Fighting the White Knight: Saving Education from Misguided Testing, Inappropriate Standards, and Other Good Intentions

By Jocelyn Turner

In her final year of teaching, Jocelyn Turner spent over one-fifth of each week administering mandatory tests and quizzes. She spent the remaining time preparing students to take those specific exams, regardless of the background knowledge and preparedness of individual students. While she was testing, she could not teach. Teachers were expected to present the same Common Core-based, grade-specific material to all their students at the same time-- whether Jake was reading at a first-grade level or Taylor at a ninth-grade level. It was a rare and lucky child who fit the profile of the year's onslaught of tests.

Since No Child Left Behind, US schools have been burying students in tests and then drawing often misguided conclusions--when sometimes the only conclusion anyone ought to draw is that student X obviously spent hours staring at a set of questions he or she did not understand and maybe could not even read.

We have been told that US education is in crisis. Ms. Turner agrees. In Fighting the White Knight, she argues that government mandates created and are now perpetuating this crisis, depriving children of remedial learning, instruction time, and personal attention.

Fighting the White Knight also looks at the $1.6 trillion student debt crisis, a consequence of today's single-minded, college-bound pipeline; vocabulary deficits left to fester due to narrowly targeted curricula; and the sneaky gutting of elective, vocational/technical, and gifted education.

Ms. Turner concludes by advocating for changes she believes can rescue American education--guiding children back to the safe, inspirational learning experiences of a more student-focused time.

• Language: English
• Publisher: Covenant Books, Inc.
• Release date: Jan 12, 2023
• ISBN: 9781638859215

Book preview

Chapter 1

Data Collection: A Numbers Game

If enough data is collected, a board of inquiry can prove anything. —Marion F. Sturkey, Murphy’s Laws of Combat

I have written bits and pieces of this book over the years, trying to identify what really matters, what matters a little, and what can be discarded as we try to level the educational playing field and close the achievement gap. I don’t exactly keep starting over, but I do keep editing as my opinions change. For example, I once believed money figured hugely in America’s educational problems. I now believe that, although a fairer funding system would help many schools, money itself will not fix this mess—at least, not if we spend that money implementing more of the same tired and tiring strategies that keep failing year after year while simultaneously demanding more money to implement new brainstorms by educational reformers who have hardly ever walked the halls of a real high school.

I want to start this book with data-driven instruction. Data-driven instruction is an educational strategy that demands teachers improve instruction by systematically using student data from tests, quizzes, daily work, and other sources to plan individual instruction. Data-driven decision-making sounds eminently sensible. The fact that instruction has always been data-driven—planned based on students’ previous grades and behavior—tends to get forgotten as formal demands for data increase.

The US push for data-driven instruction requires much more reflection than it has received. We have become enamored of numbers, numbers, and more numbers. Nearly twenty years after the failed NCLB began, schools are still gathering numbers to define educational problems and many schools are producing numbers that show a stunning lack of progress at solving those problems. Mostly missing from our discussions, however, have been the actual academic and human costs from gathering and processing the avalanche of figures we call data.

Data-driven instruction demands to be fed, and teachers and administrators just keep shoveling more numbers into spreadsheets. Students and educators today are drowning in fuzzy numbers—while students try to divibe their peppiron toppings by three for no good reason and teachers sit in hot, dusty math classrooms at the end of the day trying to explain the big red spreadsheet blotch that represents the complete lack of understanding of seventh-grade Common Core probability reflected in those massacred pizzas.

Research by the Council of the Schools, which represents 66 large urban districts across America, found that students take an average of 112 mandated standardized tests between pre-K and twelfth grade. That number doesn’t include the many pretest practices given to help students prepare for the real tests or the many hours spent learning to take the test before it arrives. That number doesn’t include the many teacher- and school-designed tests and quizzes used to assess daily instruction. The Council found one district where eleventh-grade testing reached twenty-seven days, or 15 percent of the school year, not counting Advanced Placement, college entrance or exams for career and technical education courses (Student Testing, 9–10).

Several powerful forces impede the sporadic efforts currently being made to rein in assessments. The quest for increased data-driven instruction encourages more testing. Extra data can prove useful, especially if one ignores data-gathering costs. In academically struggling districts, being able to produce improving numbers may greatly affect administrative and staff evaluations and even retention. When a superintendent, principal, or teacher does not deliver improving test scores, their job is sometimes forfeited to the next person offering yet another scheme to deliver better numbers. With the hanging sword always poised to fall, testing cannot be deemphasized.

Efforts to curb testing may also be stymied by the necessity to produce data for outside authorities and stakeholders. Multiple tests provide multiple data sources to document progress. If Measures of Academic Progress (MAP) results show only mediocre progress, AIMSweb scores may provide a greater leap forward. Both assessments are benchmark tests. In education, benchmark tests are used to find out roughly where students stand as they learn specified academic standards. These tests are typically given multiple times during the year to track student academic growth. Other benchmark tests are also available as part of this lucrative market.

Why use two tests? Scores on one test may have risen 6 percent in the last year while scores on another rose only 2 percent. That 6 percent improvement will likely make its way into the district newsletter while the 2 percent quietly disappears into a seldom-used file cabinet in a dusty interior office or a Google file never destined to be shared or opened. If test 1 doesn’t deliver the desired scores, test 2 may do the trick.

During my last teaching year, testing time loss amounted to around one-fifth of my math students’ time. All those tests represented missed classes—days when students were filling in bubbles rather than learning new material. We gave both AIMSweb and MAP multiple times. This meant that my school administered two different three-day benchmarks and two PARCC exams in one year, on top of the frequently incomprehensible mandatory unit tests and weekly Friday quizzes based on those unit tests. All these assessments were mandated so that Illinois education officials, district leaders, school administrators, and teachers could provide feedback on student progress toward meeting Common Core academic expectations.

My principal opted to give both tests because she wanted data from the better of the two for the Illinois State Board of Education representatives who regularly dropped into our school. I expect she also wanted to learn the size of the learning gap (or chasm) that she had taken on—not that my data was ever going to help that principal. Translations might have benefitted some bilingual students, but translations were originally prohibited because the tests would then no longer have been identical, which was unacceptable for data-comparison purposes. Desperate for better numbers, thought was later given to allowing translations, but I don’t think the effect would have been of much benefit. Seventh-grade mathematics can be expected to clobber students operating at a third-grade mathematics level, no matter what language we use to write the story problems. We could write those problems in Klingon, and any effect on scores would probably be negligible.

We lost ridiculous amounts of vital instructional time, minutes and hours that might instead have been dedicated to learning and reinforcing mathematical concepts these students had somehow missed or forgotten from earlier years. I like to say we were not on the radar by that point. We’d been shot down. The local school board had been fired before state oversight expanded to even include classroom visits.

If my students’ extremely rough year created a few future dropouts or give-ups, will anyone ever know? I may find out via Facebook, but my former students will be long done by then. I can’t speak for the future, but I can say that required testing wasted excruciating amounts of that academic year. The time bleed continues today. I started one early November morning in 2019 by reading an online post from an elementary teacher. She showed a picture of her many data binders and informed us that she had spent more time meeting data requirements than teaching in the fall. Given that she lost the spring to COVID-19, I fear she and her students pretty much lost the year.

Ironically, testing often hits America’s lowest-scoring kids the hardest. Academically successful districts don’t have to give extra tests to show higher scores because they have already generated numbers that make them safe from scrutiny and criticism. Those districts don’t have to lose nearly as much time preparing data for overseers. Their administrators don’t need to travel from classroom to classroom asking students, "Please, please take the upcoming state tests seriously." Those administrators can afford to relax as teachers teach division and fractions to students who are not yet ready for grade-level probability problems.

Administrators in academically stronger schools are more able to focus on kids instead of numbers. They can work on school spirit rather than test scores. They can bring in authors to discuss the process of writing nonfiction novels without worrying about lost instructional time before a test. They can provide in-school pep assemblies to kick off sports seasons, again oblivious to the fifty minutes of possible math or English instruction that just vanished. They can visit after-school clubs to watch dances and play practices.

One sad truth lost in the flurry of rapid-fire test preparation: school dances and pep assemblies have kept more would-be dropouts and tune-outs in school than extra test practice ever will. And that extra test practice regularly ends up supplanting pep assemblies in areas where pep assemblies are needed most.

US test score discrepancies must be addressed. Those score discrepancies are the achievement gap. But we ignore the costs of testing at our students’ peril. Obviously, teachers must gather data to determine if instruction is working. Gradebooks, report cards, and the traditional spring state test are all about data. But too often nowadays, districts gather copious quantities of information above and beyond those report card grades without questioning the direct costs and lost learning opportunities created by testing itself.

How many hours of instructional and lesson preparation time are being sacrificed to prepare unnecessary, unhelpful, or even inaccurate spreadsheets that show the same student growth we could infer from already-mandatory report cards or from long-term records stored in student cumulative (cum) folders? Schools keep a cum folder for every student, a file that begins when the child first enters school and contains grades, attendance, standardized test results, teacher observations, disciplinary actions, and other information for each academic year. This folder follows the student from grade to grade, school to school, and district to district.

As I write this, I want to emphasize that I am in no way against tests. Teachers must measure progress to plan instruction and to make certain instruction is working. But US educational reformers seem to have lost sight of the side effects of the current arduous testing regime—most crucially, the effects of those tests on students. Our students are not a crop of mindless vegetables that we can water to make our garden grow. Motivated, unmotivated, ambitious, lazy, happy, depressed, anxious, energetic, lethargic, ADHD, dyslexic, autistic, clinically ill—they are all individuals. They bring their personalities, families, socioeconomic status, and histories along with them. By middle school, if students decide not to cooperate with educational programs, schools may lack the resources to pull them back on track, especially in financially disadvantaged and urban districts.

When schools become numbers-focused, rather than student-focused, that’s a mistake. Students’ emotions and interests should not get shoved onto the back burner while districts frantically post numbers on spreadsheets, especially in academically failing areas. I could put research results here to support what I just wrote, but I’d rather ask readers to transport themselves back in time: Did you ever take a required class where somehow almost nothing you did ever made the professor happy? How did you feel on test days? How did you feel on regular days, as you walked to that class?

Here is what I observed while teaching my district’s new Common Core-based curriculum: students at the bottom of the learning curve became more and more confused because the hours required to teach those Common Core standards left little or no time to address lost information from the past, the unlearned lessons from previous years. In effect, the Core was harming my students rather than helping them. Making existing content more demanding without offering adequate time and resources to teach missed material from earlier years means that children who previously understood only a fraction of what was coming at them daily in the classroom now understand even less.

Sign on an elementary school wall, waiting to post individual student scores on the MAP assessment.

That increased confusion is only the tip of the iceberg however. Confusion does not exist in a vacuum. More reformers should be asking themselves what it means to be hugging the bottom of that Second Grade MAP Scores chart on the cinderblock wall above. How does it feel? How do other kids react? I guarantee similar charts have resulted in bullying and putdowns. Educators try to keep scores anonymous but don’t always succeed. What conclusions are students drawing as they view the chart?

Before NCLB, few districts taught specifically ONLY to their state’s annual standardized test. The phenomenon of scripting a whole year’s curriculum to prepare for one huge governmental whammy in the spring began as an offshoot of the severe NCLB penalties for schools that failed to improve test scores. In districts with a history of high scores, that annual test has become more important but proportionally much less so than in poor and academically struggling districts where test preparation now regularly gobbles up field trips, extracurricular activities, and even science experiments that stray from the curriculum, as schools focus desperately on gaining points. In these schools, testing becomes an end in itself, with data secondarily factored into the equation. The tests themselves become the year’s goals, with unrelated learning dropped or deemphasized (Walker).

What happens if the mismatch between our students and our tests is too great? I know a few answers to that question from experience. I had confused kids, anxious kids, resentful kids, and depressed kids. I kept propping them up, and then another unit test would knock them down again. I kept reassuring them that retakes would rescue them. At first, they looked at me like I was some kind of lunatic, but over time they understood that I was not being given a choice. At that point, I had compassionate kids. They felt sorry for me. I felt sorry for the state of US education while I filled out spreadsheets and went to meetings. If I had not been nearing a reasonable retirement point, my best move would have been to exit the world of underperforming students for an academically stronger district.

Even in the days of NCLB, higher-scoring districts could be more relaxed, knowing the worst government penalties for test-score failures would not fall upon them. Taking a school day to go to a museum didn’t threaten teachers and administrators in those schools. Test adaptations in high-scoring districts might only address the problems of subgroups such as special education students that struggled to make Annual Yearly Progress (AYP).

Individual student achievement levels are often ignored in discussions about teaching to tests, yet personal experience has shown me those levels make all the difference in how effectively teaching to the test works. If I teach the content of an eighth-grade test to two children, one operating at an eighth-grade level and the other at a third-grade level, I will have prepared the first student for the test, but I probably didn’t help the second student much. At worst, I confused that child. Tutoring my third-grade-level student should be all about remediation, even if the remediation is set years below grade level. My lower-performing student has fallen too far behind to prepare in a few months for a test set five years above his or her academic operating level, even with lengthy before-and after-school tutoring. If children could learn five years of missing knowledge in a few months, we wouldn’t be in this mess. The achievement gap would probably have been closed by now.

The net effect of teaching to the test is that most students near, at, or above grade level will probably gain significant points in the test-score game, while those performing significantly below grade level will conversely sacrifice useful instructional time for a utopian goal that they cannot reach. In the first case, teaching to the test worked and provided learning. In the second, inappropriate instruction prevented learning from taking place.

A Quick Review of Some Problems with Teaching to the Test

The US educational system’s unrelenting emphasis on boosting student self-esteem not only arises from the best of intentions but is strongly supported by research. Children with low self-esteem feel hopeless, anxious, inadequate, and isolated. They are reluctant to express their ideas and lack both motivation and resilience (25 Low). Even without reading the large body of research on this topic, teachers understand the importance of self-worth to success. What happens to self-esteem when no win is possible for a student who cannot read the test in front of him? In my experience, self-esteem takes a dive until a student sometimes detaches and chooses not to participate in this ongoing, seemingly unwinnable, game.

That plan to boost academic performance through harder tests? I am reminded at this point of mongooses and one of the classic stories of bio-control wreaking environmental havoc. In the late 1800s, the Jamaican mongoose was introduced into Hawaii to control rats, a plan that completely failed to consider that rats are active at night and mongooses during the day. The mongooses ended up eating eggs instead of rats, which had almost no impact on the total rat population but resulted in the unplanned extinction of some native bird species.

A bad plan can often be worse than no plan at all.

Fairness becomes a nightmarish issue when teacher evaluations and even job retention are based on student test-score numbers. All classrooms are not equal. Like poker hands, classrooms change with every shuffle of the cards. Students are placed in classrooms and teachers get what they get. Lucky teachers find themselves with a hand of academically strong kids. Teachers can identify student academic performance levels by examining previous test scores. Or they can skip the score review, which can be prejudicial, and look out into their classrooms. I always know within a few weeks what I have been dealt from reviewing daily work, grading quizzes, and listening to students’ answers to questions. Long before data-gathering ran amok, I listened as a second-grade teacher nearing retirement identified her current class as one of the two strongest groups she had taught in three decades of teaching. Educators become adept at sizing up their classes.

Today, benchmark assessments are often used to evaluate teachers as well as monitor student growth. Benchmark scores should show progress, of course, but that progress may be inadequate in the eyes of school administrators—especially in lower-achieving, high-turnover districts when administrators do not know their staff or the students. The desire for better evaluations, in turn, becomes a rationale for teachers to leave such districts in search of higher-scoring students and the possibility of improved performance evaluations.

To compound the problem, districts may demand that teachers use one set of test-aligned books or software. This creates its own set of problems when those books or software are effectively unreadable by many, most, or even all the children in a classroom. Information sources should be matched to student reading ability levels, but inflexible attempts to teach standards with texts aligned to tests may prevent this matching. When the mismatch is great enough, class books or iPads can become an active hindrance to learning rather than an aid—a huge waste of money in districts with little or no money to spare.

What happens when we reward students for scores instead of learning? When we teach testing strategies to compensate for lack of learning, we undermine learning. When the emphasis of education is almost solely on test scores and schools become desperate in their efforts to improve those scores, we set up some children to fail. When the test ceases to be a measuring instrument but becomes, instead, the actual goal of instruction, learning can become merely one tool used to obtain higher test scores along with other tools such as cheating.

Students sense this. They understand that the apparent goal of instruction is to beat the test. Learning is not occurring for the sake of knowledge but for the goal of testing well. This understanding devalues learning, as we subtly teach students to sort information into two categories: (a) useful for the test or (b) probably irrelevant. The problem comes later when the irrelevant information that was thrown into the discard pile becomes useful or necessary.

As I started the day’s math toward the end of my last official teaching year, one of my students said aloud, More math? Why do we have to do math? The tests are over.

Why indeed? It’s not as if we were all there to learn mathematics or something.

Snapshot: Better Instruction for the Students Who Need It Least

Before test content steered instruction to the extent it does today, teachers often took time to explore personal passions with their students. They might spend extra time on World War I or add a lesson on the Tokugawa Dynasty in Japan. It’s worth considering that this time spent off or adjacent to the curriculum nonetheless produced national test scores that suggested students learned as much or more back then, despite such deviations from routine (The Nation’s Report Card, Trends).

The National Assessment of Educational Progress (NAEP) is America’s largest national exam. It tests reading, writing, mathematics, science, the arts, civics, economics, geography, and US history and is administered to students at grades four, eight, and twelve. The assessment uses the same booklets from year to year, carefully documenting any changes. By law, the test is managed by the US Department of Education. When I talk about the lack of progress from NCLB and the Common Core, I am basing that conclusion on stagnant, sometimes even falling, NAEP scores (The Nation’s Report Card, Data Tools, NAEP).

Today, deviations such as that extra lesson on the Tokugawa Dynasty may no longer be allowed in schools losing the test-score game, despite the benefits to students. Administrators in these schools too often expect all classroom instruction to be directly pointed at that annual spring test. Units of instruction without test benefit will be jettisoned. I suspect that the breadth of learning has been declining steeply in many academically challenged districts since NCLB and frantic testing took over the agenda. Loss of breadth becomes a direct response to fear of lost test points.

In schools where that fear is minimal or nonexistent, those extra off-the-curriculum lessons about the Tokugawa Dynasty, brain function, foreign dynasties, and Shakespearian comedies may remain in daily lesson plans. The rich get richer, as the saying goes…and the kids at the top of the testing heap get the most diverse and interesting classroom instruction.

Chapter 2

Testing Special Education, Bilingual, and Other Outliers

If we are going to judge our educational efforts by our test results, we need to be using the right tests. This would seem to be the absolute bottom-line of education. But the problem with living and dying or hiring and firing based on test results is that tests are only as smart as the people who write and administer them. Tests are meaningless unless they measure what they are supposed to measure in a reasonably efficient fashion. Otherwise, they waste instructional time and lead to faulty conclusions that directly affect the lives of our students, usually for the worse.

I lived through a fascinating example of this during my first year as a bilingual teacher when I was forced to give my students the ISAT. In 2008, Illinois mandated for the first time that all bilingual students who had been in this country for over one year had to take the ISAT in English. I did my part. I administered the test. I couldn’t teach during this time since I had to give the test. My students weren’t learning since they had to take the test.

What did Illinois bureaucrats and their federal counterparts get for this classroom time loss? They learned that my students knew little English. Of course, we knew this before the ISAT. My students had already taken the lengthy, multiday ACCESS test that documented they understood too little English to function in a regular English-language classroom. This was the very reason they had been placed in a bilingual program: these students didn’t know enough English to function academically without the benefit of occasional translation.

The bureaucrats may also have decided that my students lacked basic mathematics skills. In this respect, the test results were deceptive. Since a great deal of the math section consisted of story problems that my students couldn’t read, it didn’t actually test their math skills. I know for a fact that my students could have done some of those problems if they had known what the problems said. Since I was not allowed to translate the problems, all they could do was guess at what they were supposed to do.

A quick aside on this topic. The only translation allowed that year was for test instructions. In perhaps the silliest-ever accommodation allowed for bilingual students, the test administrator could repeatedly read math story problems aloud to students in English. In other words, I could read and reread the problem to them if I never did it in a language they understood. One way to put this into perspective might be to imagine what would happen if we gave students in America their standardized achievement tests in French after they had studied French for a year or two in school. We would learn little from such an exercise because we would be testing their French comprehension at a level that remained years of study in the future. Probably the only thing we would learn for our lost time and formidable testing expenses would be that, amazingly enough, our students understood little French.

Fortunately, a few more rational testing accommodations were later added for English Language Learners (ELLs). Students first received math story problems in Spanish on the ISAT and then on PARCC. Given that parents often rely on schools to teach written language and that US schools tend to teach English literacy using Spanish verbal crutches at the start (with some dual-language exceptions), that Spanish may be no easier to read than English. But the accommodation does help the small percentage of students who are more literate in written Spanish than English.

In human terms, in 2008 I handed my eighth graders a test they could not understand. The school had spent a great deal of time telling them how important the scores on this test were. I had prepped them to eliminate wrong answers when possible, preparing them to guess since I knew that was almost all they would be doing.

Then I watched Odriel for a few hours. He was a round-faced sweet boy who had arrived from Mexico a few years earlier. School was proving difficult for him. Like many immigrant children, he had attended school in Mexico only intermittently and had canyon-sized gaps in his learning. Nevertheless, Odriel could grasp new concepts quickly when he concentrated. I watched as he reached for his test, a smirk of confidence on his face. He knew he was good at standardized tests, at least the ACCESS test he was accustomed to taking. He normally did well on tests and the flak only started later when his daily work failed to measure up to his obvious potential. Odriel took up his pencil with determination. The minutes rolled by. His smile disappeared, the pencil faltered, and a frightened look appeared in his eyes. He’d been plunged in far over his head, and at this point, he knew it. A few more minutes passed. Finally, his face collapsed. A bleak sadness crossed his features as the last vestiges of hope fled. He looked up at me, betrayal in his eyes.

Across the years, Odriel, I apologize for having done that to you. You should never have had to take that test. We might as well have handed you the Medical School Admissions Test. The whole thing was ludicrous—and cruel. By the writing portion of the ISAT, Odriel was pretty much free-associating. He wrote a bunch of gibberish. When I handed this to the principal to send to the state, I could tell he was disappointed in me. Why had I not prepared this boy better for the test? Why had he not taken the test seriously? The principal didn’t understand that Odriel had taken the test seriously, so seriously in fact that I would be unable to keep him focused in class for the next few weeks. In some respects, I don’t know that I ever got Odriel back. I don’t know if he ever got his confidence back either.

A year after Odriel had to take his first and only ISAT, he left the country to live with his grandparents on their pig farm in Mexico. His mother explained to me that school was not working for him and he was much happier back on the farm. I’d say Odriel was lucky. At least he had someplace to go.

Odriel’s problem was not testing itself. As I noted earlier, testing is essential in education. However, our increasing reliance on standardized testing to evaluate academic performance has spun out of control and resulted in irrational testing.

If most or all my students flunk a test in my classroom, the two mostly likely possibilities are that I either failed to teach