A reader takes issue with the Néw York Times about “what is working.”

He or she writes:

“A few days ago, the Néw York Times published a bizarre and illogical editorial. It went like this: black and Hispanic students are very ill-prepared for college. The numbers are appalling. However, the Bloomberg administration accountability program was a great success,–except for the simplistic and failed A-F system. So, if so few students were succeeding, why is the accountability program indispensable? I posed this question to an expert on the DOE staff.

“Here is his answer:

“In its editorial, “Getting an Accurate Fix on Schools,” the New York Times argued for keeping, with modifications, the school grading system started under Mayor Bloomberg. Two forms of argument were presented to support this proposal; illogical arguments and arguments based on made-up facts and data.

“Let’s start with the illogical arguments. The editorialists begin by noting the “striking racial disparities” in the college readiness of New York City’s graduates.

“They could have added that this is true about non-graduates as well, about 20% of White and Asian students don’t graduate while over 40% of Black and Hispanic students don’t graduate. The illogic here is obvious. If Bloomberg’s policies have failed to address this very issue why would any rational person want to keep those same policies? For further details see the comprehensive review essay at this link.

“The editorial then notes we “must now find a way to solve it [the achievement gap] by ramping up the quality of education for poor and minority children.” So far so good.

“But the very next sentence begins “for starters, the city must preserve, at least in part, the controversial school evaluation system.”

“Huh!? In order to increase the quality of education… we need to keep the school evaluation system? It is hard to come up with an analogy that is equally absurd.

“Perhaps think of a football coach readying his team for the Super Bowl. Should he focus on developing the skills and tactics of his players? Or should he spend his time tinkering with the quarterback rating formula? Which strategy would you bet on?

“Logically you would think that in order to increase the quality of education we should focus on increasing the quality of education. Well proven initiatives should be promoted such as universal pre-K programs focused on building the academic and social/emotional skills of pre-schoolers, after-school enrichment programs, extended summer school for all students, intervention programs for students struggling academically, the development and implementation of a comprehensive and rich curriculum for all grade levels, and smaller class sizes in the early grades.

“Next the editorialists write that “de Blasio has rightly decided to junk the simplistic…A-through-F grading system. Their proposed solution is “continue to report a separate rating for each relevant metric.” Why would a whole bunch of simplistic letter grades per school be any better than a single simplistic letter grade per school?

“But enough with the illogical arguments. Let’s take a look at the made-up facts and data. The editorialists claimed that “the Bloomberg administration devised a way to control for demographically driven differences.”

“Not true. In fact, according to the Independent Budget Office “the method of calculating the continuous metrics on which final progress report scores are based may not fully control for confounding variables. All other things being equal, a school with a higher percentage of black and Hispanic students or special education students is likely to have lower performance and progress scores than other schools.”

“A report by the Metropolitan Center for Urban Education found that “schools with higher percentages of Black and Latino students received lower Progress Report grades” and “school demographics play an important role in predicting grades.”

“Yet another report, by New Visions for Public Schools, found that “schools’ overall PR scores remain associated with many preexisting risk factors, suggesting that a school’s score can be influenced by factors outside of its control.”

“The editorial asserted that “the data show that over the last two years, nearly 80 percent of the lowest-performing schools improved their ratings after receiving help in the areas where they were weak.”

“Not true. A check of the actual data posted on the New York City Department of Education’s webpage as a “multi-year summary” of “Progress Report citywide results” reveals that this is inaccurate. Of the 23 schools that received grades of F for the 2011-12 school year one got an A the next year, 5 got B’s, 10 got C’s, 3 got D’s, and 4 got F’s. Another 10 schools with F’s were closed. For the 2012-13 school year 45 schools got F’s. At almost double the number of F’s than the previous year there is no evidence that 80% of schools improved. Of these 45 schools 4 were brand-new schools for which the F-grade was their first grade ever.

“It is worth noting that this is a higher failure rate than that of pre-Bloomberg era schools. Another 4 schools got F’s the previous year, 12 had D’s, 17 had C’s and 8 had B’s. The data show that over the last two years school grades appear to randomly swing in all directions. From one year to the next over half of the schools with failing grades one year can be expected to get average or good grades the next. And over half the schools with failing grades had received an average or good grade the year prior—calling into question the school grading method.

“The New York Times declared that without these grades we “will never know how well students are doing.” In fact the data show that the grades give a contradictory picture. The other high-stakes but qualitative school metric, the School Quality Review, shows zero predictive correlation (as opposed to backwards looking correlation, which Mayor Bloomberg’s Department of Education artificially produced by forcing reviewers to align their score to the prior year’s report card grade) with Progress Report grades. Of the schools that swung from a B-grade in 2011-12 to an F-grade in 2012-13 the median Quality Review score was between “well-developed” and ”developing.” Of the schools that swung from F/D grades in 2011-12 to an A-grade in 2012-13 the median Quality Review score was “developing.” The schools that ended up scoring higher on the quantitative measure scored lower on the qualitative measure. Schools that ended up scoring lower on the quantitative measure scored higher on the qualitative measure. Instead of clarifying things the grades leave total confusion in their wake.

“When the Grey Lady reads like Pravda, with complete disregard for facts and a seemingly bottomless willingness to make up data, education in the United States is in serious trouble. Instead of trying to defend poorly designed metrics using false data we need to figure out how to provide schools with better oversight and support. The non-geographic network support structure should be abandoned.

“Replacing networks with local community superintendents, who have proven capacity as instructional leaders, would be a good first step. These superintendents would oversee 15 or so schools and develop a deep understanding of each school’s successes, needs, and challenges.

“They would develop strong relationships with the community, build close ties with families, and form connections to groups that could provide families with out-of-school assistance. They would work with a re-organized Tweed to make sure that families, students and schools get the support and programs they need. Only then will we make progress in closing the achievement gap.