Blogs About Reading
Shanahan on Literacy
Literacy expert Timothy Shanahan shares best practices for teaching reading and writing. Dr. Shanahan is an internationally recognized professor of urban education and reading researcher who has extensive experience with children in inner-city schools and children with special needs. All posts are reprinted with permission from Shanahan on Literacy.
Is Comprehension Better with Digital Text?
Do we read digitally as well as we read paper texts?
I’ve been asked this provocative question three times in three weeks. Once I was presenting a workshop on how to teach college-bound high-schoolers to handle complex text on tests like the ACT. This group wanted to know if it mattered whether students were tested digitally or with paper (studies estimate significant differences in performance favoring paper).
Last week, I was on a panel at Reading is Fundamental’s National Reading Coalition, a meeting of literacy providers, policymakers, and business leaders. This time the question was posed by Kathleen Ryan-Mufson, Director of Global Citizenship for Pitney-Bowes, a major player in digital communications. She wanted to know about the importance of digital literacy in learning, which opens up issues of access, precision of understanding, and student preference.
Then Friday, I was with a particularly thoughtful group of middle-school teachers in Indiana. They asked the question straight-up and were pretty sure that digital was better than paper because of technological affordances, such as easy in-text access to a dictionary, and because these kids are growing up digitally (the so-called “digital natives”).
Must be something in the water.
My answer: We don’t read as well digitally as we do on paper. When texts are short — a page or less — and comprehension demands light (what’s the main idea?), we do pretty well with either kind of text. But as learning demands increase and the texts are more extensive, paper wins hands down.
Like those Indiana teachers, students tend to think they read best digitally; but tests of their comprehension reveal that they are wrong.
Years ago, knowing such questions would come my way, I did some self study. I read a novel silently, usually prior to bedtime, I read one aloud to my youngest daughter, I listened to one on “Books on Tape” when I drove to work, and I read Dracula on my computer (thanks, Gutenberg Project).
My personal sense of the matter was that I was hurrying when I was reading digitally. As with current research findings, I was fine with major plot points, but it seemed like my understanding was fragile and not very deep. For me, at that time, reading online was more like skimming than reading. I was moving too fast.
Since then technology has improved and I’ve grown used to such reading. Engineers have improved digital texts, in lots of ways. We can now download texts so that we’re no longer “online.” Page sizes and formatting are more similar to those of real books; and screen illumination is better, too.
There are even ways in which tech books are demonstrably better. I can increase font sizes (which, at my age, I love) and I can set screen illumination so that I can read with the lights out and Cyndie can sleep. I spend a lot of time on airplanes and portability matters, so being able to bring along tech’s version of a dozen books and as many magazines is a definite win.
These days I often read digitally, for work and pleasure, much more often.
Nevertheless, reading digitally is still a different experience.
One loses the sensory pleasures of the page, and navigation can be disorienting. I can’t always go back and locate what I’m looking for. I still have a sense that I’m going too fast and, perhaps, reading too superficially. Though that just might be me. Kretzshmar, et al. (2013) found older readers do make shorter fixations when reading digitally, but that wasn’t true of younger readers.
Dillon (1992) and Singer & Alexander (2017) have conducted the most complete and thorough meta-analyses of the issues; the former looking at all the pre-1992 studies, and latter all the work since Dillon.
Both meta-analyses concluded that we don’t comprehend digital as well as paper, and that the disparity is as true for so-called “digital natives” as for people like me (“digital geezers?”).
Apparently scrolling a screen is more memory disruptive than simply turning a page. And, digital reading is often interrupted by multi-tasking (Baron, 2015): 67% of readers don’t last ten minutes before they’re messaging or shopping during reading!
Of course, this is all a bit complicated. Reading a PDF file on one’s computer is a different from reading a test passage on an online state exam or from reading Prairie Fires for pleasure on my iPad. They differ in their navigability, their user friendliness, and how likely one is to be tempted to do other things instead of reading.
That means comprehension is not always suppressed or limited by digital text, and yet it is often enough that we all should be concerned. Mangen (2013) found students could get the major plot points of a story digitally but that they were deficient when it came to making connections of other text points with the plot.
Maryanne Wolf (see Reader, Come Home: The Reading Brain in a Digital World) has agonized over the potential losses to patience, persistence, and depth of thought that could result from a daily diet of the short, peripatetic text excursions characteristic of digital reading.
Oh, and may I add that lots of people don't actually enjoy reading digitally as much as they do text on paper. (The last couple of Scholastic surveys have found that the overwhelming majority of kids much prefer books.)
Digital reading is superficial, less understandable, and less enjoyable for most people. Sounds like we should get rid of it, and that only fools would invest in digital texts for their instructional programs, right?
I strongly disagree.
Digital text is here to stay. There are all kinds of economic and social reasons why this is likely true, but what matters is that if I’m correct, then kids — all of us really — are going to need to learn to read such texts effectively.
Two things that need to happen:
First, many other writers (e.g., Boone & Higgins, 2007; Jabr, 2013; Kieffer & Reinking, 2006; Talaka, et al., 2015) have argued that tech engineers should continue to beaver away at making digital reading environments more supportive. Instead of trying to make tech readers like books, they need to think about how to produce better digital tools. Tech environments can alter reading behavior, so technological scaffolding could be used to slow us down or to move around a text more productively.
Second, we as teachers need to make students aware of their tech fallibility. Instead of romanticizing the tech savviness of everyone born since the first Apple sprung from the head of Steve Jobs, we should be teaching humility. They aren’t as good with these tools as they think they are, and the digital tools, while solving some problems, pose others.
Kids vary in their ability to locate information on Google, to evaluate such information, or to understand it. Basic reading comprehension ability helps with these things, as does amount of world knowledge; but even when those are high, students frequently struggle to take advantage of the affordance of digital text or even to understand what they read digitally.
Interestingly, not everyone’s comprehension is impaired by digital text. Singer and Alexander (2016) found a group of college students who actually did better; they slowed themselves and became more careful when reading digitally (unlike me and the majority of the students they studied).
We should be teaching students strategies for digital reading, fostering ways of reading that allow students to overcome the limits of the ways that they tend to adopt for screen reading. We should also teach them efficient ways of navigating in different screen environments (e.g., arrows, site maps, breadcrumb trails, non-linear navigation), and how to evaluate the trustworthiness of the digital information that they do locate.
Students don’t comprehend digital text as well as they do paper text. But they could.