The Fall, and Rise, of Reading

David Jolliffe has seen how much trouble students have had in recent years doing their reading.

He has given them a full article or book and watched them struggle to read anything beyond an excerpt. He has handed them a nonfiction book and heard them call it a novel. He has asked them to tell him the main idea of a text and watched their eyes search the page for answers.

“The main idea in a text is not on the page,” Jolliffe, who retired last year from the University of Arkansas as a professor of English and of curriculum and instruction, tells them. It’s “something in your head that you need to construct.”

Jolliffe’s perspective speaks to a broader sense of disquiet among professors about their students’ reading — the rate at which they do it and how well they understand it. Almost all professors, 97 percent, think it’s “important” or “very important” that students come to class having completed their reading, according to the 2018 Faculty Survey of Student Engagement. But just 3 percent think that students actually do all of their reading.

“Students these days are not as capable as students were in previous generations as critical readers,” Jolliffe says.

While it might be tempting to write off that sentiment as another “kids these days” lament, there’s also reason to think he’s on to something: A study published in 2000 by the psychologists Colin M. Burchfield and John Sappington compared 910 psychology students’ self-reports and performance on pop quizzes over time. They found that just 20 percent of students “normally did the readings” in 1997, down from about 80 percent 16 years earlier. Other studies on “reading compliance,” or the rate at which students do the reading, have found a broader, though still troubling, range: During a normal week — whether in two-year or four-year colleges, in the humanities or STEM — about 20 to 40 percent of students do the reading.

But a deeper look at the research on reading and other evidence paints a more complicated picture of one of higher education’s most basic operations. It is a picture that reflects the priorities of students and professors, the influence of broader social trends and mores, and the ambiguous role that educational attainment seems to have in encouraging — and sometimes discouraging — reading.

The picture is also easily fogged. For more than a century, scholars have fought over whether reading has declined. They’ve warred over whether problems came from schools or society at large. They’ve blamed students, instructors, standards, technology, conservatism, liberalism, demographics, recessions, divorce rates. In the late 1980s, when a decline in reading scores was blamed on the “permissive ’60s,” two scholars pored through a century of research to settle the question. The best they came up with was an “educated guess”: It was hard to say.

“The trends are in doubt,” they wrote. “The existence of literacy problems in our society is not.”

In other words, arguments about reading — what the researcher Louisa C. Moats has called “the most studied aspect of human learning” — are contentious in the way that a lot of things in higher education are contentious. “Reading is kind of at the fulcrum of education,” says Dolores Perin, a professor at Columbia University’s Teachers College. “So anything that can affect education probably affects reading.”

When It Started

You don’t have to go back far to find a culprit for reading’s apparent decline: the digital age. In 1997, a professor in The Chronicle bemoaned his children’s entrance into “a grand experiment in which the computer, the Internet, and the World-Wide Web are redefining literacy and reshaping the architecture of how they learn.” Since then smartphones have further divided people’s time and attention.

But the data aren’t so clear. “Measuring how much people are reading is a bit like judging levels of happiness,” wrote Naomi S. Baron, a professor emerita of linguistics at American University, in her book Words Onscreen: The Fate of Reading in a Digital World. Studies vary by scope, time period, and definition — one analysis may show a steep decline in time spent reading, while another shows an increase among many young people who read online books, magazines, and news.

Reading habits for class are better understood. The average college student in the United States spends six to seven hours a week on assigned reading, according to the National Survey of Student Engagement (which started tracking the statistic in 2013). Other countries report similarly low numbers. But they’re hard to compare with the supposed golden age of the mid-20th century, when students spent some 24 hours a week studying, Baron says. There were far fewer students, they were far less diverse, and their workload was less varied — “studying” meant, essentially, reading books.

The government started studying literacy more systematically as those trends changed, in the 1980s. The percentage of teenagers who said they “never or hardly ever” read for fun — which correlates with reading ability — started rising soon after. By 2008 it had roughly tripled, to 24 percent.

Reading in the classroom changed, too: The typical 17-year-old now reads fewer pages for school than the typical 9-year-old, according to data from the National Assessment of Educational Progress.

Meanwhile, students’ preparedness for the kind of reading they would do in college buckled as they grew older. “Only 51 percent of 2005 ACT-tested high school graduates are ready for college-level reading,” the ACT wrote in a 2006 report. “And, what’s worse, more students are on track to being ready for college-level reading in eighth and 10th grade” — about 62 percent — “than are actually ready by the time they reach 12th grade.”

Government data have shown a similar phenomenon. The scores of fourth- and eighth-graders on reading tests have climbed steadily since the 1990s, according to the National Assessment of Educational Progress. But those of 12th-graders have fallen. Just 37 percent of high-school seniors graduate with “proficiency” in reading, meaning they can read a text for both its literal and its inferential meanings.

The problem seems to extend to life after college. In 1992 and 2003, the National Center for Education Statistics studiedAmerican adults’ prose, document, and quantitative literacy (respectively, the ability to do things like read news articles, to read maps and food labels, and to balance a checkbook). The results, experts said, were “appalling.” College graduates’ math skills, statistically, hadn’t budged. But their prose and document literacy had declined. While those with bachelor’s and graduate degrees maintained the highest levels of literacy overall, those groups also experienced the steepest declines. Just 31 percent of college graduates were considered proficient readers in 2003, by that test’s definition, down from 40 percent in 1992. (International studies show similar trends. More data allowing comparison of adult literacy over time is expected this year.)

Advocates were baffled. The researchers shied away from discussing causes. But speculation in 2005, when the results were released, turned to familiar arguments: Higher-education institutions, in a rush for tuition dollars, had lowered standards to raise enrollments. College students’ scores for “intermediate” performance had gone up, implying, some said, a drop in rigor. Colleges weren’t adjusting their instruction for a more diverse student population, especially the greater numbers of Hispanic Americans who had grown up speaking only Spanish.

Moreover, one of the test’s directors said, young adults had given up reading for pleasure in favor of TV and the internet.

A similar trend holds true for the college-educated. The share of adults reading books of any kind on their own held steady from 2012 to 2017, according to the National Endowment for the Arts. The only segments of the population to significantly decline in that regard were women and adults with some college or with bachelor’s degrees.

What happens to reading habits during college? That was something Mary E. Hoeft wanted to know. Having taught at the University of Wisconsin at Eau Claire’s Barron County campus for 48 years, she’d noticed a persistent reluctance among students at the two-year college to do the assigned reading. So she decided to see what she could do about it.

Hoeft surveyed 124 students in two sections of a seminar meant to help first-year students make the transition to college: Had they done the reading? Those who said yes wrote three-sentence paraphrases, which she searched for any topics or anecdotes that could show basic comprehension.

The results, which Hoeft published in a 2012 paper, “Why University Students Don’t Read,” were sobering: 46 percent of students claimed that they had read the assignments, but just 55 percent of that group could demonstrate basic comprehension of the material. In effect, three-quarters of the class had not really read it.

The students cited all sorts of reasons for not reading, but one stood out: time.

Hoeft found a possible old-school fix, the same one Burchfield and Sappington praised 12 years earlier: quizzes. By adding quizzes or graded assignments, like handouts and journals, she could improve their reading rate by double digits.

She presented her results at conferences. After her talks, many professors would be skeptical. “They didn’t want to believe it, that they have to be testing their students on reading,” she says. “They believed their students were doing the reading.”

Hoeft heard what they were implying: Barron County was virtually open-access. Almost three-quarters of her freshmen were first-generation students. Almost a fifth of them were over 22 years old.

One professor, from an Eastern university, told her: “The problem is your students. Not my students.”

The Hivemind

Eric Mazur, a professor of physics and applied physics at Harvard University, started noticing issues with his students’ reading habits many years ago.

He tried administering quizzes and assigning reading summaries. He turned to “just-in-time teaching,” in which professors use student feedback about pre-class readings to explain points of confusion. The strategy helped create meaningful engagement, he says, but it didn’t tell him how much of the reading his students had done or how deeply they had thought about it. “I had no way of checking whether students were reading just one part and coming with a question, or synthesizing the whole reading,” he says.

Mazur started exploring solutions in an applied-physics course he taught with an advisee, Kelly Miller, by posting readings on an online platform. It allowed them to track how students read online.

In their own class, they found, “a very small fraction of students were actually complying, maybe a third,” Miller says. “All of this is here, it’s like Harvard undergrad. These are pretty highly motivated students to begin with.”

They tried different ways to motivate students to read, at first making use of the reading software and its discussion tools voluntary. “We quickly realized that unless you actually assign a grade for the out-of-class component, students just won’t do it,” Miller says.

She analyzed the behaviors of students in courses across the country that used the same reading software, and found the same pattern. “Harvard students are really not that different in terms of how they behave. They’re bright, they’re academically more gifted,” she says. But they’re also “incredibly good at figuring out how to do exactly what they need to do to get the grade. They’re incredibly strategic. And I think that’s really true of students everywhere.”

So they drew on one of the lessons of Peer Instruction, a method that Mazur had developed in the 1990s to encourage students to discuss conceptual questions with one another during lectures. Part of what makes it effective is that it turns the classroom into a social-learning environment. Why not do the same out of class, he thought, with the reading itself?

Online annotation software had been around for years. But Miller, Mazur, and a few colleagues decided to make their own — a program that would combine deep analytics and simple tools to nudge students toward habits that research shows lead to better understanding and classroom performance. They called it Perusall.

Students are required to read on Perusall by midnight before each class meeting. On the platform, they highlight and annotate around one another, seeing mutual points of understanding and confusion, and voting their peers’ comments up or down. Meanwhile, the software tracks exactly how long students spend on each page. It notifies them when someone engages with their comments. An algorithm grades the “thoughtfulness” of their annotations and compares their reading habits to behaviors that predict classroom success. (Students can appeal those grades, and many did at first. But Miller says the algorithm has improved, and appeals have vanished since they introduced a feature allowing students to click on their grade and see exactly how it was calculated.)

The grades were the stick. But just as important, the instructors say, was the carrot: the explicit connection of students’ readings to what was covered in class. A Perusall-generated “confusion report” helps the instructor walk students through what they didn’t understand.

The knowledge that professors were reading students’ questions and using them to guide what they taught in class, Miller says, turned out to be “a huge intrinsic motivator.”

Reading compliance soared. More than 90 percent of students read all but a few of the assignments they were given. The vast majority read almost every page. They did better on exams. And, Mazur found, students read beyond what was required: The software showed students coming back to the text after it was no longer assigned.

Diffuse Causes

If efforts like Mazur’s demonstrate anything, it’s that teaching strategies can change how students read.

That’s part of what some experts see as the problem. Educational attainment affects proficiency more than traits like age, gender, or family background do, according to the “Survey of Adult Skills,” part of the most recent wave of international literacy research. It just matters less than it used to, the government’s 2003 survey showed, for college graduates.

The ACT has argued that many educators simply “ignore” reading instruction in high school. State reading standards, so rigorous for early education, left high schoolers on their own — so far as such standards even existed, it said. Nothing, the ACT said, was more important, or overlooked, than teaching students with the kind of complex texts they’d encounter in college. (Some scholars now feel that newer Common Core standards, while controversial, have helped draw more attention to critical reading in high school.)

The data showing that reading habits and literacy are dropping are “cut from the same cloth,” says David Jolliffe, of Arkansas, who was also a chief reader for the AP English exam. To him, trends in English pedagogy and a turn toward standardized testing have upended students’ incentives to read: “We have young people who are coming away from high school with a very sort of test-driven training — I won’t call it education — training in reading.”

Instructors’ attitudes toward the importance of reading are part of the problem. Two-thirds of college instructors think their students come unprepared to read scholarly articles, according to the ACT’s “National Curriculum Survey” — yet they’re less likely than K-12 teachers to say texts of any kind are “important” to their own courses. Teaching students how to read in college feels “remedial” to many professors, Jolliffe says.

“There was not enough attention paid to the fact that when we get students at age 18, no matter what their high-school experience has been, you need to continue to teach them to read critically,” he says. “Critical reading is a developmental ability that grows over the years.”

But how much can professors be expected to do? They’re at least as busy as their students.

Faculty members are trained in their disciplines. “They don’t want to be reading teachers. I don’t think it’s a lack of motivation,” says Columbia’s Doris Perin. “They don’t feel they have the training.” Nor do they want to “infantilize” students by teaching basic comprehension skills, she says.

“By the time students get to college,” Perin says, “there are so many different kinds of gaps.”

Meanwhile, students feel pressed in several ways.

Time is one such pressure, and it forces them to be pragmatic. Students won’t read unless they feel it’s an “integral component” of their course, wrote Maura A. Smale, a professor and chief librarian at the City University of New York’s College of Technology, in a forthcoming paper.

She interviewed 30 students, many of them urban commuters, across three CUNY campuses, to understand how and why they read assignments. “They understand that reading is important,” and they’re not surprised they have to do it, she says. “But sometimes they’re hindered by factors outside their control.”

“I hate to use the language of business,” Smale says, “but they’re doing a cost-benefit analysis.”

Money is another precious resource, and, as a Vox analysis showed last month, the cost of reading has skyrocketed. The prices of both college tuition and textbooks became uncoupled from inflation around the late 1980s, a 2005 Government Accountability Office report found. From 2006 to 2016, book prices rose faster than tuition did. And many digital texts — with access keys, subscriptions, and no resale value — end up costing students more, according to advocates with the Public Interest Research Group, who have long criticized textbook publishers.

In 2014, almost two-thirds of students surveyed by the consumer-advocacy group reported not buying a textbook because it was too expensive. Virtually all of them worried that the choice would hurt their grades.

Access problems are “really, really, complicated,” Smale says. So students get creative. They mix PDFs, online textbooks, and print. They order cheaper used copies but may have to wait weeks into the semester to get them. Many use phones to take pictures of a classmate’s book or one from the library. “But that also affects how they read,” she says, “and how they can take notes.”

6 Steps

Evan Carton uses old-school methods as well as high-tech tools to teach his students at the University of Texas at Austin to read closely and thoroughly.

He teaches English 316, an overhauled general-education course that draws about 400 students to a large lecture hall, many of them nonmajor sophomores and juniors looking to meet a requirement. The course employs some of the flashier curricular experiments that have earned UT-Austin some notoriety. Students can take it in a traditional large lecture hall or as a “synchronous massive online class,” the slickly produced video format that, as The Chronicle has reported, has shown mixed results.

Tips for Getting Students, in All Disciplines, to Read

• Tie reading to a grade: Quizzes and assigned journals, which can determine about 20 percent of the final grade, can double or even triple reading compliance — but rote formats that seem to exist for their own sake can encourage skimming or feel punitive.
• “Do away with the obvious justifications for not doing the reading,” says Naomi Baron, at American U. “If you summarize everything that’s in the reading, why should students do it?”
• Ask students to make arguments, compare, and contrast — higher- order skills than factual recall.
• Using different media is fine, but maintain rigor. “You can do critical reading of anything that has essentially an academic argument in it,” says David Jolliffe, at the U. of Arkansas. Video and audio, in fact, may sometimes be better than textbooks — what he calls “predigested food.”
• Explicitly tie out-of-class reading to in-class instruction, going over points of confusion and connecting lessons and texts to each other.
• Teach reading skills. “Hundreds” of strategies exist, all of which make “explicit the processes that proficient readers use without thinking about it,” says Doris Perin, at Columbia.

But the principles of the course are bread-and-butter English. No laptops or phones in the lecture hall. The only required materials: the two-volume Norton Anthology of American Literature; Tim O’Brien’s The Things They Carried; and a clicker.

Students complete written exercises and a quiz every week. Each lecture starts with a “bridge” section, determined from their responses and online discussions, to connect the last class’s reading content to the current class.

“Even in the quizzes, I do my best to emphasize that the point is not for them to take away a series of disconnected facts,” Carton says. “It’s for them to see how certain elements of a text … need to be retained.”

But instructors don’t just tell students to read. They show them how to do it. A series of videos, aimed directly at bridging the gap between students’ high-school reading skills and college, lays out what professors call the CRIT method: the “close reading interpretive tool.” On a given night, a student might be assigned a 20-to-50-page Norton reading, along with a video teaching them critical-reading skills.

One video introduces the six steps of Texas’ close-reading method and links to a handout explaining them: paraphrase, observe, contextualize, analyze, argue, reflect.

It gives an example using an “unexpected source”: a character’s recitation of Walt Whitman’s “When I Heard the Learn’d Astronomer” in the TV show Breaking Bad. The next six minutes show close reading at work, tossing out simple summary in favor of argument. In their assignments, students can practice the CRIT method using an online form, or with pen and paper.

On the basis of in-class behavior and quiz performances, Carton says, he and his eight teaching assistants estimate that roughly 75 to 80 percent of students keep up with the readings and show comprehension.

Carton, like Hoeft and Mazur, has found success in getting students to read more comprehensively and closely. While these professors work in different disciplines and institutions, and their strategies and methods vary, they share a pedagogy recommended by scholars of reading and writing, one that makes how to read — and why — apparent.

For many instructors, such a pedagogy means focusing students’ attention so that they read carefully and critically. Sometimes this means assigning less reading. Naomi Baron, of American University, says she would prefer assigning three articles during the course of a semester if doing so would change how students think. It would be better, she believes, than having them read two dozen articles that they won’t remember — even if you can prove that every student read each one.

Professors have a choice. “A lot of faculty members, myself included, are saying, If they’re not doing the reading, we can get unhappy, we can get angry,” she says. “Or we can do something about it.”