An explosive student opt-out movement on Long Island and across the state may reshape standardized testing for years to come, a top state education official says, though exams based on the Common Core academic standards aren't going away any time soon.
Even the staunchest defenders of the state's English Language Arts and math tests, which affect more than 1 million students, acknowledge that pressure for change is mounting. Statewide, anti-testing activists estimated that more than 155,000 students in grades three through eight refused to take this week's English exams, and some school officials expect numbers to rise higher next week, when the math test is administered.
Merryl Tisch, head of the state's policymaking Board of Regents, said in a recent radio interview that growing test refusals could force a switch in the type of tests that are given.
"That is particularly why I am very interested in curtailing this opt-out movement, because in order for us to be able to have a viable test, we need a viable number of students in every district showing up to be tested," Tisch told WNYC's Brian Lehrer on April 7, a week before the English Language Arts test started statewide. "In the absence of that critical number, we will be forced unfortunately, and I truly believe unfortunately, to adopt a national test."
New York already is positioned to make that leap. The state is part of a multistate consortium known as PARCC, which has developed a separate set of computer-based tests that could be introduced within two or three years, if the Regents decide to take that route.
Most on the 17-member Board of Regents are familiar with PARCC -- the Partnership for Assessment of Readiness for College and Careers -- which also bases tests on the Common Core standards. In February 2014, amid an outcry from educators and parents over the rushed rollout of curricula tied to the Common Core and its first, more rigorous tests the previous spring, the board loosened some requirements for testing and teacher evaluations.
As part of that retreat, the Regents backed away from a prospective timetable for implementing PARCC. At the time, a growing number of districts had expressed deep concerns about who would pay for the computers and bandwidth needed for the tests, and had raised questions about practical matters, such as where schools would find the space for scores of students to take tests on the machines.
Since then, PARCC tests have been field-tested in school districts across the state -- including some on the Island -- but the Education Department has not unveiled a concrete proposal for full-scale usage of the computer assessments.
Over the same period, the opt-out movement, far from being curtailed, has grown exponentially.
This week, 71,764 students in grades three through eight in Nassau and Suffolk counties -- more than 42 percent of those eligible -- refused to take the English Language Arts test, a Newsday survey of 110 of the Island's 124 districts found. Last year, nearly 9,500 students in 67 districts that responded to the newspaper's survey declined to participate in the ELA.
Several New York City news outlets have described the movement as driven largely by high-income suburban parents. Newsday's survey found, to the contrary, that districts with the highest percentages of test refusals were mostly middle-class -- for example, Comsewogue, with an opt-out rate of nearly 82 percent; Rocky Point, with 77.3 percent; Sayville, with 67.9 percent; and East Islip, 62.9 percent.
"We want our classrooms back, and I want my children to regain their love of learning," said GiGi Guiliano, an East Islip mother active in the movement. Two of her sons stayed home for three days this week while English tests were administered.
Guiliano, who works part time as an MRI technologist, said she was particularly angry that time spent on test prep and testing was restricting her sons' opportunities to pursue their interest in music.
Many parents who have rallied against state testing and the use of test scores in evaluating teachers' job performance express hope that the number of opt-outs will grow to the point where the state's teacher evaluation system will be dropped altogether.
Can still gather data
Statistical experts said, however, that the state still collects enough data to keep the system functioning. Under state rules, 16 student scores are considered adequate to calculate a teacher's state rating known as a "growth" score -- meaning that eight children taking both English and math tests in a single class would be enough.
State Education Department officials voiced confidence this week that the latest rounds of testing would generate scores valid for assessing both students' progress and teachers' job performance.
Ken Wagner, the state's senior deputy education commissioner, said in an interview that he expects more teachers to fall short of the required 16 scores this year than last year, when an estimated 447 teachers statewide were in that category. More than 38,000 teachers were issued growth scores in the latest round of evaluations.
Teachers in grades 3 through 8 with fewer than 16 student scores are rated according to another measure known as a Student Learning Objective, or SLO. This often consists of a composite of English and math scores for all students in a school building.
The rationale behind this schoolwide measure is that all teachers in a building -- even those involved in subjects such as art and music -- are broadly involved in improving students' academic literacy. Not all educators are convinced, however.
"It is absurd to evaluate teachers based on students that they don't even teach," said Brad Lindell, a school psychologist and union leader in the Connetquot district. "Why are they doing that?"
Tremors from what some describe as a parent "revolt" have been felt in Washington, D.C., as well.
On Thursday, a U.S. Senate committee approved a revision of the federal education law from President George W. Bush's administration, known as No Child Left Behind, on a bipartisan vote of 22-0.
One provision in the revised proposal would ban federal officials from setting rules for the way in which school districts rate the job performance of teachers and principals. New York State currently has a federally approved system that uses English and math scores to help rate teachers and principals -- a particular sore point that has prompted many of those educators to support the opt-out movement.
Bob Schaeffer, a former Port Jefferson resident and now one of the nation's leading critics of standardized testing, noted that No Child Left Behind was supposed to expire in 2007, but that Congress delayed action until this year. He credits the impact of parents' test refusals on Long Island, in New Jersey and elsewhere for the sudden turnaround.
"A major reason why 'No Child Left Behind' amendments have finally appeared -- eight years after they were due -- is that a huge grassroots movement including parents, teachers, school board members and others has put pressure on federal officials to stop test misuse and overuse," said Schaeffer, who is public education director for FairTest, a Boston-based advocacy group.
Still, the amended bill leaves in place the central components of No Child Left Behind: annual testing in English and math for all public school systems nationwide, coupled with annual reports on score results.
Even testing experts and local school leaders who worry that the increasing number of opt-outs will render test results less reliable than in the past say that they have to continue working with whatever data the system provides.
"So it is skewed," said Lorna Lewis, superintendent of Plainview-Old Bethpage schools and co-chairwoman of a curriculum committee for the New York State Council of School Superintendents. "Nevertheless, when we talk to people in [the] state [Education Department], they still say their figures are pretty robust and that the results will stand up."
David Morganstein, president of the Virginia-based American Statistical Association, the world's largest organization of statisticians, noted that when the percentage of people responding to a survey drops below 80 percent, results have to be carefully analyzed to check for statistical "bias." The same principle applies to testing, he said.
Thus, if 20 percent or more of students skip a test, results should be checked to see if those opting out included disproportionate numbers of students who were rich or poor, academically gifted or academically struggling, and so forth. Such aberrations can be adjusted statistically to validate results, Morganstein added.
"It certainly reduces the value of results," he said, "but it doesn't invalidate them."