Passing the Test
A collection of schools in New York have been quietly staging a constructive revolution against standardized tests for decades.
By the time New York students have hiked through the public school system, climbing from pre-kindergarten to graduation, they have had an up-close-and-personal relationship with standardized tests.
According to a 2015 study from the Council of Great City Schools, the average big city student takes 112 standardized tests by the time they graduate.
In New York, it’s no different. It starts with the Common Core assessments, the state-sponsored tests required by the federal government every year from third to eighth grade. It ends in high school with the Regents tests, the final hurdle for students searching for a diploma. All New York high school students must pass at least five of these Regents to graduate, with one each in English Language, math, social studies and a fifth test in any of several remaining categories.
At the elite high schools like Stuyvesant in Manhattan — where students are accepted based on yet another state test administered to eighth graders throughout the city — students must pass 10 Regents exams to earn an advanced diploma.
All these tests are seen as a measuring stick, a way for policymakers, educators and parents to assess how the state, city or a single school is performing. They produce digestible numbers that can be easily sorted and quantified.
That’s what those in favor of the tests claim. But critics, who have become increasingly vocal, disagree.
High-stakes exit exams, some educators argue, were designed as a means to evaluate student progress and school success. But too often they have become the central focus of the curriculum itself. The alternative, according to City-As-School High School Principal Alan Cheng, is project-based learning.
“It’s a belief that standardized testing is an ineffective way of measuring and supporting student learning,” said Cheng. “Performance-based assessments are a good alternative to that.”
City-As in the West Village is one of 38 schools in the New York Performance Standard Consortium, a collection of public schools that get a waiver from the high-stakes Regents tests. Instead their students are measured on Performance-Based Assessment Tasks, including student projects such as research papers, or science projects developed independently by students and presented PhD-style to a panel of teachers. Educators believe performance assessments give a more accurate picture of what a student knows. They require a student to apply deep knowledge rather than regurgitate isolated facts. Standardized testing, performance-assessment advocates claim, doesn’t simulate learning at all.
“There is a better way to assess what kids know and what they really know,” explained Ann Cook, a co-director at Urban Academy High School as well as the executive director of the Consortium. “The tests really don’t do that.”
There’s mounting evidence that standardized tests fail to predict much more than socio-economic status. Studies have found that tests like the SATs measure family income level more accurately than they do actual learning.
Advocates of testing point to graduation rates in New York City, which are on the rise as students are passing the five exams to earn a diploma, yet college readiness and completion rates are dire. Only 35 percent of high schools graduates were deemed “college ready” by the New York City Department of Education in 2015, a figure reflected in the high number of graduates repeating high school courses at city community colleges.
At City University of New York community colleges, nearly 80 percent of new students from New York City high schools enrolled in remedial courses for math and English. Additionally, only 45 percent of students who enrolled at CUNY schools earned a degree in five years, according to figures from the Research Alliance for New York City Public Schools. That rate sinks to 26 percent for students seeking an associate’s degrees at two-year colleges.
But at Consortium Schools, students are flourishing in high school and beyond. These schools outperformed traditional public high schools in nearly every category, according to a 2015 report from the Consortium. Minority students from these schools graduate at higher rates. Same for English-Language Learners, who nearly double the graduation numbers from Regents-based schools — 71 to 37 percent. After high school, minority students are enrolling and staying in college longer as well.
“This is far better preparation for colleges than the test scores,” Cheng said.
City-As, which joined the Consortium in 1997, two years after the coalition started, has been a model of progressive education since its formation in the 1970s. From its start, City-As built a curriculum around credit earned through internships outside of the classroom, which worked especially well for the under-credited students it attracted as a transfer school, or a school specifically designed for students who have dropped-out or fallen behind in credits.
The consortium model made perfect sense for a school built on principles of learning outside the classroom and teacher-created courses. Students at City-As don’t just take social studies, they take “Gun Violence in America” or a gender studies course analyzing why Hillary Clinton lost the 2016 presidential election. They don’t just review math, they learn how to decipher and manage credit.
“I would imagine that if I am a high school student, any high school student, I would rather get to be able to have a choice with the topics that I study,” Cheng said. “I’d rather design parts of my own research topics and go much more in-depth in particular things than study them in a broad way.”
The roots of the Consortium date back to the early 1990s, when then New York State Commissioner of Education Tom Sobol recognized about 30 small public schools for their exemplary performance and asked the Department of Education to use them as potential models for struggling schools. With Cook as one of the leaders at Urban Academy, these schools were defined by their progressives models of education, mainly basing student evaluation on projects rather than tests. In 1995, Sobol officially granted 24 of these schools waivers from the Regents for a five-year period and the Consortium has fought to save their status ever since.
In 2001, more than 1,000 students and parents flocked to Albany to demonstrate in front of the Board of Regents, the panel selected by the state legislature who reviews and sets standards for all New York schools. After another lengthy review process, state lawmakers confirmed the waiver again in 2005. To receive the waiver today, schools first seek approval from Cook and the Consortium, who then submit a bid to the Board of Regents for approval.
Like all consortium schools, students at City-As replace Regents exams with performance tasks in the same subject (except for the English Language Arts assessment, which they still take). Students first complete a research paper on a topic of their choice, typically with a hyper-specific focus, such as review of organic fertilizer for a science assessment or a profit analysis from an internship they completed. Students then present and defend their findings in front of a panel of three teachers. The presentation lasts 40 minutes to an hour, sometimes with the help of PowerPoint, as students outline their research, how that informed their project and their findings. After the student finishes, he or she takes questions for about 15 to 20 minutes. The entire process mimics the same a way doctoral student might defend a thesis.
“People need to be able to think things through,” Cook said. “They need to take evidence and develop their own voice based on evidence.”
The tests, then, reflect a specific philosophy of education. “It’s the matter of whether you see education as the collection of a lot information, or whether you see it as a process,” she added.
The push for performance assessments as an alternative to standardized testing by Cook and the consortium is not a new phenomenon. Resentment toward high-stakes accountability testing in the 1980s led to an increase in states using performance assessments for students. In 1990, at least eight states were using some form of performance assessment in math or science, wth 16 more either developing or piloting similar programs. These templates were like hybrid versions of New York’s consortium model today, with states simply reducing or removing multiple-choice questions from tests (like Kentucky), or submitting student samples of work within a subject in addition to a standardized test (like Vermont).
Nearly all of these programs, though, were phased out by the early 2000s, because of concerns about the reliability and feasibility of evaluating complex, content-based answers. The thought was that the performance-based assessments allow too much variability between evaluators. Exams, particularly multiple-choice ones, are straightforward. They provide objective answers.
With a bubble test, there’s right or wrong, explained David Bloomfield, professor of education at Brooklyn College and City University of New York graduate center. “The Regents are defended by being objective and adhering to a broad factual knowledge,” Bloomfield explained. “It’s a more routinized, standardized type of teaching assessment. It’s kind of a factory model and easy to compare apples to apples. Everybody is taking the same test.”
Politics have also contributed to the rise of high-stakes testing, as well. In New York, Regents exams have been around since 1876, but it wasn’t until 1998 that they became requirements for high school graduation. Four years later, the federal government enacted No Child Left Behind, which used testing to evaluate school and teacher performance as a means to bring poor performing schools up to speed.
The results have been underwhelming. Since the passing of No Child Left Behind in 2002, math and reading scores have displayed little to no gains, while gaps between white and black and hispanics students has remained about the same, according to National Assessment of Educational Progress.
Today, as in the 1990s, there’s again a growing backlash toward standardized testing, particularly from parents. In response to over-testing, students have chosen to “opt-out” of the state Common Core tests in third through eighth grade. In 2015, about 20 percent of New York students declined to take the Common Core tests. Teachers Unions were also upset with the tests, citing a report from the American Statistical Association that argued it’s unfair to grade teacher performance based on testing.
Common Core and the Regents are not directly comparable. Opting out on the former has zero repercussions so far for students (though schools risk losing federal funding), while missing Regents means no diploma for individual children.
Cook sees the opt-out movement as a welcome dent in the army of standardized-testing supporters. In December 2016, the Board of Regents placed a moratorium on linking teacher performance to testing and said it was researching optional graduation requirements for failing students, particularly performance-based assessments.
Other signs of test fatigue are showing elsewhere. Nearly 1,000 accredited colleges and universities do not require or de-emphasize the SAT or ACT for admissions — up from 280 in 1999 — including nationally recognizable institutions like George Washington and Wake Forest universities.
Cook and the Consortium go to great lengths to ensure their schools are effectively evaluated, she said. To be considered a consortium, a school must first ask all teachers and administrators to vote on whether the school wants to switch. The school then has to agree to a set of standards, such as the amount of time allotted for teacher professional development and transparency to the consortium for curriculum and evaluators.
The Consortium also has each school within the group submit a sample of assessments at the end of the year that have been marked as outstanding, good and competent. Then, about 200 teachers re-evaluate those assessments, without knowledge of the original score. Those results are then sent back to the high school. The process allows schools to judge whether they’re assessing their students too hard or too soft and gauge how they match up with similar schools.
For his part, Bloomfield is reserved about the implementation of a citywide consortium for all of New York public schools, a system of 400 high schools encompassing 250,000 students (although he’s an advocate of the model and sent his child to the Beacon School, a consortium high school in Hells Kitchen). Bloomfield imagines some sort of compromise between the two ideas, where a school can adopt aspects of the performance tasks without completely abandoning tests.
Cook, though, is a purist. Many private and parochial schools have eschewed Regents exams for years without the assumption that they’ll somehow be unable to produce educated, college-ready students. The statistical evidence in favor of the performance-based model exists already, she said, not to mention the insistence on college admission officers who cite curriculum and grade point average as the best predictors of college success, not standardized testing scores.
For the performance-assessment model to persist, it all comes down to trust, Cook explained “In order to do the kind of work that consortiums do, you have to trust teachers,” she said. “You have to support them.”
“We think teachers are worth the effort,” she continued. “And they’re probably the ones who know the students the best.”
