Log in  |  Search

Joel Klein Turns a Blind Eye to His Own Data on Charters and Test Scores

The Atlantic just published a long opinion piece by Joel Klein, including a repetition of his long-standing argument that New York City’s charters perform miracles with “students who are demographically almost identical to those attending nearby community and charter schools,” and that anyone who claims differently is a blind supporter of the “status quo.” A closer look at Klein’s own numbers, however, tells a very different story. According to the progress reports released by his Department of Education just last year, New York City’s charter sector did not outperform similar district public schools. And the Harlem Success Academy — the school which he specifically holds up as “almost identical” to neighboring district schools — actually serves dramatically lower proportions of the city’s neediest students and of English Language Learners than other Harlem schools.

As most observers of the city’s schools know, each year the Department of Education releases progress reports with “grades” for each of its district and charter schools, which take into account the progress that students at each school made when compared to students at “peer schools” (those with similar student bodies in terms of poverty, Special Education status, and the proportion of English Language Learners, as well as other factors.) On the newest school Progress Reports, which were released by Klein’s office in 2010, 58% of district schools got an A or a B in 2010, compared to only 34% of charters. In Districts 4 and 5 in Harlem, more than half of district schools got either an A or B (27 out of 53), compared to only 8 out of the 21 charters in those neighborhoods.

One of the factors driving these relatively low grades for the majority of the charters in the city is the fact that they did not serve the same high-needs students as their local schools. On Klein’s progress reports, very few charter schools were identified as enrolling the same kinds of students as schools in their neighborhoods in Harlem, the Bronx, and Brooklyn. Instead, the DOE concluded that most charters served student bodies similar to those of district “peer schools” with more advantaged populations. In addition, only one charter school in the entire city received any credit on its progress report for successfully educating students in self-contained classrooms (the highest need special education students). This could mean that the other schools enrolled few or none of these students, or it could mean that they were not as effective as educating these students as district schools.

These conclusions are backed up when we look at the newly released state “report cards” for the city’s schools. Based on the data charters reported to the state last year, the city-wide difference in poverty between charters and district schools almost doubled — from 2.5 percentage points in 2008-09 to 4.3 percentage points in 2009-10. In addition, poverty at public schools rose 2 percentage points from 2008-09 to 2009-10, while at charters the increase was only a tenth of one percent. Across the city, 15 percent of district students were English Language Learners, while in charters, English Language Learners made up only 5 percent of students.

The state and city data also shows that, in contrast to Klein’s claims, Harlem Success Academy does not serve “almost identical” students by any reasonable standard. According to the data Harlem Success released to the state, only 3% of its students are English Language Learners, compared to 12% in the average Harlem district school. No students at Harlem Success require the highest levels of Special Education services, compared to 31.4% of students in local district schools, and almost 80% of Harlem district school students qualify for free lunch, compared with just over 60% at Harlem Success. In addition, Klein’s claims for high performance at Eva Moskowitz’s Success Charter Network are based on a single school. Harlem Success Academy is the only school in the chain which includes tested grades, for a total of 120 students’ results. This single case is what Klein refers to when he argues that “Harlem Success has 88 percent of its students proficient in reading and 95 percent in math.”

As we’ve noted before, some of the city’s highest performing charter schools also have very high student attrition, and don’t replace students in upper grades. For example, based on testing data, KIPP Infinity lost 20% of the students from its 2006-07 5th grade class by 8th grade in 2009-10. Whether this drop is due to students leaving the school or to holding weaker students back in earlier grades, these schools’ rising scores may be due in part to their choice not to fill these open seats with students who wish to transfer in, an option that district schools do not have. As noted in this recent New York Times article, district schools are legally required to admit and serve all students, including those who transfer in the middle of the year or in upper grades.

Overall, Klein should have been more balanced in his portrayal of the city’s charter sector as a model for district schools to follow. While he may feel that “collaboration is the elixir of the status-quo crowd,” his arguments should have reflected some awareness of the data that had been released by his own office, some humility about the limitations of charters’ ability to address deeply rooted issues of inequality — and some acknowledgement of the success of district schools (and their unionized teachers) in overcoming challenges to academic achievement. Taking a look at his own facts about charters and test scores might have complicated his simplified story of the heroes and villains of urban school reform, but it would have made his piece more effective in persuading those who are working to improve education in our city that he has all students’ best interests at heart.


1 Comment:

  • 1 Bill Bean
    · May 16, 2011 at 8:13 am

    I tend to agree with most of Klein’s conclusions in the Atlantic article, but I also know to be skeptical when it comes to statistical arguments. We all know how to use them to our advantage.

    While I am not well read on the subject, you seem to have made some reasonable challenges to the way he handled or interpreted the stats. I certainly appreciate that, and hope that all involved in the discussion/debate will do better with their arguments. So much is at stake.