Many fields suffer from the persistence of received ideas that are unsupportable, but the problem is especially severe in American education. In many cases, systematic research has illustrated that common current practices could be vastly improved on, but such improvements have not been implemented widely.

A 2019 report by Emily Hanford for American Public Media, “At a Loss for Words,” described how millions of students in the United States were still being taught to read using a flawed approach known as “three-cueing.” This method, Hanford writes, sprang from a 1967 paper that theorized that fluent readers recognized words not merely from their spellings but also from “contextual” factors such as sentence structure and accompanying illustrations. Primary education, the theory held, should teach students using the same process: not “sounding out” words but using only isolated phonetic components, such as a first letter, and combining these with contextual clues.

As Hanford describes, subsequent psychological research disproved the underpinnings of the three-cueing theory. It was actually less fluent readers who relied more on contextual clues while reading. Fluent readers may not have to sound out words explicitly, but that’s only because they were so practiced at associating sound to spelling that the whole process was automatic. Students taught with the three-cueing system, furthermore, often made bewildering mistakes while reading basic books, coming up with sentences that fit stereotyped grammatical patterns and the illustrations but diverged wildly from what was actually written.

One would hope that this debate would have been resolved by now. The three-cueing system, for instance, differs only mildly from older ideas about how to teach reading through “whole language,” which Rudolf Flesch fiercely attacked in his influential 1955 defense of phonics instruction, Why Johnny Can’t Read. Flesch noted, for instance, that the 1949 book Children Learn to Read gave a list of “seven different ways to recognize new or partly known words,” which included “the use of picture clues” and “the use of context clues” and, only last, “phonetic and structural analysis of the word.” And a number of systematic studies have found substantial benefits to phonics instruction, including for racial-minority students and as an intervention with low-performing students.

The debate over phonics—which Flesch’s work should have put to rest 66 years ago—reflects a broader problem. Consider Direct Instruction, an elementary teaching methodology developed in the 1960s by Siegfried Engelmann based on his experiences working with disadvantaged primary schoolchildren in Illinois. Direct Instruction delivers consistently positive results, and no better concise introduction to its ideas exists than several articles that Shepard Barbash has written for City Journal. Direct Instruction lesson plans teach a specific skill through explicit demonstration from the instructor and choral call-and-response with students; lessons follow careful scripts that teachers should not deviate from. By contrast, as Barbash describes, most early-childhood teachers take a different approach. “Central to the typical early-childhood educator’s worldview,” he writes, “are three ideas: that it’s better for young children to learn through play than through work; that children learn best and are happiest when they can help direct the pace and content of their own learning; and that a child’s mental abilities develop at a natural pace that adults cannot do much to accelerate.”

Direct Instruction succeeds with disadvantaged children, whose “language deficit” motivated Engelmann’s emphasis on oral call-and-response. A recent 2018 meta-analysis of hundreds of studies on Direct Instruction found universally positive effects. Even more promisingly, after several months of follow-up, the beneficial effects of Direct Instruction programs showed only insubstantial signs of fading away—a common problem with educational interventions, in which many promising-seeming interventions have only temporary benefits.

At least phonics and Direct Instruction have footholds in regular teaching, which cannot be said of another useful invention: spaced repetition. One of the oldest fields of research in psychology, dating back to Hermann Ebbinghaus’s calculations of “forgetting curves” in the late nineteenth century, is the quantification of memory: How likely is someone to remember a piece of information, such as a fact or a word in a foreign language, after reviewing it some number of times and then waiting for some period? Research in this field consistently concludes that “spaced” practice is superior to “massed” practice—that is, retention of old material requires continual review at intervals after originally learning it. This is the opposite of typical school lectures, which often explore material over a short period and then leave it behind until a review session.

Research on spaced practice, moreover, has produced one undeniably practical offshoot: “spaced-repetition software,” such as the free program Anki. Spaced-repetition software maintains a deck of virtual flashcards and shows users each day a subset of cards that they are likely to have forgotten. Almost any school subject that rests largely on the acquisition of a body of factual knowledge—for instance, history, geography, biology—could be improved by structuring courses around spaced practice and the use of spaced-repetition software, allowing students to retain vastly more information with the same amount of instruction and study time.

Yet spaced repetition has remained relatively obscure. I myself learned about Anki only from online hobbyist language-learning communities, which almost universally recommend it for building vocabulary. (I can also attest to its efficacy: after just one year of daily Russian vocabulary reviews with Anki and only casual study beyond that, I can now—with some difficulty—read nonfiction such as Russian-language Wikipedia. I made significantly less progress with the same amount of effort invested in studying German, a much easier language for English speakers, through Duolingo.) One of the few exceptions is medical school, where use of Anki to review for exams is widespread.

Why have clearly effective pedagogical techniques not seen universal or even broad use? One culprit is a set of deeply ingrained received ideas. The education profession has been averse to rote learning since the time of John Dewey, and even before; techniques such as spaced repetition, though effective, are undeniably tedious. Flesch speculated that the 1950s reluctance to teaching phonics—grounded in spurious notions of developmental “reading readiness”—stretched back to Rousseau’s Émile, the founding text of progressive education, which argued that children should never be made to learn things for which they do not feel ready.

Trends in education schools to emphasize teachers’ political mission over their pedagogical one—and even to dismiss the notion that teachers’ principal job is to teach a set of concrete, testable skills—make it unlikely that these ideas will get dislodged in the near future. Harvard’s education school, for instance, recently updated the website advertising its master’s degree to note that “the case for transformative education has never been stronger” and promises students immersion in a community with “a deep commitment to social justice.” A recent New York Times article about the low efficacy of remote learning in teaching basic skills such as reading and mathematics quoted a professor at the University of Washington’s education school who called concerns about learning loss “premised on a set of racialized assumptions” and advocated instead that schools refocus on lessons that students learned—in the Times’s paraphrase—“about loss and grief, about racism and resistance, about cooking and family traditions at home.”

Whatever the case, it is clear that better teaching methods could improve American students’ performance in many domains. Conscientious teachers and school board members—not to mention parents considering educational options for their children beyond the neighborhood public school—would do well to take note.

Photo: cglade/iStock


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next