because it does not have open doors; it admits only those who meets its standards, and they are high. Eva Moskowitz pretends that her schools get superior results with exactly the same population because of her superior methods, when in reality the success of her schools is built on a deliberate policy of winnowing out low-performing and nonconformist students.

Why did Bergner insist on obscuring this crucial difference between SA charter schools and public schools? Public schools can’t remove students with low scores. They can’t refuse to enroll students with severe disabilities and students who can’t read English. They can’t close their enrollment after a certain grade. Unless they have a stated policy of selective admissions, they must accept everyone who seeks to enroll, even if they arrive in February or March. Their doors must be open to all, without a lottery. It is not honest to pretend that public schools can imitate Moskowitz’s practice of selective attrition. And it is not honest to overlook that difference.