Report of the UFT Task Force on High Stakes Testing April 2007
Apr 20, 2007 2:26 PM
Introduction
Anyone who enters the teaching profession has to have a commitment to educating students. Without that commitment the job is too demanding, the conditions too demeaning and the rewards too small. Teachers want their students to succeed. They hold themselves accountable for their students' performance. No teacher goes to work with the goal of failing a student. The biggest reward in this profession is seeing a student succeed. The question then is what kinds of tools, as well as knowledge, methods and support services will help teachers to do this. One tool that teachers need and value is reliable data about student progress and achievement.
For more than a decade efforts in New York City and New York State public schools to raise academic standards, improve the quality of education for all students and to help students succeed have been accompanied by an equally vigorous movement to develop and implement a variety of tests and assessments that ideally would not only measure student achievement but also would be useful in improving the quality of instruction on a day-to-day basis.
The Recent History of Testing
In the 1990s the New York State Education Department revamped the high school Regents exams and phased out an easier alternative for some students, the less demanding Regents Competency Tests. They introduced standardized tests in grades 4 and 8 that were intended to provide benchmarks for evaluating schools, not students. The passage in 2001 of the federal No Child Left Behind Act (NCLB) required all states to implement testing of reading and math in grades 3-8 as a school accountability provision (though not as a tool for making many high stakes decisions such as those about promotion or graduation). This requirement pushed New York State to expand its existing standardized testing program in reading and math from the 4th and 8th grades to grades 3, 5, 6 and 7. At the same time the New York State Board of Regents moved to make the standards for graduation more rigorous by requiring all students to pass a number of Regents exams as a prerequisite for graduation. The increase in the number and importance of standardized tests NCLB generated has had an enormous impact on schools throughout the state.
Here in New York City, at the same time as the state was revamping its testing program to conform with NCLB, the city school system continued to administer its own grade 3-8 testing program as required by the 1968 decentralization law and added another series of assessments in kindergarten through 2nd grade (most notably ECLAS) that exceeded the requirements of both the state and NCLB. For a time this led to duplicative testing by the state and the city of students in the 4th and 8th grades and the reporting of often conflicting student data. This duplicative testing ended in 2005 before it could expand to all students in grades 3 through 8 largely due to the public outcry of parents and teachers.
Commenting recently on the state of testing in the city a former head of the DOE’s Division of Assessment and Accountability on reviewing the testing calendar issued annually by his former office remarked that in New York City's schools someone is being tested on four out of five days every week of the year.
Testing for Accountability
Although NCLB mandated testing to measure school performance, the increased number of tests students must take and the importance given these tests as the sole determinant in such high stakes decisions (even though the tests used are not intended for these purposes) as eligibility for gifted programs, promotion of students, and entrance into select schools, has also sparked angry responses from parents, teachers and students.
The testing and accountability provisions of NCLB delineate consequences for schools and districts that fail to make adequate yearly progress (AYP) towards a uniform level of proficiency for all students. The law only allows schools to make AYP if a certain percentage of students overall, and a certain percentage of students in all the subgroups based on race, gender, poverty, special needs and English language ability in the school achieve an absolute level of proficiency. The federal law designates schools that fail to make progress as "schools in need of improvement" (SINI) and school districts must give students the right to transfer to a non SINI placement. Designation as SINI, if ongoing, can result in other sanctions and punitive measures including ultimately closure or reorganization as a charter school under private management. This federal accountability system operates parallel to but independently of New York State's existing accountability measures which uses its 4th and 8th grade reading and math tests as the basis for state designation of schools as low performing or SURR (schools under registration review).
This year the New York City Department of Education (DOE) began its own citywide accountability system, using measures that differ from those of the federal and state governments and giving schools a letter grade of A, B, C, D or F. This rating system is modeled on a system in use in Florida for more than five years and that, opponents there claim, has had a demoralizing and punitive effect on schools. (An October 2006 Zogby poll showed that 61% of Florida voters disagreed with using test scores as a basis for funding and grading schools.) New York City officials acknowledge that these different systems of evaluation can result in a school receiving, for example, a letter grade of A yet have a designation under NCLB as SINI or by the state as SURR.
The Misuse of Tests
The intuitive appeal of test scores as measures of student performance and of the tests as representative of high standards of achievement has prevented a meaningful discussion of their limitations and the negative effects of attaching high stakes consequences and sanctions to test results. It is not possible to explain all of the sometimes very technical limitations and consequences of high stakes testing in a few short sentences. Nevertheless concern and controversy around testing competing accountability systems have grown as the federal government and state and local school systems continue to misuse them to determine promotion, graduation and other high-stakes decisions for students as well as a basis for evaluation of schools and districts.
This task force sponsored a November 11, 2006 UFT conference on high stakes testing at which James Popham, former high school teacher, test designer and distinguished author, pointed out that those who advocate for the misuse of student test scores to evaluate individuals, schools, and entire school systems are ignorant of or choose to ignore the fact that the makers of these tests never intended them to be used for those purposes. The use of these tests for making these decisions is questionable at best, he said. He is not the only expert decrying this use of tests. Professional organizations such as the National Academy of Sciences, the American Psychological Association, the National Council on Measurement in Education, the National Council of Teachers of Mathematics, the National Council of Teachers of English and the National Parent Teacher Association, have all come out against high stakes testing. The American Education Research Association has stated that tests are always fallible and should never be used as high stakes instruments.
Yet wrongheaded proposals from Chancellor Klein, elected officials, corporate heads and other non-educators who do not understand the limitations of the test data continue to call for the misuse of student test scores in order to make important decisions about children as early as kindergarten. They are also proposing misusing these test results as an evaluative tool for teachers, as a factor in determining teacher salaries and as a basis for granting tenure.
The UFT Task Force
Rationale for its creation
As a result of the widespread and ongoing concerns that parents, teachers and students have expressed, and given the lack of real opportunities the public has had to question decisions the city and state have made, President Randi Weingarten announced in the late spring of 2006 the formation of a UFT Task Force on High Stakes Testing. The UFT leadership charged the task force to provide a forum for public debate and discussion and to make recommendations to help teachers and guide the UFT leadership in its ongoing discussions with city and state officials, and, in collaboration with the American Federation of Teachers, the federal government. Membership was open to all UFT members as well as officers and staff who expressed an interest in joining the task force. Vice-president Aminda Gentile chaired the task force.
A context for the work of the task force
Opponents of testing frequently remark, “Pigs don’t get heavier merely by weighing them everyday.” The proponents of testing reply, “You have to weigh the pigs to know if the feed you’re using and the way you’re feeding them is having the desired result.” This task force sought a proper balance between a rich instructional program that focuses on the whole child and an accountability system that allows us, as professionals, to evaluate whether we are accomplishing our purpose.
As a union of professionals we believe that our members help students learn and grow in many ways and we are professionally accountable for providing a rich educational environment to help the students in our classes succeed. This does not mean that our students or our performance should or can be judged by a single test score from a single day. The negative consequences of a high stakes testing process are more deleterious than any apparent gain in accountability measures. Proponents of testing are quick to portray any attack on high stakes testing as an attempt to avoid accountability. That is not our purpose.
The work of the task force
Over the past several months this task force, composed of teachers from all levels, clinicians, teacher center specialists and UFT staff, made an intensive study of the topic, familiarizing themselves with the actual assessment tools in use in the city’s schools as well as federal, state and city regulations and policies regarding testing and assessment. The task force members also studied testing and assessment policies in other states. They utilized and shared articles, opinion pieces and research from a variety of sources and perspectives to help them in their discussions. In addition, task force members designed and participated in the aforementioned November 2006 conference as well as the citywide forums described below in order to gain as wide a perspective possible about the impact of high stakes testing and to provide information about the issue to the public.
Among the questions for which the task force sought answers were:
- How has the emphasis on tests in the city's Children First agenda and the federal NCLB legislation affected what goes on, day-to day, in schools and classrooms in our city?
- What are alternatives to standardized high stakes tests in assessing student achievement?
- What are the proper uses of tests and assessments?
- How can the proper use of time, resources professional development and student data help improve teaching and learning and ensure a quality education for all our students.
To guide them, task force members shared articles, opinion pieces and research on the broader issues of testing and assessment. Their own professional experiences based on the situation in the schools in which they teach or with which they are familiar, also guided the members of the task force. The task force was fortunate that among its members, all of whom were knowledgeable and passionate were teachers from Urban Academy, a member of the New York State Performance Standards Consortium. Teachers at this school, and others in the consortium, have extensive experience in developing and using performance-based and alternate assessments such as projects and portfolios for the evaluation of student learning.
The UFT Forums on High Stakes Testing
In a series of public forums on high stakes testing this task force held throughout the city in December 2006 and January 2007, teachers, parents, students, elected officials and others with a stake in public education spoke out about the adverse affect the current high stakes testing culture is having on instruction and learning. While recognizing the importance of testing and assessment as an indicator of student progress and a valuable resource for guiding instruction, these forums showed that many members of the public are concerned about excessive testing as well as inappropriately linking the results of these tests to important decisions about a student's or a school's future. Concerns were also expressed about the pressure to raise student test scores, the narrowing of the curriculum in order to focus upon only the skills and knowledge needed to pass the tests, the demands that testing and test preparation puts on instructional and professional time, the decreased flexibility in professional decisions teachers experience as a result of the emphasis on tests, and how these high stakes tests are affecting the education of special populations of students such as English language learners and students with disabilities. .
"We're teaching students to be test takers"
One speaker, summing up the feelings of many, noted, "Students are being taught to take tests, to figure out what the test wants one to say and then to say it. They are not being taught to be critical thinkers."
A 4th grade teacher complained, "We are losing classes due to test prep. Before we dropped science and social studies to prepare for the ELA (English Language Arts) exam; now we are preparing for the math [exam]. Now there is no art."
Many teachers bristled at the lack of respect shown for their professional autonomy and decision making regarding the education and the future of the students they teach every day. Principals and administrators prefer, they said, to look only at student test score data on a single test when making decisions about that child, ignoring teacher input. "My experience is worth a lot more than 'open up to page five'" said one teacher.
In addition to the citywide reading and math tests teachers must now also administer pre-packaged interim assessments such as ECLAS-2, EL SOL and Princeton Review. These assessments often have little relationship to the actual material being taught and publishers provide little or no information regarding their technical adequacy as predictors of future achievement. Many principals also mandate the use of ongoing assessments such as portfolios, anecdotal binders, running records and reading assessment charts that are often duplicative, generate excessive paperwork and are ill suited as diagnostic tools for certain subjects and students. Although they might occasionally provide useful data, given the current focus on testing there is, paradoxically, little opportunity to translate that data into real, non-test preparatory instruction. One forum participant characterized herself and her colleagues as "The Stepford Teachers” because they are forced to follow a lock-step teaching model focused solely on test preparation.
"What good are the test results if we never get them?"
Teachers also said the results of the many tests and assessments New York City's public school students must take far too often do not come back to them in enough time to help them guide individual student learning. This includes the interim assessments whose stated purpose is to give teachers guidance on what the student needs to learn to do better on the real test. Teachers cited delays of up to eight months between test administration and release of scores. To be useful, teachers said, the turnaround time for releasing results of assessments conducted during the school year should be as brief as possible, a matter of days at most, in order to provide timely information to teachers, parents and the students themselves. It is not clear that any test publisher or outside vendor has the capacity to do that for a single school, let alone for a school system the size of New York City.
"The Bubble Student"
Teachers at the forums also discussed the phenomenon of the "bubble student," the student who scores within a few points of the next level of performance and who with extensive test preparation can move up a level and "improve" the overall performance of the school. These students are often singled out for intensive preparation for the test, in the guise of tutoring, to try to ensure that they make it to the next level. The students who need the most help, paradoxically, get less because they are less likely to make the jump from level one to proficient. In this climate where scores are everything test preparation is becoming a subject unto itself.
"I am reading test preparation booklets, not Shakespeare," said a middle school student.
Teachers at the forums said their students are no longer individuals with strengths, interests, talents and challenges but are looked at by data analysts as "a 1, 2, 3, or 4," the number corresponding to the student's level of proficiency on a reading or math test. Nor is the "bubble student" phenomenon limited to this city's schools. Schools around the country, operating under the NCLB sanctions for failure to make AYP are desperate to improve test scores. The Washington Post of March 4, 2007 reported in an article "A Concentrated Approach to Exams" that one Maryland school singled out Black, Latino and Asian students who had the best chance of improving their [and the school's scores] and offered them intensive test preparation while ignoring those who needed the most help but were seen as unlikely to improve substantially.
This emphasis on test scores and the corresponding data they yield about levels of student proficiency as the most important (and the defacto only) indicator of school quality above all other possible indicators means, unfortunately, that teachers are unable and, according to forum participants, often prevented from giving students at the lowest and highest levels of test score performance the individualized attention they should get. English language learners, gifted and talented students, transfer students, students with histories of erratic attendance, special education students and others with specific needs are too often ignored. Administrators, whose evaluation unfairly depends on the overall performance of the school, as determined by student test scores, focus their attention on test preparation rather than on a high quality educational program for all with appropriate student supports and individualized instruction.
"Most AIS (academic intervention services) instruction is devoted to pull out every day for the tests coming up," said one elementary school teacher.
Students sense this too. One parent said her middle school age son said, "If you get a 4 (the highest level of proficiency) then you don't even need to stay awake for anything else."
"What about students with specific needs?"
This excessive testing is harmful to specific groups of students. Teachers and parents talked about the NCLB requirement that special education students must take tests at their grade level, a requirement that prevents them from demonstrating true progress, reinforces feelings of inadequacy and frustration and alienates them from school, increasing the likelihood that they will ultimately drop out. The law permits only a small percentage of the most profoundly disabled to be exempt from the standardized testing program, though states must develop alternative measures to evaluate their progress. The law requires students who are new to this country and whose native language is not English to take grade level tests in English after only one year, despite research that shows it can take five to seven years for English language learners to acquire the English language skills necessary to perform on a level with their peers. This requirement generates the same feelings of alienation and frustration among English language learners as it does in other students with special needs.
"If it's not tested it doesn't matter"
Teachers and parents in all five boroughs also spoke about the elimination of and cutbacks in art, music, social studies, science, physical education and foreign language programs, as well as field trips and after-school recreational programs in order to provide more time for test preparation for the high stakes math and English language arts tests. A member of a Community Education Council noted that only 40% of the elementary schools in his borough now offer physical education. The city's announcement on March 6, 2007 that it is implementing a citywide science curriculum is a welcome acknowledgement that a well-rounded curriculum has to include more than test prep for math and English but other subjects remain eliminated, ignored or curtailed. But the announcement does nothing to prevent principals from canceling science in order to find time to maximize performance in English and math by maximizing test preparation.
Consider the comments of a middle school teacher from Brooklyn who described the emphasis on test preparation in her school. “Subjects like social studies are put on the back burner so that everyone can focus on the tested subjects of ELA and math. In our school a science lab was disbanded to make more room for ELA and math classes even though the school has an accelerated science program where 8th graders can take the biology regents but now they don’t even have a lab to do dissections.” In a test driven climate only subjects for which there are tests are valued.
"I'm concerned about the health of my child"
Parents spoke with understandable emotion about the adverse affect excessive testing has had on the physical and emotional health of their children. Students as young as eight and nine have characterized themselves as "failures" due their score on a single math or English language arts standardized exam. According to parents their children often experienced sleepless nights, vomiting, severe headaches and anxiety attacks during the days leading up to the fateful high stakes test. A school psychologist said she teaches students how to “de-stress the test". She described how children tell her they become so anxious before the tests that they have stomach aches and can’t sleep, and she pointed out that test anxiety causes students to forget what they know and affects how well they listen to directions and to stories read aloud to them. A PTA president said, "Our kids are getting lost in these tests."
Once again, the issue is national. A January 12, 2007 article in USA Today "How Bush Education Law has changed our schools" quotes Carmen Melendez, a bilingual language arts teacher, who has since left teaching:
"It was insane. The kids were all jaded. They were tired, they hated school. They're 8 years old, and they're so worried about a passing score. I think that's inhumane."
David Keyes, a second grade teacher in Maryland writes in "Crying Over A Test" in the March 2007 NEA Today:
"The test prep program completely changed the classroom culture. In writing I might expect a well-edited paragraph from one student, two simple sentences from another. The test prep program sends a very different message: each question has one correct answer, which all students must find. Recently two struggling students who had failed to get a single answer right all week broke down in tears."
"Is it higher student achievement or easier tests?"
Finally, the law of unintended consequences was evoked by classroom veterans as they spoke of the "dumbing down" of tests and the intentional design of tests in order to yield higher scores, not surprising now that high test results are so explicitly and inextricably linked to the careers of principals, superintendents, a chancellor and even a mayor. "How do you get students to run five miles faster?" asked a former principal. "Make the five miles three."
"It's not a new problem"
The criticism of high stakes testing is not new. In 1906 a New York State Education Department official made these comments as the legislature contemplated establishing a high stakes test:
“It is an evil for a well taught and well-trained student to fail in an examination. It is an evil for an unqualified student, through some inefficiency of the test, to obtain credit in an examination. It is a great and more serious evil, by too frequent and too numerous examinations, so to magnify their importance that students come to regard them not as a means in education but as the final purpose, the ultimate goal. It is a very great and more serious evil to sacrifice systematic instruction and a comprehensive view of the subject for the scrappy and unrelated knowledge gained by students who are persistently drilled in the mere answering of questions issued by the Education Department or other governing bodies."*
In ignoring this prescient warning, we are reaping exactly the consequences that the education department official foresaw. In a June 18, 2000 op-ed piece in the New York Times "When Testing Upstages Teaching," Kate Zernike quotes Sue Bastien, the head of Teaching Matters, a group of former teachers devoted to teacher training, as saying:
"The standards which should have guided us have bowdlerized into standardized tests which are then used to punish low-scoring schools rather than improve the curricula or teachers' skills."
The problem is not new, but obviously forum participants and educators from around the country have said the problem is that it has gotten worse.
* (from Sharon L. Nichols and David Berliner, (2007) Collateral Damage: How High Stakes Testing Corrupts America’s Schools, Harvard Education Press, Cambridge, MA, p. 5)
