October 15, 2019

It's Time to Rethink Education Policy:

Advice for the Democratic Presidential Candidates

By Alfie Kohn

More than 50 million children attend public elementary or secondary school in the U.S. The fact that so many voters spend so much time thinking about what happens to their kids in school means that the topic of education -- specifically, what comes between preschool and college -- should be a priority for a Democratic candidate.[1] And the focus should transcend political and economic considerations in order to address how our children are being educated.

The standard positions of political progressives are absolutely worth repeating and embracing: support for public education, which is a cornerstone of a democratic society; adequate pay for teachers; a commitment to equity and integration; and so on. But many people who talk this way ignore or even endorse policies that are troubling to thoughtful educators.

In a word, many political progressives are not educationally progressive -- or even mindful of the difference.

That's why we need to call attention to -- and offer a plan for changing -- how the current narrative in education is shaped by the same corporate perspective that colors mainstream views of taxation, health care policy, and other issues. The models, methods, and metaphors of business predominate in talk of "school reform." And this is uncritically accepted by too many left-liberals.

The standard story sounds something like this:

We need to educate our students to be competitive in the 21st-century global economy. Alas, they (and our schools) have fallen behind their counterparts in other countries. We must hold students and teachers accountable by raising the bar, employing tough and uniform standards for what is taught, measuring the results with frequent testing, and using rewards and sanctions (for states, districts, schools, teachers, and students) based on those scores.

Every single phrase in that paragraph is misguided. As a whole, it is ripe for challenge even when it isn't packaged with the rhetoric of privatization and competition that so often accompanies this narrative -- promotion of "school choice" with vouchers (mostly Republicans) or charters (Republicans and Democrats).

Here's what the presidential candidates ought to say instead:

1. It's time to affirm that the primary value of schooling can't be measured in dollars and cents. Education should be about supporting a vibrant democracy and doing right by our children -- nourishing their curiosity, helping them fall in love with ideas, promoting both the capacity and the disposition to think critically. That's very different from what we hear from most politicians and corporate executives, who frame education as an economic "investment," the point being to instill in students whatever skills and attitudes will enrich their future employers.

2. It's time to stop judging high-quality schooling by whether we're beating other countries who are also trying to improve education. Talk of "competing in the global economy," when applied to education, not only reduces learning to crude financial terms; it also treats people who live elsewhere as rivals to be bested rather than as potential collaborators. To rank countries by test scores is to treat education as if it were an athletic event where the point is to be able to yell "We're Number One!" This implies that we want children in other countries not to learn effectively -- a stance that is both intellectually and morally bankrupt.

Incidentally, for whatever these comparisons (and the exams that drive them) are worth, U.S. students actually do reasonably well, contrary to popular belief. But it makes no more sense to talk about the "quality of American schools" than it does to talk about the quality of American air. An aggregate statistic is meaningless because test scores are largely a function of socioeconomic status.  Our wealthier students perform very well when compared to other countries; our poorer students do not.  And we have a lot more poor children than other industrialized nations do.  For example, U.S. schools with fewer than 10 percent of students in poverty ranked first among all nations on PISA tests of reading a few years ago, while those serving a predominantly low-income student population ranked about fiftieth. The problem isn't bad teachers who need to be "held accountable"; the problem is poverty.

3. It's time to stop defining high-quality schooling by its difficulty level. Macho talk about "rigor," "tougher standards," and "raising the bar" confuses harder with better. What we give students to do can be too difficult as surely as it can be too easy -- and the main effect of overvaluing rigor is that the system is rigged to make sure some of them must always fail. (Worse, those who fail are disproportionately low-income children, children of color, and children whose first language isn't English.) Judging a school, a teacher, a book, or a test on the basis of whether it's sufficiently "rigorous" is like judging an opera based on whether it contains enough notes that are really hard for the singers to hit.[2]

4. It's time to stop defining high-quality schooling by how similar it is from one district or state to the next. The corporate model is all about standardization: a one-size-fits-all set of mandates that amounts to creating a centralized power over ideas and assumes teachers are not capable of working with their students to plan a curriculum. This trend reached its apotheosis with the Common Core, which was essentially adopted because a billionaire named Bill Gates thought it was a good idea and bankrolled it. Are all kids entitled to a great education?  Of course.  But that doesn’t mean all kids should get the same education.  Uniformity is not the same thing as excellence – or equity.  In fact, one-size-fits-all demands may offer the illusion of fairness, setting back the cause of genuine equity.

5. It's time to stop defining high-quality schooling in terms of its specificity. Harold Howe II, the U.S. commissioner of education under President Johnson, was once asked what national standards should look like if we had to have them. He summarized a lifetime of wisdom in four words: They should be "as vague as possible." Instead, a corporate approach to school reform -- based on B.F. Skinner's outdated behaviorist model of learning -- consists of transmitting a long list of narrowly defined facts and skills to children. This list features material that most of them will not need (given that virtually any information can be summoned on a phone) and that even successful students may not care about, remember, or be able to use.

6. Most of all, it's time to stop defining educational "success," "achievement," or "accountability" in terms of scores on standardized tests. These tests are used to monitor and enforce compliance to top-down standards and also as indicators of progress. The result has been untold damage to children and schools, to educational excellence and equity.

The less someone knows about how children learn (and how learning can be assessed), the more likely it is that he or she will insist on the use of standardized tests -- and then proceed to cite their results. The reality is that these tests mostly serve to make dreadful teaching appear successful. As a rule, they measure two things: (1) how much time has been taken away from meaningful learning in order to train students at taking tests, and (2) the size of the houses near the school. As much as 80 percent of the variance in test scores -- the difference between schools, between districts, and even between states -- can be explained purely in terms of socioeconomic status. Tell me how much a kid's parents earn and I'll predict her score with chilling accuracy. To use those scores to evaluate teachers or schools is therefore not only unhelpful; it's unethical. Politicians and economists like to talk about student achievement -- aggregate data based on the results of fill-in-the-bubble tests. Good teachers, by contrast, are concerned with students' achievements -- the ideas and accomplishments of specific children over time.

The damage of widespread and continuous testing is compounded if the tests are high-stakes, which means the results determine whether students are allowed to graduate, or how much teachers are compensated ("merit pay" is one of the most egregious features of corporate-styled school reform), or how schools are funded. This pressure, which is the engine of both the No Child Left Behind Act and Obama's Race to the Top, has driven many of our best teachers out of the profession since they want neither to be mere technicians nor to be blamed for factors outside of their control. (This exodus of talented educators is yet another way that a demand to "raise standards" has actually made schools worse -- especially schools in low-income areas, where the scores are lowest.) It has led to poor students dropping out. It has taken second-rate schools and turned them into third-rate test-prep factories.

An emphasis on "accountability" -- again, adopted by Democrats and Republicans alike -- blames students and teachers for what are primarily problems of unequal opportunity and inadequate funding. Low-income and minority students are thus punished twice. And their failings are measured with tests that are neither reliable nor valid instruments.

*

Senators Warren and Sanders have called for "big structural change" in our economic and political system. What does that look like in education? I believe it entails (1) a sharp reduction in federal micromanagement, coupled with (2) active efforts to support states and districts in developing and implementing more progressive practices, and (3) a shift from standardized tests to more authentic forms of assessment.

No Child Left Behind (unaffectionately known by many educators as No Child Left Untested), a reauthorization of the historic Elementary and Secondary Education Act (ESEA) of 1965, vested authority over education policy in the federal government to a degree that was historically unprecedented. Liberals old enough to recall when "states' rights" was code for segregation, and when federal troops heroically enforced court-ordered integration, were slow to understand the problem here. "Let the feds fix the schools" is at best a tenuous position because the likelihood of Washington's being a force for good depends on who is in charge at the moment. (Think of federal versus state initiatives on the global climate crisis right now.) More to the point, confidence in the federal government to bring about progressive changes in education is overstated during any political era: That there may be a role in enforcing integration doesn't mean good classroom teaching ever can, or should, be mandated by distant authorities.

A wry sign on display in some classrooms reads "The beatings will continue until morale improves." This is the position that corporate school reformers have adopted without irony. While there is surely room for disagreement about what constitutes pedagogical progress, federally enforced "accountability" and "tougher standards" is rooted in an intolerance for disagreement. It is profoundly undemocratic. Moreover, accountability seems to point in only one direction: Politicians and corporate titans dictating policy are never held accountable for the forces they've unleashed, no matter how damaging the results. The reliance on uniform standards and high-stakes testing has had approximately the same effect on learning that a noose has on breathing.

To be clear, the federal government does have an important role to play in

* enforcing civil rights and pressing for desegregation. (Lately it has more often been on the wrong side here, undermining promising local initiatives.)

* ensuring protections for English Language Learners and students with special needs

* equalizing funding. (As Richard Rothstein at the Economic Policy Institute points out, poor states spend much less per pupil than rich states, even adjusting for regional variation in spending power, whereas they ought to be spending more because they have more children living in poverty. But Washington exacerbates this inequality because ESEA aid is proportional to states' own spending. The states that need the most get the least.)

On other matters, however, such as curriculum and assessment, the first step is for the federal government to stop doing harm (of the kind listed in the six numbered items above). Ultimately it should simply do less -- which includes refraining from trying to compel local school districts to adopt even the ideas that progressive educators endorse. Such coercion rarely turns out well because, while teachers and school administrators may not resist change, they do resist being changed.

Calling for a scaled-down role is likely to find favor with conservatives who have been adamantly opposed to the Common Core curriculum standards in particular and are allergic to federal government initiatives in general. That alignment gives some of us pause, but, frankly, it also offers an opportunity to create alliances and perhaps even attract some conservative voters to a Democrat’s candidacy.

At the same time, it is important to realize that many states have essentially been saying, "We demand that the federal government stop imposing its terrible standards and tests on our communities. It’s the states’ job to destroy critical thinking and curiosity, and we’ll do that with our terrible standards and tests, thank you very much.” If you’re a teacher, it may not make much difference if oppressive dictates originate in Washington, in the state capital, or even in your district's central office. The point is still that your skills, and the unique interests and needs of your students, don’t count for much. So I suggest shifting the narrative from "The feds are bad" (which has disturbing implications when generalized beyond education) to "Let's support the autonomy of professional educators. We won't tell them how they must teach, but we will provide support for local communities to rethink the direction of education and offer to help them get there."

That support, as distinct from mandates or manipulation through financial incentives, may take the form of pilot projects, research initiatives, professional development events, and publications -- all of which can encourage experimentation with alternatives to the traditional behaviorist, "bunch o' facts," teacher-centered model with its reliance on grades, tests, homework, worksheets, textbooks, and lectures. There is a substantial literature demonstrating the value of more collaborative, student-centered learning that helps students to understand ideas from the inside out and become more enthusiastic about what they're doing.

The monumental failure of NCLB and Race to the Top did lead to a diminution of federal pressure in the form of the latest reauthorization of ESEA in 2015: the Every Student Succeeds Act (ESSA). The relief was modest, however, and it failed to challenge the continued reliance on standardized testing. This brings us to one of the most important features of a more humane and rational federal policy on education.

Standardized assessments are necessary only if our question is "Who's beating whom?" If our question is "How well are teachers teaching and students learning?" then assessments do not have to be identical from place to place. In fact, they do not have to -- and usually should not – consist of tests at all, let alone one-shot, stressful, paper-and-pencil tests. Testing is not only unnecessary; its results are profoundly misleading. Researchers have discovered that higher test scores are often associated with shallower learning. To that extent, those of us concerned about equity -- for example, ensuring that African-American and Latinx students are not receiving an education inferior to that provided to whites -- would find only misinformation if we depended on standardized tests. (Disaggregating garbage by income or race produces only more granular garbage.)

There is a huge literature in education on the theory and practice of "authentic assessment," almost entirely unknown to policy makers (who find it more expedient to reduce learning to a single number, pretend it is free of human judgment and therefore more objective, and assume that a bigger number signals progress). Authentic assessments track the quality of students' real classroom learning over time, providing "exhibits of mastery" that show what students can do with what they know. Sampling these rich individual learning records provides an overview of a whole school or district without the need to include results for every student. Some examples of authentic assessment strategies, as well as of schools that have shown that eliminating standardized testing is not only feasible but preferable, are available here from the organization FairTest.

Thus, the federal government should

* support states and districts in creating authentic assessments that support learning

* strongly discourage states and districts from using standardized tests -- particularly those that are (a) high-stakes, (b) norm-referenced,[3] (c) timed, (d) multiple-choice, (e) administered to children below fourth grade, (f) administered more often than once each in elementary, middle, and high school, or (g) designed to rank districts, states, or nations.

* roll back national standardized tests with the possible exception of the National Assessment of Educational Progress (NAEP), which should be broadened to include more performance-based items, as it once did. (Instead, the federal government is moving in the opposite direction, announcing plans just this year to cut funding for the one assessment that is potentially valuable.)

Politically speaking, a strong statement of opposition to standardized testing is likely to be timely. Impatience with the absurd overload of tests has been growing for years among students, parents, and teachers. Many parents not only resent having their children's intellectual capacity or accomplishments reduced to a single score but are alarmed at how much of their schooling has been hijacked -- in some cases, particularly in low-income areas, for virtually the entire school year -- in order to prepare them for these tests. This resentment has led to a grassroots “opt out” movement in some states to boycott tests.

But real leadership on this issue would consist of steering the conversation beyond problems with how the tests are used, to problems with the tests themselves; and from talking about how often students are tested to talking about whether these tests are needed at all (in light of the harm they do as well the availability of more informative and less destructive alternatives).

The last national politician who embraced these principles was the late Senator Paul Wellstone. As the thunderclouds of NCLB were gathering, he declared, “Making students accountable for test scores works well on a bumper sticker and it allows many politicians to look good by saying that they will not tolerate failure. But it represents a hollow promise. Far from improving education, high-stakes testing marks a major retreat from fairness, from accuracy, from quality, and from equity.” It is time to carry on that fight.

_________________________________________

NOTES

1. There are also 3½ million public school teachers, and they tilt to the left. Adopting the suggestions in this article would likely unleash their enthusiasm and capacity for organizing.

2. The eminent educator Deborah Meier offers us "Meier's Mandate": "No student should be expected to meet an academic requirement that a cross-section of successful adults in the community cannot." (Of course, many states' graduation exams require exactly this, with a diploma hanging in the balance.) "Kohn's Korollary to Meier's Mandate": Any public officials who talk sanctimoniously about the need to "raise the bar" and demand "tougher standards" should be required by law to take these exams themselves...and have their scores published in the newspaper.

3. A norm-referenced exam such as the Iowa Assessments or the Stanford Achievement Test isn’t designed to tell us how well the test-takers (or their teachers) are doing. It’s designed to tell us how each test-taker compares to all the other test-takers. So no matter how well or poorly they were taught, no matter how difficult or easy the questions, exactly 10 percent of the students (or schools) will always score in the top 10 percent. And exactly half will fall below the median and look like failures. Always.