Significance of Tweed Insider's Analysis of Bloomberg's Department of Education Legacy

*Demonstrates tale of two educational school systems, akin to de Blasio's case of two cities of rich and poor  *Vacuous graduation claims of Department of Education dispelled
New York City education observers have been pining for the day when a Chelsea Manning or Edward Snowden type dumps tons of documents from the bowels of Tweed. Instead, similar to the Pentagon Papers we have a profound essay in Diane Ravitch's blog from a Tweed insider.
True, the Pentagon Papers were valuable in revealing much information that was damning to the cause of the hawks of the Vietnam War. Yet, many of the details were suspected by war critics and were visible to participants on the scene. The Pentagon Papers, aside from revealing important information, were important for providing the basis for a detailed, authoritative analysis of the outrageous pursuit of the Vietnam War.
In the present analysis we have a very authoritative analysis of the nature of the corporate education program as pursued in New York City under mayor Bloomberg. This account by someone in the central Department of Education office confirms many of the allegations made by other sorts of insiders (teachers) such as the Chaz School Daze blog (e.g., more recently here, here, here  here), the present blog, and others, such as NYC Educator and Assailed Teacher, who saw general patterns, but did not have an intensely documented scope as Tweed Insider exercised. No commercial media outlet has as of yet dared to address the total fallacy of the emperor's clothes chimera that is Bloomberg's education record.
The Tweed Insider's analysis confirms that mayor Michael Bloomberg's new schools were skewed towards greater chance for success with channeling stronger students to them. This confirms the recurrent recent analyses at Chaz School Daze. Furthermore Tweed Insider demonstrated that resources were skewed to the new schools in an unfair manner.

Bloomberg has created a dual school system, one flush with resources and varied curriculum, the other, resource-starved with the narrowest of curriculum. As is apparent in the following essay, the favored schools have had fewer students with disabilities. In this respect, this is a class-divided system, an indeed more sophisticated counterpart to the Jim Crow American apartheid of old, but refined or not, this is unfair and deeply insidious. The great challenge to incoming mayor Bill de Blasio is to reverse the disparity, and to provide to all students the richly diverse curriculum that his son is enjoying at Brooklyn Technical High School.

The following is the Tweed Insider analysis, reprinted from the great Diane Ravitch's blog, December 20, 2013:

Tweed Insider: Where the Bloomberg Administration Went Wrong on Education

The following was written by an insider at the New York City Department of Education, who requests anonymity, for obvious reasons.

“Bring This City To A Place Where Our Children Really Are Put First:” Implementing Mayor-elect de Blasio’s Education Platform in New York City

The above quote by Mayor-elect Bill de Blasio captures his commitment to addressing the growing opportunity and income divide in New York City (NYC). Mr. de Blasio’s education platform focuses on improving all schools via principles of equity, engagement, collaboration and support. He has been attacked for this. In order for us to move forward, the time has come for a comprehensive review of the outcomes of Bloomberg-era education policies.

Some of the Bloomberg educational initiatives and programs have shown potential for improving the educational experiences of students. The number of uncertified educators teaching in NYC has decreased and the qualifications of teachers, especially in high poverty schools, have increased. Programs such as the Interagency Taskforce on Chronic Absenteeism, the Expanded Success Initiative, and the Middle School Quality Initiative have shown promise based on early data. These programs should be continued and broadened throughout NYC. Alongside the significant strides in public health policies in NYC such as the ban on smoking in public spaces, the nutrition value of school food improved, in some cases thanks to advocacy by community groups. Other policies such as creating a centralized bureaucratic structure focused on accountability, evaluating schools and teachers based primarily on test scores, a recurring cycle of closing schools and opening replacement schools, and an emphasis on market-based reforms and charter schools have been less successful. The successful policies have been empirically responsive, thoughtful, and well-designed. They were crafted in response to issues of genuine concern to parents and educators and pulled together and coordinated expertise and resources from multiple agencies, partners and community groups. The less successful policies are characterized by an ideology that attempts to have incentives and market forces address issues without direct responsibility and ownership on the part of the NYC Department of Education (DOE). As we will see, they also do not acknowledge the data and empirical evidence.

The implications of the failure of these policies are of national importance. Similar corporate-style reforms are sweeping the nation. A recent survey of the last dozen years of education data in NYC showed the poor overall outcomes of Mayor Bloomberg’s policies. If these policies have failed after 12 years, under a single mayor with no checks on his power and the backing of his personal fortune, it is obvious that the time has come to implement a very different reform platform. The Big Apple now has the opportunity to blaze a different and better path.

This essay has five parts.
Part I. Have the current organizational structures and funding policies formed an environment necessary for a successful education system to thrive?
Part II. Are measures of school and teacher performance currently in use valid, reliable, and fair?
Part III. Is the policy of closing schools and replacing the closed schools with new schools working?
Part IV. Has the portfolio strategy based on market-choice and charter schools improved student outcomes?
Part V. What new policy directions should be considered for implementation?
Each fact and data point cited is sourced at the embedded links throughout the essay.

Part I. Have the current organizational structures and funding policies formed an environment necessary for a successful education system to thrive?
NYC’s non-geographic school support network structure and the central bureaucracy at Tweed are bloated, ineffective, inefficient, and lead to patronage. Of the non-geographic networks a full 1/3 received ineffective or developing quality ratings. An audit by the NYC Comptroller’s Office found that “it is difficult to determine whether or not that support increased the efficiency of the school’s day-to-day operations… a network’s contribution to the scores allotted to the schools cannot be directly ascertained.” This is not surprising. We do not assign police precincts to police blocks in different neighborhoods, even if those blocks have similarities. We do not have fire stations cover fires in different boroughs. So why should our schools be supported by teams responsible for 30 schools spread all across the city? A policy report by the NYC Comptroller revealed that this structure blocks parent influence in local school governance. The opportunities for conflicts of interest, patronage, and corruption, supposedly a concern under a geographic structure are, unfortunately, all too plentiful in the centralized structure. Favoritism is rampant. For example, schools were forced to hire Aspiring Principal Program (APP)/Leadership Academy principals despite the data showing that student suspensions increase, the” performance drop… is larger at the schools hiring an APP graduate” and over 40% of APP graduates are no longer principals in the same school after 3 years.
While community input was stifled the central office headcount increased by 70% and the salaries by 79%. The number of non-pedagogues employed by the DOE increased to the highest levels since 1980. According to the Independent Budget Office an ever increasing share of money budgeted to “total classroom instruction” actually went to central offices. In 2007 about $550,000,000 went to central offices and in 2012 about $793,000,000 went to central offices, approximately a 45% increase in total classroom instruction dollars going to central offices (figured derived from the bar chart in the IBO report). Tens of millions of dollars were spent on recruiting and at least $200,000,000 on subsidized tuition and training for the Teaching Fellows, a fast-track teacher preparation program. This despite the fact that the evidence showed Teaching Fellows are slightly less successful and have higher attrition rates than traditionally certified teachers. More money was spent on the “Innovation Zone,” though an audit by the NYC Comptroller found that the initial pilot had “no clear specific measurable criteria to use in assessing the effectiveness.” A United States Department of Education denial of an Innovation Zone grant application stated “NYCDOE does not provide a high level of transparency in… processes, practices, and investments.”

Financial and organizational mismanagement is rampant. An audit by the NYC Comptroller found that the DOE spent $67,000,000 on a special education information system that “is not meeting its overall goal.” Another audit of an $80,000,000 “achievement and reporting information system” concluded that the DOE “lacks effective measurements for gauging whether ARIS [Achievement Reporting and Innovation System] is an effective tool.” The DOE’s data system was so poorly designed that schools paid thousands of dollars out of their budgets to procure better systems. The cost for developing a High School Application Processing System increased from an originally contracted $3,600,000 to $23,000,000, after which use of the system was discontinued. An audit by the New York State Comptroller’s Office noted the “lack of documentation supporting the justification for non-competitive contracts submitted…vagueness of the categorization of “other special circumstance” which constituted the majority of the $342.5 million of non-competitive contracts…significantly diminishes assurance that DoE’s non-competitive contracts are justified.” Other audits “found significant control weaknesses, which prevent DOE from effectively monitoring its individual consultants for mandated services” and “DOE failed to provide Related Services to 72,302 of 285,736 students-more than 25%.” Nonetheless, Tweed was not held accountable for any of this, nor for the bungled centralized Regents scoring this past June. When a parent advocate requested performance evaluations of the DOE’s leadership team she received a reply stating that “no such records have been created. Accordingly, there are no records to provide.”

Schools are funded in a deliberately unequal and unfair manner. Under the current approach schools are not budgeted the money they are promised by the Fair Student Funding formula. The “percent of formula” that schools actually receive varies greatly between schools and is not based on clear and objective educational factors. Some schools get 80%. Other schools get over 100%. For example: Flushing High School receives 80.8% of their funding. Queens High School for Language Studies, a new school co-located in the very same building, gets 100% of their funding. Veritas Academy, a new school co-located in the very same building, gets 100% of their funding. Dewitt Clinton High School receives 82.38% of the funding that they are supposed to get. Bronx Collaborative High School, a new school co-located in the very same building, gets 100% of their funding. World View High School, another new school co-located in the very same building, gets 100% of their FSF their funding. Lehman High School receives 81.44% of their funding. Westchester Square Academy, a new school co-located in the very same building, gets 100% of their funding.
Schools with a more impoverished student body receive, on average, 6+% less of the funds they should be receiving than schools with a relatively affluent student body. Schools with the most academically struggling student body receive, on average, 13% less of the funds they should be receiving than schools with an academically privileged student body. A Daily News story revealed that new schools were overfunded at the expense of other schools. One small group of schools was underfunded by $30,000,000. A senior thesis from Princeton analyzed FSF at elementary and middle schools and concluded that “it does not appear that that relationship [between student need and school funding] generally increased with the implementation of Fair Student Funding.”
There are inequities in the construct of the formula itself. It does not distinguish between Beginner, Intermediate, Advanced, and Long-Term English Language Learners. The formula also does not provide any weights for Students with Interrupted Formal Education. This means, for example, that schools that have many beginner and long-term ELL students are funded at the same level as schools with advanced ELL students, even though the instructional needs are dramatically different. The formula deliberately overfunds Integrated Co-Teaching/Collaborative Team Teaching ($7173.81 of additional funding per student) at the expense of self-contained ($2407.91 of additional funding per high school student) settings. There is no additional funding for students in temporary housing, students who are involved with the juvenile justice system, or other students requiring additional supports to succeed. A report by the Independent Budget Office found that Fair Student Funding never fully kept its already limited promises. Black and Hispanic students are less likely to have a gym, medical office, and library and have fewer art and music rooms and science labs in their schools than Asian and White students.

Part II. Are measures of school and teacher performance currently in use valid, reliable, and fair?
NYC school report card letter grades do not reflect actual school quality. School grades on the Progress Reports fluctuate wildly from year to year and do not reflect genuine changes in school quality. One year’s data showed that 75% of the schools that received F’s the previous year got A’s or B’s the next. Another year, 60% of schools moved 200 places or more in the rankings as compared to the prior year. Yet another year, over 55% of schools moved a letter grade or more (out of only 5 possible grades) as compared to the prior year. One analysis demonstrated that this year-to-year change in grades is only slightly less than would be expected by random luck. One year’s grades showed a correlation of -.02, in other words no correlation at all from one year to the next, in schools’ progress scores. Another set of grades showed low moderate stability in school growth scores, with small schools showing very little stability. Schools with fewer than 500 students saw the largest swings. A fascinating experiment demonstrated that a random number generator was 4-10% more accurate in predicting a school’s grade than the actual grade the school received the prior year. Summing up, it is obvious that the wild swings in school grades are not tracking real changes in the quality of education students are receiving. Despite claims to the contrary there is no evidence that school report cards have improved school quality.

The metrics used in the school report cards are not valid. Many of the measures used in the report cards are easily gamed by schools and have nothing to do with the quality of education students receive. Up until last year schools graded their own Regents exams, leading to a situation where the number of passing scores was inflated by 4-8.3%, depending on the exam, and “the manipulation of Regents scores was noticeably more common in NYC than elsewhere in the state.” Allegations of test-tampering and grade-changing have “more than tripled” under mayoral control. Over 25% (in years past it was 33%) of a high school’s grade is based on how many students earn 10+ credits. Given the curve in these report cards, this can mean the difference between an “F” and a “B” or between a “D” and an “A”. There are, of course, no uniform criteria for what a student has to learn in order to earn a credit. As a result this measure has been entirely corrupted. In the first four years of school report cards the number of students earning 10+ credits a year jumped a remarkable 16 percentage points citywide. The increase at schools serving disadvantaged students was even greater. Some schools practically jumped over the moon and increased student credit accumulation by over 50% from one year to the next. New small schools in particular have corrupted this measure. An analysis of the data in this spreadsheet shows that new small schools as a group rank 10% higher on the credit accumulation measure than would be expected based on their Regents pass-rate ranking. This fact, along with the recently uncovered data showing disproportionate numbers of struggling students not accepted to a high school through the application process are sent to large/medium sized schools, raise serious doubts about the claims that new small schools are performing better than other NYC high schools.
Another 10% of the grade is based on responses to a school Learning Environment Survey. A report from New York University found that the “category scores were not strong measures for distinguishing between schools…the survey provides…less information about how that school differs from other schools…scores were not powerful predictors of other performance indicators.” It turns out that teachers inflate their responses by significant margins as compared to student responses in schools that received a “D’ or “F” the prior year. In addition to the potential for gaming, many of these measures are flat out invalid. The report cards use test scores to compare student performance across years in a manner the tests were not designed to do. The reports cards do not account for the statistical noise in test results, meaning that schools whose test scores are statistically indistinguishable nonetheless receive very different grades. The report cards use standard deviations in ways that make tiny differences in school performance seem bigger than they really are. For example a student attendance rate .8% lower than that at other schools ends up being unfairly graded as 18.4 out 100. Summing up, it is clear that the metrics used in these report cards do not really measure the quality of education students are receiving.

The school report cards penalize schools working with disadvantaged students. It is widely acknowledged that it would be unfair to compare a school whose students enter scoring above grade level on exams to a school whose students enter scoring well-below grade level. It would also be unfair to compare schools that work with disadvantaged and struggling students to schools that work with only selected students. The report cards claim to account for this by only comparing schools to other schools with similar student populations. But is this, in fact, the case? The data say no. A report by New York University found that schools serving higher proportions of Black and Latino students, English Language Learners, and students with disabilities received lower grades. The same pattern was found by a professor at Columbia’s Teachers College and by NYC’s Independent Budget Office. They also found that the less selective a school was in accepting students the lower the average grade. Other analyses found that the report cards favor schools that start with higher student baseline scores, that as the percent of self-contained special education students at a school increases from 0% to 14.5% the average report card grade falls by over 20 points, and that schools that get “D” and “F” grades have many more students entering overage than schools that get an ”A.” But that is not all. Schools in the lowest quartile of grades have lower entering student scores and 2.5x more self-contained students than schools in the top quartile. An analysis of another year’s data found that the median proportion of self-contained special education students at schools with “F” grades was 2,100% greater than at schools with “A” grades. Schools with the lowest levels of students receiving free lunch were 3.5 times more likely to get A’s than schools with the highest levels. So it is obvious that the report cards do not account for differences in incoming student characteristics.

The peer index does not account for demographic differences between schools. The creators of these report cards invented “peer indexes,” designed to compare similar schools to each other. But the data show that schools with higher peer indexes receive higher average grades, that the peer indexes lump together very dissimilar schools, and that the peer indexes do not really control for incoming student characteristics. Summing up it is clear that the report cards are blatantly unfair and do not truly measure the quality of the education at schools.

Value-add use of test scores to rank and evaluate teachers is unsupported by the data. For years NYC created value-add reports for math and English 4^th-8^th grade teachers. As of 2013 New York State will be creating value-add reports for even more teachers as part of the Race to the Top teacher evaluation system. What does the data from NYC tell us about the reliability of such reports? The data used to create value-add teacher rankings in NYC was often inaccurate. The DOE itself admitted that a third of all value-add reports were not reliable and that in 30 schools the reports for every single teacher were not reliable. Scores for teachers of classes at the top or bottom were particularly unreliable, with 3,900 of 11,800 multiyear ratings falling into this category. Even with multiple years of data, up to 70% of teachers could not be distinguished from average. A .01 change in either direction changed a teacher’s percentile ranking up to 63%, making small real world differences appear larger than they really were. Scores changed by large amounts from year to year, with 49% of teachers moving downwards. There was almost no correlation between the English and math scores of teachers who taught both subjects during the same year. 98% of teachers fell in a very narrow range, meaning that the numbers should not be used to create rankings of teachers. Looking at the same exact teachers, in the same exact schools, teaching the same subjects there was no correlation between teachers’ 2005-06 scores and their 2007-08 scores. There was no correlation between a teacher’s value-add score from year to year. It was, in fact, close to random. A teacher in the 90+ percentile one year had only a 1 in 4 chance of remaining there the next. A teacher in the bottom 10% had only a 7% chance of remaining there the following year. Only 7% of teachers landed above the median for 3 years in a row, with lots of movement between the upper half and the bottom third. Predictions about future student achievement assumed by the formula were not accurate. Scores were biased against teachers of high performing students. There was a 3+:1 ratio of teachers who taught high-performing students rated below average versus above average. A single extra question correct on the exams of a teacher in this group raised their value-add score by 10-20 points while an incorrect answer lowered their score by 20-50 points. The reports did not control for school level factors and class size. For example, a teacher was 7.3% less likely to receive a good rating for each additional student increase in average class size. A teacher’s score one year predicted only 5-8% of the next year’s score. The value-added scores of teachers who taught similar groups of students with similar pre-test scores for two years in a row showed almost no correlation. 43% of teachers with very high value-add scores in 2009 did not meet that mark in 2010. Of the thousands of teachers in the top 20% in 2005-06 only 14 math teachers and 5 ELA teachers remained there each year through 2009-10. The educator rated as the worst teacher in the city taught the highest need English Language Learners, very few of whom took the exams the rating was based on. What’s worse, 40% of her students had “imputed” scores which are wholly unreliable.
This year New York State is going to create such value-add rankings for teachers and principals in NYC. A review of the NYS model found significant biases in the model such as higher incoming student scores in both English and math correlating to higher growth scores. Yet another report found that teachers of high achieving students are more likely to get higher ratings and teachers of students growing up in poverty to get lower ratings. All this messiness means that the value- add scores have no practical value.

Part III. Is the policy of closing schools and replacing the closed schools with new schools working?
The student population at closed schools had significantly greater needs than other schools in the city. Schools were punished for working with disadvantaged students. According to a report by researchers from Brown University the data on all schools closed since 2003 shows that they had more special education students, more English Language Learners and a higher poverty rate than the citywide average. They also found that schools that were closed had 4x as many (15% more) students entering overage. The number of high needs students increased dramatically in the years before the schools were closed. Schools with the lowest peer indexes were closed. Within schools in the top 1/3 of student need 40% of the D’s and F’s closed, none of the D’s F’s in the middle 1/3 of student need closed and of the schools in the lowest 1/3 of student need none got D’s and F’s. Within the top 1/3 of student need the schools that are closed had higher levels of poverty, special education students, high-needs special education students, overage students, and boys (note that poverty level and % boys are not factored into school report card grade). Among Persistently Lowest Achieving schools selected for school reform models the schools selected for closure had lower average incoming 8^th grade scores, more students entering overage and a lower peer index (meaning higher student needs) than schools selected for the less punitive transformation or restart models. A report by the Independent Budget Office “found that on nearly every measure the closing high schools had greater concentrations of high needs students.” A second Independent Budget Office report found that “the share of their enrollment in some high needs categories,
such as the share of students in special education, has been increasing in recent years.”
As the percent of self-contained special education students in a school increased from 0%-14.5% the graduation rate declined from 68.6% to 54% (i.e. by an almost identical 14.6%). School report card scores, used to make closure decisions, also fall as the rate of self-contained students increase at a nearly 1 to 1 ratio. Of the 69 schools that received A’s in the peer group of schools to be closed 40 had 5% or fewer high need special education students. Of the 18 schools that received D’s all except 2 had 25+% high need special education students (and 1/3 had over 55%). Closing schools had 25% or more students entering overage than schools that received A’s in the very same districts. Looking at 11 districts with closing schools the schools that received A’s had significantly fewer students entering overage than the schools that got D’s or F’s.
A review of two closed schools found that on the DOE’s own Regent performance metric they were doing much better than would be predicted based on incoming student need. One high school in Manhattan scored 10 percentile points higher than expected across all Regents exams and another high school in Brooklyn scored 19 percentile points higher than expected. The 17 high schools selected for “turnaround” by the DOE in 2012 had Regents pass rates 8.5 percentile points higher than would be expected based on incoming student performance for the two years prior. Nonetheless the DOE wanted to “excess” all the teachers at those schools, a process an independent arbitrator halted.
The Deputy Chancellor responsible for closing schools started a school that had virtually no self-contained students (.2%), only 11% overage students, and average incoming scores of 2.8, a much more advantaged student population than that in the schools he closed.

DOE did not follow its own set of criteria for school closures and ignored the needs of students in the schools they closed. One year 14 out of 20 schools that Bloomberg’s DOE wanted to close, in order to make room for other schools, scored above the criteria for closing a school as set by the very same department. Yet another Brooklyn school was closed even though it received a “proficient” on its Quality Review which should have protected it from closure according to the DOE’s criteria. Another school built a website showcasing extensive data demonstrating that the school in fact did a good job with its students. The department ignored the data and closed the school anyway. A report by a student group on “the abandonment of students in closing schools” found that of 33,000 students attending 21 closing high schools in their final years “5,612 dropped out, 8,089 were still enrolled, 9,668 were discharged, only 9,592 actually graduated.” They attributed this in part to the resources that were pulled from these schools.

Many new schools are able to selectively screen their students despite the official “limited unscreened” admissions process. New schools have access to prior student performance and attendance data and can use that information to rank students for admission. Some limited unscreened schools actually required students to write essay. A report by researchers at Columbia University’s Teachers College found that new schools accepted 9-10% more students proficient in reading and math, with 4% average higher prior attendance who were 15% less likely to enter overage, 6% less likely to be ELLS, 5% less likely to be students with disabilities, and 7% fewer males. Overall they found that the closing and opening of schools did nothing to reduce the segregation of students by academic need. A New York University researcher found that 2 of the 3 new small high schools she studied managed to screen their students, even though they were officially not allowed to. Other new schools seem to lose vast numbers of students from each cohort, in one instance from 26-37% of their incoming students over 3 cohorts.

New schools that replaced the closed schools have a much more advantaged student population than the schools they replace. New schools (for example those on the JFK and Columbus campuses) serve significantly fewer self-contained special education students than the schools they replaced. At the same time new schools with the higher proportions of self-contained special education students performed poorly on the DOE’s metrics. A review of new small schools on 8 campuses found that the proportion of self-contained students was on average 9% fewer than the schools they replaced. When new schools (for example Gateway and School for Community Research and Learning on the Stevenson Campus and UAA for History and Citizenship) do serve a similar proportion of self-contained students they are closed as well. In yet another example, at the closed Far Rockaway HS over 20% of the students had special needs while the new schools that replaced it had an average of 11% (of which, on average, 2% were self-contained students compared to 55% at Far Rockaway). At the closed Beach Channel HS school 19% of the students had special needs compared to 9% in the co-located school (of which 28% were highest need at Beach Channel vs. 0% in the new school). A review of 10 campuses showed that the schools that replaced the closed schools had students with incoming math/reading scores over 29 percentile points higher than those of the closed schools. In Queens, the large schools the DOE has targeted for closure admit overage students at about four times the rate of new schools in the same neighborhood. In Brooklyn, the rate is three to one, and in the Bronx it is double. The disparities are even greater when comparing closed schools to new schools by borough on the proportion of the highest need special education students. The new schools that replaced Morris HS in the Bronx have fewer overage, Limited English Proficient, free lunch and special needs students and more students who have passed reading and math exams in middle school with higher prior attendance as well. The same is true of new schools that replaced the closed Bushwick HS in Brooklyn and Evander Childs HS in the Bronx.

DOE deliberately concentrated high needs students in specific schools and did not send high needs students to new schools. A report commissioned by the DOE found that concentrations of high needs students in schools was the most important factor by far in predicting low graduation rates. Nonetheless DOE continued to concentrate high needs students in specific schools, while other schools accepted very few high needs students. Advocacy groups found that ELL students were deliberately excluded from new schools and that special education students were deliberately excluded as well. A follow up report found that while the DOE formally reversed the policy of exclusion they did not actually ensure that ELL students were given access to new schools.
Researchers from Brown University found that the DOE deliberately sent the highest-need over the counter students to a specific group of large high schools and that most small schools were not sent such students. “Higher percentages of OTC students are assigned to struggling or persistently low-achieving high schools. Significantly higher percentages of OTC students were also assigned to high schools targeted for closure in the years before their closures were announced… our study identified a substantial group of high-performing high schools that are assigned very low percentages of OTC students and a similar-sized group of struggling high schools assigned very high percentages of OTC students.”

As a whole new schools perform no better than old schools. Despite some, thoroughly debunked, claims to the contrary new schools do not have very good outcomes. The Center for New York City Affairs released a report that noted “of 158 new schools for which data is available, 90 saw their average daily attendance decline by at least 2 percent, and 37 saw their attendance decline sharply, by 5 percent. Only 15 had attendance rates that were increasing… a large proportion of the new schools achieved high graduation rates for their first class but sharply lower rates for their second class. Of 30 Bloomberg-era small schools that had graduated at least two classes in 2007, 13 had graduation rates that declined in the second four-year cohort.” However, the report did not take into account the more advantaged population of new schools when comparing their outcomes to old schools. Taking into account the more advantaged students that new schools serve the Daily News reported that “on 2012 state reading test scores for 154 public elementary and middle schools that have opened since Mayor Bloomberg took office, nearly 60% had passing rates that were lower than older schools with similar poverty rates…Of 133 new elementary and middle schools that got letter grades last year, 15% received D’s and F’s — far more than the city average.” Old schools have a student college readiness rate 10 percentage points higher and a graduation rate 1% higher than those of the new schools. Schools considered for closure by the DOE have the same proportion of new schools as old schools (about 5% of all the schools in each category). When new and old schools, with comparable percentages of self-contained students, are compared old schools do better. The old schools have a 3% higher graduation rate and 11% more schools getting A’s or B’s on the school report cards. New middle schools are overrepresented on the NYS Focus list of struggling schools. They are 58% of the schools on the Focus list, although they are only 43% of schools overall. Looking specifically at schools working with the most disadvantaged students (the quartile of highest student need) the gap is even wider, with new schools being 66% of the Focus list and only 40% of all schools. Comparing new and old schools by decile of student need, new schools have lower college readiness rates across the board while passing a much larger proportion of students.

Part IV. Has the portfolio strategy based on market-choice and charter schools improved student outcomes?
The DOE’s portfolio/choice strategy has not addressed extreme divergence in school outcomes and the lack of diversity in schools. The DOE’s data sets show that schools in NYC are not providing students with equal opportunity. SAT scores- in only 28 out of 422 schools with reported data did the average critical reading score match or beat the national average score of 496 in 2012. In only 31 out of 422 schools with reported data did the average math score meet or beat the national average score of 514. Only 28 schools had scores that meet or beat the national average of 422 in writing. Advanced Placement exams- in over 40% of schools not a single student took and passed an AP exam last year. In only 56 schools, out of the 468 with reported data, did more than 50% of students pass the AP exams they took. Eight schools account for over half of the number of AP exams NYC students passed last year. High school Advanced Regents Diploma graduation rate- only 20 schools out of 419 with reported data had 50% or more of their students graduate with this college preparatory diploma last year. College readiness- in only 30 schools out of 407 with reported data did 50% or more of students graduate with Math and English score that New York State consider indicative of college readiness.
De facto education redlining continues to exist in NYC with extreme inequities in educational opportunity across districts. One report concluded “low-income children and children of color are not receiving the benefits of school integration.” A report from New York University on the school choice process concluded “the decline over time in the number of academically mixed, educational option high schools is notable.” A report from Brown University found that “eighteen of the twenty-one neighborhoods with the lowest college-readiness rates are in the Bronx…thirteen of the fifteen neighborhoods with the highest college readiness rates are in Manhattan” and concluded that “high school choice seems not to have provided equity of outcomes for the city’s high school students.” The Independent Budget Office report analyzed data and found that African American, Asian, and White students in NYC now attend less diverse middle and high schools than in the past. A New York Times infographic shows that although neighborhood diversity increased in NYC the typical African-American student’s school decreased in diversity. An audit of the high school admissions process by the NYC Comptroller’s Office concluded “we do not have reasonable assurance that the possibility of inappropriate manipulation of the student rankings, favoritism, or fraud is being adequately controlled.” The number of Black and Hispanic students at the city’s specialized high schools has decreased although the tests used to determine admissions decisions are seriously flawed. One analysis found that the schools serving the most advantaged student populations have over 70% more “proficient” students than the schools serving the most disadvantaged students. Another analysis found not a single school serving and holding onto (i.e. without large cohort attrition rates) 90+% students living in poverty in the top half of city schools for English/Math proficiency.

Cohort attrition at charter schools is so high that parents end up with limited choice. At some charter schools 24%-68% of the students are lost from each cohort. Up to 7 out of 10 parents at these charter schools do not see their child complete schooling at the charter school they chose. Other “high performing” charter schools suspend 25%-40% of their students a year in order to see gains in test scores. This means that each year up to 2 in 5 parents at these charter schools have their choice forcibly taken away by the very charter school they chose to send their child to. In one particularly egregious case a charter school pushed out 1/3 of its student body in order to improve test scores. Unlike public schools, charter schools are able to expel students and generally do not backfill the vacated seats in the cohort.

Charter schools do not serve a representative student population. If you are the parent of an English Language Learner or of a student with special needs you won’t have much choice since charter schools tend to accept very few of those students. And if they do accept your child it seems that at least some charter school chains will attrite English Language Learners and students with special needs at very high rates. An academic research paper found that “English language learners are consistently underrepresented in charter school populations across 3 academic years.” An analysis of two districts found that the charter schools in those districts served 31% fewer students with low incoming Math scores, 18% fewer students with low incoming English scores, and 16% fewer special education students.

Charter schools are not transparent about their data and finances. The DOE under Mike Bloomberg refused to share data on special education services in charter schools. A charter school chain sued New York State to prevent an audit of how it used public money. New York State backed down. Joel Klein, former DOE Chancellor, falsely claimed that charter schools “closed the longstanding achievement gap.” He made this claim even though the data showed it to be false. In 2007, when the big political push to open up more charter schools began, the data showed that charter high schools had an on-time graduation rate less than half that of public schools. Even so more charter schools were opened. As many sources have pointed out very little of the data that can be found for public view on the official web pages of public schools can be found on the official web pages of charter schools.

Charter schools do not have better scores than public schools. In 2009 a report showed that students in charter schools made less progress than those in public schools. In 2010 the data showed that public schools were 24% more likely to get A’s or B’s on the NYC school report cards than charter schools. In 2011 yet another analysis showed that charter schools are more likely to get D’s or F’s on the progress section of the NYC school report cards than public schools. In fact, charter schools were twice as likely to get F’s as public schools. Charter high schools had half the college readiness rate of public high schools. This past year charter schools saw bigger drops in performance on the Common Core exams than public schools. Additionally charter schools performed worse on average than public schools in English and the same as public schools in math. This is all the more concerning given the creaming, the extremely high suspension and alarming attrition rates. Despite these competitive “advantages” charter schools overall do worse than public schools. A report, funded by conservative groups, claiming the opposite had significant flaws and a review of the data found that, in fact, charter schools had student outcomes 6.5% below that of similar schools. In 2012-13 Charter schools on average were at the 46^th percentile in English and the 53^rd percentile in Math growth. Focusing on the students that charter schools claim to be dedicated to serving, namely students who most need great schools and great teaching, they do even worse. The data reveal the sort of job charter schools are doing educating students who scored in the lowest third the year prior. Looking exclusively at progress with these high-needs student charters are at the 41^st percentile in English and the 45^th percentile in Math. This means that they are doing a below average job as compared to other schools in serving this population of students. According to teacher value-add metrics, an admittedly unreliable measure as we will see below, charter schools on average are not adding as much value as non-charter public schools in ELA and are adding about as much value in Math.

Charter schools are funded at higher levels than public schools. As a whole charter schools in public buildings receive almost $650 more per student in public money than public schools. When the fact that charter schools have fewer high needs student is accounted for charter schools in public buildings receive $2,200 more per student in public money than public schools. Many charter schools spend a lot more money per student than public schools. KIPP spends over $3,000 more per student. Other well-known charter chains spend $4,300 more per student than public schools. When charter schools are “co-located” with public schools they take resources such as libraries, science labs and computer rooms from the existing public school.

Student outcomes have not improved compared to similar districts, which did not implement the market-based reforms reviewed above, over the past 12 years. On the National Assessment of Education Progress (the only multi-year national measuring stick) Trial Urban District Assessment, NYC had lower growth in 8^th grade reading and math by an average of 5 points and a single point of improved growth in 4^th grade reading and math as compared to other large urban districts. The achievement gap increased by 3%. Sorted by demographic group NYC is second to last among large cities. NYC’s SAT scores declined by 20 points over the past 12 years, a larger decline than would be expected even with more students taking the exam. An educational impact statement prepared by the Coalition for Educational Justice found declining SAT and AP outcomes for Black and Hispanic students.
The high school graduation rate increased and there is plentiful evidence that this is due to changing student demographics, the lowering of standards, and the manipulation of metrics rather than educational progress. The under 18 population in NYC changed from 2000-2010 with a 15% decline among Blacks/African Americans, a 15% increase among Whites and 28% increase among Asians. Yonkers, the only demographically similar “Big 5” city in NYS has seen its graduation rate increase by 9% since 2008 while NYC’s has stalled.
The grading curve on the Algebra exam was lowered by over 25 points over this time period and the curve on the United States History exam by 13 points. At the same time, the content grew less rigorous and multiple choice questions compromised ever larger proportions of the exams. As noted earlier, credits were granted at a rapidly increasing rate and Regents exam scores in NYC were inflated as compared to the rest of New York State. A 2009 NYC Comptroller’s Office audit “identified significant weaknesses that DOE has not addressed to help prevent or detect the manipulation of test scores.”
There is also evidence that schools started to cut corners to increase the graduation rate. A 2009 audit by the NYC Comptroller’s Office found that “schools 1) awarded students multiple credits for passing the same course two or more times 2) made numerous changes to transcripts without sufficient explanation and 3) did not maintain evidence that all transcript changes were properly approved.” It took another 3 years and pressure from the New York State Education Department (NYSED) for the DOE to follow-up. A 2012 internal audit “found problems at 55 out of 60 high schools reviewed… including the improper grading of Regents exams, the graduation of students who did not meet credit and testing requirements, the awarding of credits for work not performed, and gaps in reporting about students who supposedly switched to other schools.” It took additional pressure from the Commissioner of the NYSED for the DOE to finally begin to address the abuse of “credit recovery programs,” four years after the New York Times had reported on the issue. Data released after a Freedom of Information Law request revealed that in schools using credit recovery 2.6% of all credits were earned through this often unrigorous process. The graduation rate in 2012 got a bump of at least 2% from credit recovery. A 2009 report by a researcher at Columbia University found that from 2000 to 2007 the number of students discharged (and therefore not counted against the graduation rate) from NYC public schools increased by 3.5%. The 2007 graduation rate, reported as 62% by the DOE, was actually 43.6% when discharged students are factored in. A 2011 audit by the New York State Comptroller’s Office found that 14.8% of randomly selected general education students and 20% of special education students the DOE coded as discharged should, in fact, have been classified as dropouts. It is disturbing to note that many of the fixes designed to address these issues were only put in place over the last year, when they will have no impact on the “numbers” for the Bloomberg-era.

Part V. What new policy directions should be considered for implementation?
A review of market-oriented reform policies in Chicago, Washington D.C., and New York found them to be largely failures. An in-depth review of the impact of the reforms reviewed found that the results in NYC as trumpeted by the media did not reflect the facts. Looking ahead, what specific alternative and improved policies should be implemented in the Big Apple?

An overall change from the zero sum game that characterized the Bloomberg years to a collaborative approach in which educators, parents and the DOE work together to improve all schools will definitely help. Improved tone will allow for an open and honest conversation on how to improve the experiences of our children in our schools. Parent groups have put forward specific proposals to bring back parent voice to schools. Limiting the focus on “accountability” will save hundreds of millions of dollars that can then be spent on supporting students and families via after school and summer programs. It will also help address the perverse incentives created by high-stakes accountability which encourages schools to avoid serving high-needs students, English Language Learners, and students with disabilities. As long as schools are penalized for working with such students a sad unfortunate version of “pass the hot potato” ensues and initiatives such as the special education reforms will have limited success.
It must be acknowledged however that many policy decisions are out of the DOE’s hands. NYSED signed on to Race to the Top with its emphasis on a free and largely unregulated education market, a form of corporate-style management not used by most high-performing corporations, and an extreme emphasis on test scores. An investigative reporter revealed many of these reforms are driven by a privately funded group within NYSED “not bound by Public Officer’s Law or ethics rules imposed on government officials.” As a former NYSED analyst recently testified “testing has been a defective engine driving the train.” Changing this is beyond the authority of NYC. Nonetheless, NYC should advocate with NYSED to allow students flexible options, in addition to standardized exams, to meet graduation requirements. This should include portfolios, demonstrations, and presentations and should be open to all students at all schools, not just the NYC Consortium Schools. Students should be able to be promoted and to graduate without having to score above a random cut-off number on a particular day on a specific test, as long as they can demonstrate learning in other ways. NYC should advocate for changes to the teacher/principal evaluation system with its reliance on invalid value-add metrics. Instead, teacher-driven goal setting on demonstrated student mastery of course standards should be used as a valid local measure. Educators would find the professional conversations around student success in their courses more meaningful than constant standardized testing. At the least, NYC should demand that NYSED prove that they have corrected all the problems with value-add metrics described earlier. Additionally, NYC should demand that NYS fulfill its funding obligations to NYC schools to the tune of the $3 billion shortfall they have refused to pay out.

There are a number of policy changes, entirely in NYC’s hands, that should be part of the conversations with parents, educators, students and communities as we blaze a new path:
Governance and Organizational Structures

• Strengthen the role of the Panel for Educational Policy. The members should be selected for single fixed terms of office. The number of members on the panel should be increased to ensure greater diversity and more representation by the citizens of the city. The panel will then collaborate with the mayor in ensuring that community voice is heard. The Community Education Councils will be able to independently bring issues and concerns before the PEP.
• Create a Youth Council, building on existing groups such as the Youth Researchers for a New Educational System, so that student voice is heard. Research in NYC shows that student policy level involvement and participation increases equity, opportunity, and engagement.
• Create an independent research office to evaluate educational initiatives. Embed independent researchers in central and district offices to help evaluate the success of initiatives and to suggest new directions if necessary. This office should report to the Panel for Educational Policy. Complete comprehensive data sets of all DOE generated data, with student identifiers removed, should be posted online so that independent researchers can analyze and identify patterns and trends for possible follow-up by the DOE and the PEP.
• Re-organize Tweed. The numerous program managers, directors, senior directors, deputy executive directors, executive directors, deputy CEOs, and CEOs should be streamlined so that every employee has clear goals and responsibilities. Only experienced educators with proven track records or people with proven oversight/control experience should be hired. Their work must be completed more efficiently and must directly support the work of schools. For example, enrollment and budgets should be finalized well in advance of the next school year (March at the latest). This will allow schools to complete hiring earlier, a strategy that research from NYC suggests results in positive outcomes.
• Networks should be disbanded and a geographically-based structure that combines the support role of network leaders with the supervisory role of superintendents should be developed. Under the current split-function, both supervision and support functions receive less than full attention. Superintendents are overwhelmed by the number of principals they supervise and therefore often have little on-the-ground knowledge of the schools. Network leaders are overwhelmed by the compliance and back-office responsibilities that have devolved to the networks. Additionally, it is widely understood that although network leaders do not officially rate, select, or fire principals they are deeply involved in such decisions. Combining these functions so that supervisors are focused on school improvement will pay dividends. In order to accomplish this, the nearly 1,000 current network personnel would be divided between instructional/youth development functions and compliance/HR/back-office functions. Each of the new superintendents will oversee/support no more than 15 or so schools. This would allow the superintendents to visit each of their schools multiple times each month, providing increased support and oversight. This will also allow the new superintendents to produce yearly short narratives of each school, on a staggered schedule, sharing strengths and areas of growth with parents. Another narrative report will be developed for the educators at the school. This will replace the School Quality Reviews, announced visits by outsiders that have become a dog and pony show. The savings from ending Quality Reviews will be returned to schools to support enriching field trips for students and families. Research has shown that well-designed field trips have significant positive impact on student critical thinking, historical empathy, and tolerance. The new superintendents will focus on instruction and youth development and will be supported by a team of 4-5 experts in these fields and in school data analysis. Each high school superintendent will have a dedicated programming/data specialist on the team to ensure that each school’s course sequencing/flow meets student needs. There have been more than enough disasters in this area. Programming is an often overlooked but very powerful lever for school improvement. After all the program determines where every single student and teacher is every period of every day. The remaining back-office functions around compliance, HR, budgets, legal, and the like would be carried out by a handful of regional geographically-based teams staffed by the remaining 600 or so network personnel. This structure provides a differentiated structure of autonomy, support, and supervision for schools. A recent petition to keep networks was signed by fewer than 7.5% of principals, many of whom were selected to their positions by the very same networks and some of whom have said that they do not agree with the petition. Superintendents will need to be recruited through citywide and national searches, as many of the current superintendents, in their sidelined and minimalistic role, were not selected for their instructional/school support abilities. Finding and recruiting talent should not be difficult with the role and scope (even groups of only 15 schools is larger than most entire districts outside of NYC) of the new superintendent positions. Principals will form non-geographic affinity groups, facilitated by senior successful principals, to discuss and share practices across other non-geographic dimensions.
• Implement school budgeting in a fair and consistent manner. Fund all schools at 100% of the funding formula. Correct the weights to reflect true student needs such as weights based on ELL level and temporary housing status. Fund students with disabilities at the actual cost of the program each student receives. The money to do this can be found by ending the practice of special grants, programs, appeals and other off-budget items that are handed out with little transparency and accountability. Re-consider the policy of charging schools for the salaries of each teacher. This policy was designed to hold schools fiscally responsible for having a more senior staff, but there is no evidence that it helps students. Instead provide schools with a weighted teacher formula so schools receive an extra weight for 1st and 2nd year teachers on staff to support them, through one period of team teaching with a senior experienced educator, as they build capacity. A report on other large districts that fund schools based on student need reveals that fewer than 15% of such districts charge schools directly for the salaries of their teachers. NYC should provide each school with a specific number of faculty “units” that are not dependent on salary/experience level. Additional units should be given for the early grades at high needs elementary schools to lower class sizes in a targeted way. All other funds should be distributed via FSF for schools to use as they determine is most beneficial which would include funding additional staff and any other resources. Currently school budget data can only be found through a tedious school-by-school process. Make the data public and easily accessible so that New Yorkers can see how schools are funded.
• Discover and spread innovative practices from the ground up. Close the Innovation Zone office as top- down innovation is unlikely to work. Create challenge prizes modeled after the examples here http://challenge.gov/ and here http://www.xprize.org/. Educators and citizens would be able to submit specific challenges for inclusion on the DOE challenge list with prizes for workable and scalable solutions.

Student and School Support

• Develop comprehensive early intervention and support services for students. Increase the number of speech teachers and math and reading intervention specialists in elementary schools and train all teachers on specific literacy and math intervention programs. This will require developing a citywide early warning system using indicators that have already been identified for the NYC context and specialized curriculum to identify and provide quality remedial opportunities to students who are falling behind.
• Assist elementary schools in adding middle school grades and middle schools in adding elementary school grades. Research in NYC has shown that K-8 configurations work better for students.
• Provide schools with expert support and guidance in curriculum. Curricula should, of course, include the fullest range of courses for students, including the arts which have intrinsic value and have also been shown to support improved graduation rates in NYC. Curricula should also include guidance and post-secondary counseling. We cannot take a sink or swim approach to teaching and learning, with every school left to their own devices. The Office of Teaching and Learning must be re-opened after having been shuttered under Bloomberg. Truly expert teachers must be identified at each grade level and subject area, their lessons videoed, their materials copied, and all of such resources must be shared with teachers throughout the city. Curricula must be developed at all grade levels, in all subjects and for all student populations. Curricula development should especially focus on developing interdisciplinary connections and projects. We should work towards the creation of interdisciplinary courses such as combined math and science or English and history classes that can be taught by teachers with certification in one of the disciplines thereby decreasing the number of students each teacher is working with. There is no one curriculum that works for every student and every school. Options and specific supports for students at different levels must be part of any curriculum. A platform should be developed for teachers to share and evaluate the usefulness of the customized changes they make for their students.
• Tap into recent retirees, an underutilized resource, to support schools that are struggling. It is wrong to continue to close schools just because they serve a high-needs student population. Teams of experts must be formed to work directly with such schools in the areas of programming, data, and instructional cohesion. Each team must be assigned to one school to ensure quality support. The DOE and the UFT should collaborate on staffing teams with recently retired educators willing to work part-time who want to continue to contribute to schools and on evaluating results for further interventions if necessary.
• The new administration has prioritized pre-K. This is long overdue as research shows that the achievement gap begins at the very earliest stages of the lives of our children. Growing income inequality over the past dozen years has led to poorer early childhood outcomes such as the increasing proportion of pre-term births in NYC since 2000. All the resources devoted to caring for and supporting parents and children from pregnancy through school enrollment should be combined in a single agency. We need to do everything possible to provide quality care such as home visits to parents in NYC. We need to partner with CUNY to ensure that we train and then hire the highest quality pre-K teachers and workers.
• Provide schools with continuous feedback on how they are doing throughout the course of the year. Do not grade schools with a single letter, months after the school year ends. No teacher would ever use such a grading practice in the classroom.
• Provide students with the additional quality learning time they need to succeed. The school year should be extended with a shorter summer break in the month of July and the new school year beginning again at the start of August. Summer learning loss is a huge factor in lower student outcomes and we must address it system-wide. After-school programs should be expanded and should primarily offer students enriched social and cultural experiences.
• Build comprehensive learning communities offering a full array of advanced and AP courses, remedial coursework, College Now, CTE programs, GED and adult education options at all high school campuses. The data show that new small schools are often unable to offer students a variety of college preparatory courses. Many schools with specific CTE themes don’t have enough students genuinely interested in the program to develop a quality CTE option. Opening up such specialized courses to all students on a campus will build an academic community, communities that until now have only formed around the campus sports teams. A temporary moratorium on establishing most new schools, other than in exceptional cases of not enough seat capacity in specific neighborhoods, would allow resources to go towards building these rich campus-wide options, including 21^st century CTE programs, for hundreds of existing schools. This will improve college and career readiness rates across all NYC schools. Ensuring that GED and adult education courses are also offered through every single high school and campus will increase the responsibility of every school to support each student through to graduation. It will also help address the poor outcomes of the current GED and adult education programs in the Alternative Schools and Programs District. These campus-wide programs will be ably supervised and supported by the new geographic-based superintendents. They will have the authority to ensure that these programs are well-aligned and that all schools on a campus are collaborating on providing access to these programs to all students.
• Ensure that every single school has as diverse a student body as possible. Whether G&T programs, screened or specialized high schools, all schools must have a student body that reflects the diversity of NYC. The Office of Enrollment must improve their systems so that diversity is a crucial element of the process. Analyze the data on student characteristics to ensure that each school has a student body representative of the diversity of NYC. The Office of Student Enrollment should be held accountable for preventing the clustering of specific sorts of students in specific schools.
• Negotiate contract changes to allow middle and high schools later start times that are better suited for the circadian rhythms of adolescents. Negotiate changes to allow elementary teachers to develop experience in and then have priority rights to specific grade levels. Both of these contract modifications are supported by the research in NYC.
• Hold charter schools to the same standards of transparency and accountability as public schools if they want free rent and other NYC resources. They must be transparent about their budgets and their spending. This information must be posted on each school’s website, just as it is with public schools. Charter schools must be held accountable for every student who walks through their doors. They must no longer be allowed to transfer out 50+% percent of their students, but must work to educate and provide supports for every single student. An ombudsman position must be created so that parents of students in charter schools are able to have any issues, such as mistreatment of students with disabilities, addressed. Any new charter schools must be dedicated to the mission of serving truly underserved student populations as do the ROADS and Urban Dove charter schools.
• Improve information sharing about schools with parents and students. Share a broad array of information about schools transparently and clearly. This should include, in addition to how students do on tests as compared to similarly situated students, such information as arts offerings, clubs, years of teacher experience, suspension rates, % of students leaving the school prior to natural transition point, and videos of classes for parents and students to view. Develop a website and apps that allow parents and students to weigh this information at the level of priority important to them. Websites like this already exist, such as this one that allow the user to rank graduate programs based on individual priorities. Publish test score data using ranges to account for levels of statistical significance and include multiple years of data to account for meaningless year-to-year fluctuations. Create a system so that parents and students can write reviews of schools and publish that information on the website after a peer vetting and review process. Ensure that parents from every single school have timely updates on the children’s progress via an online grading system.
• Use data in positive ways to identify specific teachers and departments that have outstanding results year after year. Data is crucial for improving education, as long as it is not limited to test scores and as long as it is used in transparent and honest ways. Given the misuse of data, of which numerous examples were cited above, it is critical that this be done in a way that educators find honest and professionally supportive. For example the Global History departments at the following schools have had results in the top 25% of all schools every year for the past 7 years accounting for, at least partially, incoming students characteristics: Bedford Academy HS, HS for Arts and Business, UA School of Design and Construction. The United States History departments at the following schools have had results in the top 25% of all schools every year for the past 7 years: HS for Arts and Business, Manhattan Bridges, Renaissance HS for Musical Theater and Technology. Let’s identify and do qualitative studies on what, if anything, these schools are doing well. Use technology platforms to have those teachers and departments share their practices and lessons across the city.
• Use data to improve NYC schools. Instead of using data for political and ideological ends let’s start using data, only the statistically significant and meaningful data that is, to support and improve schools. Analyze the data to see if some schools have large gaps between course pass rates and Regents exam performance (including students who took a course but did not sit for the Regents exam). Support such schools in clarifying grading practices. Analyze the data to see if some schools have large gaps between graduation rate and student persistence in college. Support such schools in increasing the rigor of their academics and in building life-skills of students. Analyze the data to see if some schools lose, perhaps as a deliberate strategy to make their numbers look good, a large proportion of their students from each cohort. Support such schools in working with the every student who enters their doors and in lowering their attrition rate. Provide every school community with a data narrative identifying the long-term, multi-year trends and support each school in working to shift practices if necessary.

Teacher and Principal Development and Support

• Select only proven and successful teachers to Assistant Principalships. Only Assistant Principals with proven track records should be hired for Principalships. Close down the Leadership Academy, which has had poor outcomes. Once the organizational structure suggested above has been implemented and quality superintendents have filled those positions they should carefully evaluate and make retention decisions about the existing principals. Quality school leaders must then be identified and developed over time. Current principals should offer sessions after school and over the summer in their areas of particular expertise for all educators. Teachers interested in administration positions would then be able to demonstrate their growing capacity for school leadership over time through performance tasks and on- the-job leadership.
• Develop a clear job description for principals. Right now the DOE has created a paradoxical and hypocritical situation in which principals are told they should be instructional leaders while, at the same time, they are held responsible for numerous reporting and compliance duties that take up tremendous amounts of time. This has come about due to the fact that principal autonomy was more of a slogan than a reality. A clear set of responsibilities focused on the principal as chief teacher and coach must be communicated. A Director of Compliance role should be established to handle the other managerial aspects of running a school. The principal must become the leader of the school community and all the members of the community, including school safety agents, custodians, and food service staff, must report to the principal. This will allow the academic and youth development goals of each school to be the primary focus of all adults in the building. Under the current structure school safety, for example, has become counterproductive with overemphasis on policing and punishment rather than restorative justice approaches. A simple change in reporting structure can help fix that.
• Establish specific and clear outcomes required for the granting of tenure from the start of each teacher’s career. These outcomes should demonstrate the development of professional skill and expertise. Criteria should focus on the development of high-quality courses and class materials for each subject area in a teacher’s content area. The National Board for Professional Teaching Standards certification process should serve as a model.
• Develop a career ladder for teachers, supporting teachers in developing an area of professional expertise such as literacy interventions, remedial math instruction, or the use of technology in lessons. These teachers would teach 2-3 periods a day in their home school and then facilitate professional development in their area of expertise across school in their superintendency. This will save tens of millions of dollars yearly that are now paid out to consultants and for-profit groups. We have and can further develop expertise among the 75,000 NYC teachers and do not need to regularly call on outsiders without a deep current understanding of the classroom.
• Negotiate additional pay for educators with stellar attendance (the contract refers to 7 sick and 3 personal days per year that are paid out at ½ their value upon leaving the DOE). Research shows significant impacts of teacher absences on student achievement. Paying teachers from 50%-75% of the value of days not taken (with the higher rate given to teachers with a higher proportion of unused days) above a floor (say 3 unused sick days) at the end of each school year will have positive impacts on students.
• DOE and UFT should collaborate on a required pre-service summer training course for all new teachers. Data show that the first year of teaching is the most difficult for teachers resulting in the poorest student outcomes. Much of this is due to the fact that teachers are struggling with figuring out the curriculum and deciding what/how to teach content. Requiring every single new hire to take a summer course, taught by expert experienced teachers who share their teaching resources, on the specific curriculum the new teacher will be teaching over the school year will pay dividends.
• DOE and UFT should collaborate on interviewing and selecting teachers into a hiring pool. The data shows that currently many teachers are hired at the last minute without going through a well-designed screening process. Much has been written about Finland’s approach to teaching as a profession, suggesting that improving the selectivity and quality of our teacher training programs would be of great benefit. That is, of course, not something NYC can address. But we can improve the selection and hiring of teachers. A mix of assessments, performance tasks, portfolio presentations, and group problem-solving scenarios/interviews should be created to screen teachers. This should include the Haberman Screener , the Math Knowledge for Teaching performance assessment, Principals, parents, school staff, teachers, and students should form screening interview teams that will evaluate NYC teacher candidates on evidence of personal efficacy and other critical traits. All hiring must then be done from within this pool. Funding for this can come from the closing of the NYC Teaching Fellows program, a program that may have served a purpose when it was difficult to find credentialed teachers to teach in NYC and no longer has a purpose.