Class Size Matters Testimony on SHSAT before the Committee on Education and Jointly with the Committee on Civil and Human Rights

Testimony before the Committee on Education and Jointly with the Committee on Civil and Human Rights

PDF of the testimony is here.

May 1, 2019

 Thank you for the opportunity to testify before you today.  My name is Leonie Haimson and I’m the executive director of Class Size Matters.  I would like to testify in support of Res 0196-2018 and T2019-4317 and in opposition to Res 0417-2018.

No other school district in the country bases admissions to any one of their schools on the basis of a single high-stakes test.  Moreover,  this practice has long been opposed by the American Psychologic Association, the National Council on Measurement in Education, and the American Education Research Association who write: “Any decision about a student’s continued education, such as retention, tracking, or graduation, should not be based on the results of a single test, but should include other relevant and valid information.” [1]

As the National Academy of Sciences has explained, “current psychometric standards… recommend that a decision that will have a major impact on a test taker should not be made solely or automatically on the basis of a single test score, and that other relevant information about the student’s knowledge and skills should also be taken into account.”[2]

To make things worse, the SHSAT is an invalid and biased exam.  While nearly all of the discussion and debate has so far revolved around the issue of racial disparities, it has also been shown conclusively to be gender biased.  Though NYC girls receive higher average test scores on the state exams in both ELA and math and better grades, they are accepted into the specialized high schools at much lower rates.

Here are this year’s results by gender, revealing an admissions gender gap of eight percentage points.

Gender #stud tested % students tested #got offer % of total offers
F 14,116 51% 2,206 46%
M 13,405 49% 2,592 54%
Total 27,521 100% 4,798 100%

I discussed this gender bias in an article last year in the Gotham Gazette (also attached)[3]; as did Jonathan Taylor in more detail, in a subsequent piece in the  Gotham Gazette. [4]

Jonathan Taylor has also published his findings in a peer-reviewed journal, showing that grades are  more predictive of student success at the specialized high schools than test scores; and that girls who enter Stuyvesant with the same test scores as boys do better on their course work and receive higher grades, including in the most advanced courses.[5]

If we really want more diverse, integrated schools, we should eliminate the use of a single high stakes exam for admissions and instead rely on multiple measures, including grades and more holistic factors.  In addition, we should discourage separate gifted programs and tracking as much as possible – another form of segregation that occurs within schools that merely widens the achievement gap between racial and ethnic groups. [6]  Screening children by their purported “ability” at significantly disadvantages those who are concentrated in the lower–performing classes.[7] Moreover, the identification of children at an early age who are ostensibly “gifted” is highly unreliable; the majority of children who score in the top percentiles in first grade do not retain this status for more than a year or two. [8]

Teachers often understandably complain that it is too difficult to individualize instruction with students of different achievement levels, and indeed it is especially difficult given the large class sizes we have in NYC.  But if class sizes were lowered, this would make teachers’ jobs much easier.  Even more importantly, class size reduction is one of very few reforms proven to work to narrow the achievement gap. [9]

In NYC, our class sizes have increased substantially since 2007 and are 15% to 30% percent larger on average than class sizes in the rest of the state. More than 336,165 students were crammed into classes of 30 or more this fall. In the early grades, the number of first-through-third-graders in classes of 30 or more has ballooned by nearly 3000 percent since 2007.  Our schools will never be able to  provide students with an equitable chance to learn with classes this large.

In Finland, when the government decided to stop tracking, the national teachers union successfully demanded systematic reductions in class size, to ensure that they could meet the needs of all students of different academic levels.  Both the elimination of tracking and the concurrent lowering of class sizes contributed to the rapid improvement of Finnish schools in the 1970’s, along with the elimination of most standardized tests.[10]

If instead, as some have suggested, our schools were to add more test prep, more gifted classes and/or more specialized high schools,  we would instead be moving backward as a city.  We would be replicating the same damaging practices that have undermined educational opportunity in our schools and further exacerbating stratification and segregation by race and class.

[1] see also:

[2] National Research Council. 1999. High Stakes: Testing for Tracking, Promotion, and Graduation. Washington, DC: The National Academies Press, p169 at:


[4] ; Taylor’s findings were also reported here:

[5] Jonathan Taylor, Fairness To Gifted Girls: Admissions To New York City’s Elite Public High SchoolsJournal of Women and Minorities in Science and Engineering, 25(1): 75–91 (2019).

[6] ; see also


[8] David Lohman and Katrina Korb, “Gifted Today but Not Tomorrow? Longitudinal Changes in Ability and Achievement during Elementary School,” Journal for the Education of the Gifted, June 1, 2006

[9] See the numerous research studies at:

[10] Samuel Abrams, Education and the Commercial Mindset, 2016 , p. 281 and footnote 3 on p. 382. Also


Missing Pieces of the Discussion Around Specialized High Schools and City Education

June 22, 2018 | by Leonie Haimson

On June 2, Mayor Bill de Blasio announced in Chalkbeat that he was urging the State Legislature to change the admissions system at the city’s eight specialized high schools, which relies on a single high-stakes exam called the SHSAT. Only 10 percent of students admitted to these selective high schools are black and Hispanic, while these students make up 67 percent of the overall public school population. This year, only 10 black students were offered admission to the city’s most selective of these high schools, Stuyvesant, out of 902 students admitted.

The Mayor and the Chancellor have proposed that admissions instead depend on a combination of a student’s school ranking in terms of grades and state test scores. In the meantime, before the state law is changed, de Blasio plans to expand the “discovery program,” a special program for disadvantaged students near the cut-off score on the SHSAT, to admit them into these schools after extra academic preparation.

As City Council Education Chair Mark Treyger later pointed out on Twitter, the entire effort was announced with little preparation; and ”key stakeholders were also not consulted” on something “dropped 11 days before the end of the [Legislative] session.” The proposal has aroused much controversy, and will not come to a final vote this year, as neither the Assembly nor the Senate was prepared to pass it so late in the session that just ended. Yet the issue will be surely be reconsidered again when the Legislature reconvenes next year, and the debate continues, including over what the City can and should do on its own.

Despite the fact that much ink has been spilled and emotions aroused since de Blasio made his announcement, several aspects of this hot-button issue remain under-discussed:

  1. This move is long overdue. New York City is the only school district in the entire nation where admissions to any high school depends solely on the results of a single high stakes test. This was confirmed by Chester Finn of the Hoover Institute, a conservative education advocate who co-authored a book,“Exam Schools: Inside America’s Most Selective Public High Schools.” Advocates have made efforts for at least 50 years to open up the admission process to these schools and make it more fair.  The Hecht-Calandra Act of 1971 in New York, which specified that specialized high schools must rely solely on a single exam for entry, was passed in the first place in response to such a campaign. Bill de Blasio also promised to reform this admissions process  when he first ran for Mayor in 2013.
  2. The reality is that relying solely on a single high-stakes test for admissions, grade retention, or any important decision in a student’s educational career is unfair, unreliable, and likely to have a racially disparate impact, as pointed out by the National Academy of Sciences nearly 20 years ago in its seminal report, High Stakes, Testing for Tracking, Promotion, and Graduation.
  3. The SHSAT, produced by Pearson, is a highly peculiar exam that has never been independently assessed for racial bias. This was pointed out by Joshua Feinman in 2008, and confirmed more recently by the NAACP Legal Defense Fund in its 2012 Civil Rights complaint about the use of this exam. The results of this test also appear to be gender biased, as girls tend to score significantly higher on state exams and receive better grades, but score lower than boys on the SHSAT. (Girls were only admitted to Stuyvesantand Brooklyn Techin 1969-1970.) The test is quirky in other ways and is scored to give extra points to students who do exceptionally well on the ELA or the math section – rather than those students who score well on both subjects. It also has poorly worded questions – see, for example, the first question in this sample exam.
  4. The mayor could alter the admissions tests at five of the eight specialized high schools immediately – without any act of the state Legislature. Only Stuyvesant, Bronx Science, and Brooklyn Tech are named in state law. The other five schools — Staten Island Tech, the High School of American Studies, the High School for Math, Science, and Engineering at City College, Queens High School for the Sciences, and Brooklyn Latin School — were designated as specialized high schools by Joel Klein when he was city schools chancellor, and could be undesignated as such by Chancellor Richard Carranza.  All it would take is a vote of the Panel on Educational Policy at one of its monthly meetings. It is a panel made up of a supermajority of mayoral appointees. (The best explanation of how only three schools are mandated to use this exam was published in Gotham Gazette. While many other news outlets have gotten this fact wrong in the past, they have recently improved their reporting on this issue.)
  5. Despite the huge amount of time and money many students invest in test prep for the SHSAT exam,  there is research to show that attending a specialized high school has little or no impact on a students’ future SAT scores, chances of enrollment in selective colleges, or college graduation rates – which in turn casts real doubt about the value of attending one of these schools.
  6. Whatever happens to the admissions system at the specialized high schools, there are myriad problems with how admissions to New York City schools have been designed. Many schools utilize a complex and competitive application system overly reliant on test scores. This makes the process of admissions stressful to kids and their parents, and leads to excessive test prep. In many cases, admissions to middle schools are based heavily upon students’ scores on the 4th grade state exams, and 7th grade exams for high school. The state tests and their scoring methods have their own problems in terms of accuracy and validity. In addition, some New York City middle and high schools have developed their own special tests that they rely upon for admissions.
  7. The competitive nature of this process worsened under Mayor Michael Bloomberg and Chancellor Klein. The number of high schools that admitted students through academic screening increased from 29 in 1997 to 112 in 2017, while the proportion of “ed-opt” high schools, designed to accept students at all different levels of achievement, dropped sharply. Even so-called unscreened programs actually do screen students, in covert ways. Moreover, the Gates-funded small schools that proliferated after 2002 initially barred students with disabilities or English language learners from their schools, prompting a civil rights complaint in 2006.

Most of these small schools also required prospective students and their families to attend special “open houses” in the evening, which also tends to box out families with the fewest resources. While  Bloomberg and Klein often bragged about expanding school “choice,” too often this has meant that schools make the choice, not students or families. There is something to be said for comprehensive high schools as exist in most of the country, where students automatically have a right to attend when they enter 9th grade and don’t have to compete to get into.

  1. The method de Blasio now proposes to use for admissions at the specialized high schools would be to give preference to students at the top of their class in terms of grades wherever they attend middle school, as long as their state test scores are also good enough. Yet this method, based upon the admissions process at the University of Texas system, will likely only work to effectively integrate the specialized high schools if city middle schools remain largely segregated.
  2. Even if all New York City high schools became less stratified according to race and class, one would still be left with the biggest problem of all: too many elementary and middle schools are simply not providing the quality of education necessary – especially for students of color. This was revealed by city students’ recent results on the NAEP exams, which showed scores have stagnated – with the only significant change since 2013 being a seven-point decrease in the proportion of fourth-graders proficient in math. The achievement gap among racial and ethnic groups is also larger than ever, in 4th and 8th grade reading.
  3. Class size reduction is one of very few reforms proven to work to narrow the achievement gap, and yet New York City class sizes remain excessive. In fact, class sizes have increased substantially since 2007, and are up to 50 percent larger than class sizes in the rest of the state. More than 290,000 New York City students were crammed into classes of 30 or more this fall. In the early grades, the number of first-through-third-graders in classes of 30 or more has ballooned by an amazing 3800 percent. Yet despite promises to reduce class size when he first ran for office, de Blasio has done nothing to accomplish this.

In January, Class Size Matters, along with nine New York City parents and the Alliance for Quality Education, brought a lawsuit versus the City and the State to require the Department of Education to lower class sizes in our public schools, as the state’s highest court deemed was necessary for students to obtain their constitutional right to a sound basic education. Our lawsuit will be heard in State Supreme Court in July. Whether the city’s public school students will receive an equitable chance to learn may depend on the outcome of this lawsuit, as well as the education policies pursued by Mayor de Blasio and Chancellor Carranza going forward.

Leonie Haimson is the Executive Director of Class Size Matters. On twitter @leoniehaimson.

Note: this column has been corrected to accurately reflect when the schools admitted girls.

Categories Reports, Testimonies, Etc., Testimonies & Comments, Uncategorized, Updates | Tags: | Posted on May 1, 2019

Social Networks: RSS Facebook Twitter Google Stumble Upon Digg Reddit

Comments are closed.