Does Education Secretary Cardona Recognize the Two Huge Problems with High-Stakes Testing?

Education Secretary Miguel Cardona insists that federally mandated standardized testing will go on as usual in this COVID-19 dominated year. While his decision feels particularly impractical, intrusive, complicated and disruptive in the midst of COVID-19, the decision is of much deeper concern for two reasons.

This blog will take the holiday weekend off. Look for a new post on Wednesday, April 7.

One would like to think that Dr. Cardona is familiar with the huge debate that has consumed education experts and also many parents who have been opting out for years now.  But when Dr. Cardona explained why testing must go on as usual, he didn’t even bother to offer a rationale that addresses any of the reasons experts have insisted he should cancel the tests once again this year. Instead he said we need the tests so that the Department of Education can ensure that federal investment goes to the school districts that need it most. That is such a lovely thought, and if tests were designed and used to gauge needed investment in the poorest communities, it would be wonderful. 

But standardized tests, as mandated by No Child Left Behind and its successor, the Every Student Succeeds Act, were not designed to drive a system of test-and-invest. They are instead the very foundation of a maze of policies at the federal level—and now federally mandated across the states—to identify so-called “failing schools” and to punish them.

The first kind of damage caused by high-stakes testing is pedagogical. Standardized testing and its preparation have deeply affected what happens in the classroom itself. Dr. Cardona’s decision to insist on tests in this schoolyear will undermine what students need most when school resumes in some sort of post pandemic normal.

At his Rethinking Learning blog, Rich tenEyck explains: “For more than 20 years now, we have been told that a major component of the ‘standards movement’ was the creation and use of large-scale assessments required by federal funding programs. These were sold as a critical source of information about how much our kids are learning… These annual tests are far more reliable predictors of family wealth than as tools for helping teachers better respond to student needs. Educators have known this and have frequently tried to alert us to the misunderstanding and the misuse of these tests. What has happened as a result?  These teachers and school leaders have been vilified… But what if the tests required by various pieces of federal legislation never really tested learning at all? What if they tested the recall of many isolated and disconnected facts?… What if the tests provide almost no insight into the real learning needs of kids?”

Educator and blogger Steve Nelson diagnoses the special problem with standardized tests this spring when some students are online, others in hybrid settings, some disconnected: “In keeping with the illogical, inhumane, and ineffectual practices of the recent past, the testing industry will look for all the deficits it can find, so as to identify the mythical ‘learning losses,’ so that the least privileged can be remediated using materials produced by the testing industry, thereby further depriving them of the experiences they need most… Renowned cognitive psychologist Jerome Bruner and many others have determined that social context is the most important variable in real learning. Relationships among and between teachers and students determine the quality of school experience. Now more than ever kids need to be back in the good company of their friends and their teachers.”

And Alfie Kohn reminds us: “John Dewey described how a curriculum that’s based on students’ questions and connects with their experiences has ‘an inherent attracting power.’” Kohn continues: “The whole standards-and-testing edifice of our education system consists of expectations and outcomes that have been devised by distant authorities, imposed on students (and teachers!), and enforced by exams to ensure ‘accountability.’ These standards are often breathtakingly granular in their specificity because the whole approach is rooted in an outdated behaviorist model of learning.”

Standards and test-based accountability have moved us far away from the progressive philosophy of education advocated by Kohn, Nelson, and tenEyck. But there is also a second problem is that is structural and systemic: Standardized testing has damaged the very foundation of our entire system of public education. Ohio’s Bill Phillis captured the extent of the problem in his daily comment on Tuesday: “The No Child Left Behind Act has put the nation at risk… After four decades of reform by politicians, teachers are demoralized. Poor school districts are still poor with test scores lower than rich districts. Billions have been largely wasted on charters and vouchers. The voucher and charter advocates have developed powerful lobbies and billionaire partners. The future of the time-honored common school system is in jeopardy.”

Today states are required by the Every Student Succeeds Act to identify the bottom performing schools according to their standardized test scores and to submit to the U.S. Department of Education a plan to turnaround these schools. This system attaches high stakes to the standardized test scores as a tool to blame and punish educators and make them work harder. The punishments it imposes are severe.

  • Many states create and publish school and school district report cards which rate and rank schools and school districts.  
  • Some states take over so-called failing schools and school districts and impose state appointed overseers and academic distress commissions to manage low scoring schools and school districts.
  • Other states, or sometimes the administrators of school districts, shut down low scoring schools and, ironically, call the shutdowns a turnaround strategy.
  • States use test scores to hold children back in third grade if their reading scores are too low.
  • Many states deny students who have passed all of their high school classes a diploma when they don’t score proficient on the state’s graduation test.
  • Even though statisticians have shown that students’ test scores are not valid as a tool for evaluating teachers, and even though the federal government has ceased demanding that states use test scores for teachers’ evaluations, a number of states continue this policy.
  • School districts with F grades are the places where many states permit the location of charter schools or where students qualify for private school tuition vouchers—sometimes with dollars taken right out of the school district’s budget.
  • Because test scores tend to correlate closely with a community’s aggregate family income, the federal high-stakes standardized testing regime brands the schools in the poorest communities as “failing schools.”
  • The branding of poor school districts causes educational redlining and middle class flight.

A lot of people are watching Education Secretary Miguel Cardona carefully to gauge whether he grasps the depth of the problems with high stakes testing, first, pedagogically within our nation’s classrooms, and second, through the test-based system itself that punishes instead of assisting the schools that need the most help.

Diane Ravitch summarizes why people are so concerned that Secretary Cardona has not acknowledged the damage of the high-states testing regime: “These tests have high stakes for students (who might fail to be promoted), teachers (who might be fired if their students’ test scores don’t rise), and schools (which might be closed if test scores don’t go up)… The challenge for Miguel Cardona, Biden’s Secretary of Education, will be to abandon two decades of high-stakes testing and accountability and to remove any federal incentives to create privately managed charter schools.”

Should the Federal Government Be Determining How States Evaluate Teachers?

Senator Lamar Alexander, chair of the U.S. Senate Health, Education, Labor and Pensions Committee, held another hearing this week on the potential reauthorization of the federal Elementary and Secondary Education Act, since 2002 called No Child Left Behind (NCLB). The subject of this week’s hearing was federal requirements for evaluating school teachers. While it is early yet to predict any sort of outcome for the NCLB deliberations, Lauren Camera of Education Week speculates: “Although members of the Senate education committee agreed at a hearing Tuesday that teacher evaluations are essential for a thriving public education system, it’s unlikely that the forthcoming reauthorization of the No Child Left Behind Act will include specific requirements.”

Removing requirements for tying evaluation of teachers to students’ test scores would be a radical shift in federal policy. The Obama administration conditioned qualification for its competitive grant program Race to the Top on states’ basing evaluation of school teachers on their students’ standardized test scores.  And the Obama Department of Education’s waivers from the onerous punishments of NCLB have also been contingent upon states agreeing to connect teachers’ ratings to their students’ standardized test scores.

Describing Tuesday’s hearing of the Senate HELP Committee, Camera continues: “Republicans, including Chairman Lamar Alexander, R-Tenn., said Washington shouldn’t mandate such policies, while Democrats, including ranking member Patty Murray, D-Wash., were wary of increasing the role student test scores play in evaluations and how those evaluations are used to compensate teachers.  The lack of language in the reauthorization requiring teacher evaluations will likely stop in its tracks the Obama administration’s efforts to push states to adopt evaluation systems based in part on student test scores and performance-based compensation systems, both of which were at the heart of U.S. Department of Education’s NCLB waivers.”  Camera reports on testimony presented to the Senate HELP Committee by Terry Holliday, Kentucky’s education commissioner, that evaluation of teachers should be collaborative and “not done to teachers and leaders.”

Coincidentally, the day after the Senate HELP’s hearing on evaluation of teachers, I attended a packed meeting here at home where a panel of teachers from my own school district’s elementary and middle schools and our high school spoke about how their teaching practice has been affected by standardized testing and the evaluation of teachers based on their students’ scores.  All of them were able to examine these relatively new experiences in the context of long careers that stretch before the passage of No Child Left Behind.

Natalie Wester was chosen as Ohio’s teacher of the year in 2010, but she told the crowd that she worried even as she received the award, because that year only 4 of her third grade students had passed the autumn practice exam leading up to the official state test. In the spring only 50 percent of her students achieved the “proficient” rating. What the state’s examination did not recognize and what no official rating will ever show is that every student in her class that year grew two or three performance levels. The test, like all the standardized assessments since the passage of NCLB, recognizes achievement only when children cross the passing benchmark. If a non-reader enters a third grade classroom in the fall, learns to read, and becomes a second-grade-level reader in that one year, the child still counts as a failure according to the assessment that credits success only when a child reads at grade level. Wester declared, “I fear that in a very real sense we are squashing dreams, confidence, and children’s belief in themselves through testing.”

Another teacher reported he is working this year with a small group of third graders whose reading test scores are so low the students are likely to fail the state mandated Third Grade Reading Guarantee test. Students who fail will be required to repeat third grade. This teacher says he watches his students “shut down” when they realize how far behind they are. “I see that spark of wanting to learn dying in my students. I feel we are abusing our students.”

A high school teacher of special education worried that some of her students are so far below the basic level at which the standardized test is constructed that the testing experience itself is emotionally defeating.  All of the teachers who spoke affirm the value of informal quizzes and check-ins with students—formative assessments—that provide the teachers with feedback to plan interventions, support students, readjust the lesson, and add extra challenge as the lesson is expanded.  Very often, according to all the speakers, standardized test scores come back a semester or a year after the test, long after a particular teacher can use the data to address challenges faced by the students who are no longer enrolled in their classes.

A teacher from a neighboring school district framed the evening by explaining the details of the Ohio Teacher Evaluation System, created as a requirement when Ohio applied for the federal Race to the Top competition and included as a requirement for the state to receive a federal waiver from some of the worst problems in NCLB. This system evaluates teachers in large part based on their students’ standardized test scores.  In the context of listening to a panel of professionally expert teachers speaking to their long experience working with children, it was almost baffling to try to follow the details of the plan by which Ohio’s teachers are rated “accomplished, skilled, developing or ineffective.” Teachers are spending hours filing reams of data about their teaching and their students. These reports along with formal observations of their classes count for 50 percent of their evaluation with another 50 percent from their students’ standardized test scores. A new revision of the Ohio Department of Education’s evaluation rubric will allow a school district to create alternative components for 15 percent of the overall rating and then award 42.5 percent on reports and observations and another 42.5 percent for students’ test scores.

As I listened to  the description of the burdensome evaluation system set up by the Ohio Department of Education, I know I was not the only person thinking about Natalie Wester’s students.  Each one of them gained at least two or three performance levels in her class, but only 50 percent of her children passed the state’s proficiency benchmark that year. Even if they have made substantial academic progress, children’s failures to reach a particular cut score affect not only them and their confidence and will to persist, but also shape the formal state evaluation scores of their teachers—even for Ohio’s teacher of the year.

Controversy over Federal NCLB Waivers Creates Opening for Pressure Against Test-and-Punish

The shifting of public opinion sometimes happens while we aren’t paying attention, and then it becomes clear that something has changed.  Arne Duncan, the U.S. Secretary of Education, recently bowed to growing pressure against his policies when he told states they can delay for another year the requirement that they evaluate teachers based on students’ test results.

A couple of weeks ago Motoko Rich in the NY Times presented the history of the use of students’ scores for evaluating teachers:  “Over the past four years, close to 40 states have adopted laws that tie teacher evaluations in part to the performance of their students on standardized tests… These laws were adopted in response to conditions set by the Education Department in the waivers it granted from the No Child Left Behind law, which governs what states must do to receive federal education dollars.  The test-based teacher evaluations were also included as conditions of Race to the Top grants that have been given to states by the Obama administration.”

But recently controversy in several states about whether the U.S. Department of Education will renew or revoke the waivers from ill-conceived mechanisms of No Child Left Behind (NCLB)  have drawn media attention to NCLB’s misguided policies, along with problems in what the waivers required states to do, and  inconsistencies in the way Arne Duncan’s Department of Education is managing the waiver renewal process.

First came the extraordinary early August letter sent to all parents in Vermont by the state’s Secretary of Education, Rebecca Holcombe.  Vermont is one of a handful of states that never applied for a waiver.  That means NCLB is still operating in Vermont, and Holcombe sent the letter required by the law to all parents whose children attend schools considered “failing” by NCLB’s Adequate Yearly Progress mechanism.  The problem is that, because NCLB required states to raise ‘cut scores’ for student proficiency higher every year at the same time NCLB mandated that all schools make all their students be proficient by 2014, virtually all schools across America (and all schools in Vermont) are now “failing schools” according to the way the law evaluates schools.  In her letter to all of Vermont’s parents, Holcombe explained why her state has never sought a waiver, and then explained very clearly why NCLB’s ” failure label” is meaningless.  She also explained how the whole test-and-punish regime of NCLB has been a fiasco, how the rules of NCLB and the waivers offered by Arne Duncan are all messed up, and why Vermont simply refuses to play the game. Her letter was a refreshing development!  (This blog covered Holcombe’s Vermont letter here.)

No Child Left Behind and the NCLB waiver have also been in the news in Washington state, where the U.S. Department of Education revoked the state’s waiver because the state sought to let school districts choose the exam which would be used to evaluate the state’s teachers, when the waiver requirement is that the state test required by NCLB will be used for the evaluation of teachers. Reporters for the local Kitsap Sun on August 28th found themselves trying to explain—now that NCLB is operating again in Washington due to the loss of the waiver—why 88 percent of Washington’s public schools failed to make Adequate Yearly Progress under the newly reinstated standards of NCLB by which all students in 2014 are to reach proficiency or their schools be awarded a “failing” label.  The reporters explain that the only schools in Washington state not deemed “failing” accomplished that feat under a little known “safe harbor” rule, “which credits schools for gaining significant ground in student achievement since 2011, the last time Washington had to calculate progress.  Schools that reduce the percentage of students not meeting standard by 27 percent are deemed to be making adequate yearly progress, even if they didn’t hit the 100 percent target.”  Only 260 public schools across Washington state met the “safe harbor” Adequate Yearly Progress standard.  Like Vermont’s state superintendent, educational leaders across Kitsap County’s school districts are quoted listing the honors accomplished by many of the so-called “failing” public schools.  It is refreshing to read discussion about unworkable federal education policy in a local newspaper.

Then Oklahoma  lost its waiver on September 2, after the school year had already begun.  This time the U.S. Department of Education revoked the waiver after the state’s legislature passed a law repealing the state’s adoption of the Common Core standards.  Arne Duncan’s Department of Education had made a state’s adoption of  “college and career-ready” standards a condition for receipt of a NCLB waiver, and federal regulators deemed the standards Oklahoma presented as an alternative to the Common Core—the state’s old Oklahoma Priority Academic Student Skills Standards—inadequate because too many Oklahoma college freshmen have needed remediation.  Caitlin Emma, writing for Politico, explains that other states may be in line to lose waivers:  “With Oklahoma and Washington out of the picture, 41 states and D.C. have waivers from No Child Left Behind.  Thirty-fve of those waivers expired this summer and 22 states have received one-year extensions of their waivers so far.”

The Tampa Bay Times recently editorialized about the U.S. Department of Education’s threat to cancel Florida’s NCLB waiver: “Duncan’s staff has put Florida on notice that the state is at risk of violating NCLB standards that require all children to be counted equally in accountability formulas.  Earlier this year, with the support of educators and advocates, the Legislature agreed to give non-English-speaking students two years in a U.S. school before including their standardized test scores in school grading formulas.  The change was an acknowledgement of the huge learning curve such children face and that schools should not be penalized if those students can’t read, comprehend and write English at grade level within a year.  Yet to the federal bureaucrats enforcing the unpopular NCLB law, such common sense doesn’t matter… The last thing federal enforcers should be doing is punishing a state for embracing a commonsense reform.  Education Secretary Duncan needs to find a better solution.”

It is a very good thing to see the press exploring and exposing the local problems arising from of our federal testing law No Child Left Behind and from the state laws legislatures had to pass to enable their states to meet Arne Duncan’s conditions for receiving waivers.  As more reporters cover these problems, perhaps like the editors of Tampa’s paper, more people will begin to cry out for commonsense reforms.

At the same time, there is still much confusion among reporters and the public about whose policies are driving what is becoming recognized as ill-conceived public education policy.  Across the states debates are becoming heated about plans  to evaluate teachers—based on students’ test scores.  (You’ll remember that states had to pass legislation to this effect just to qualify for a waiver.)  Despite that Arne Duncan has delayed for a year the requirement that states with waivers start such evaluations, discussion of the evaluations—at the same time states are launching the Common Core tests—is causing confusion, anger, and controversy.  Many people (citizens as well as reporters) do not realize such test-based-evaluation of teachers is a federal waiver requirement.  A recent article in the Athens (Georgia) Banner-Herald reflects the confusion. “Georgia’s new teacher assessment system is getting bad grades from Clarke County school administrators and school board members—really bad grades,” reported Lee Shearer late last week.  “The new system is expensive–the school district has spent $10,000 in printing costs alone for newly designed pre-tests…. Another problem is the system’s heavy reliance on test scores.  Under a 2013 state law sponsored by a bipartisan group of legislators, ‘student growth’ as measured by standardized tests accounts for half a teacher’s grade—even though the tests won’t count at all for individual students this year.”

That the Athens Banner-Herald is publicizing problems with excessive standardized testing is an important development.  It is up to those of us who have been tracking federal policy, however, to continue to connect the dots to ensure that political pressure is not merely on state legislators, when state legislators passed the laws merely as a requirement to qualify for a federal NCLB waiver.  Arne Duncan and the U.S. Department of Education are vulnerable right now due to all sorts of problems with the waivers. While controversy about the waivers continues to grow, we must ensure that reporters and citizens understand that federal waiver requirements are the reason for a number of problems in their states and local school districts.  Political pressure needs to press the U.S. Department of Education to turn away from test-and-punish accountability.