Reading: Screen vs. Paper
Screen Reading Worse for Comprehension, Leads to Overconfidence, New Meta-analysis Concludes
By Benjamin Herold on April 6, 2019
More evidence is in: Reading from screens harms comprehension.
According to a new meta-analysis of nearly three dozen research studies published over the past decade, reading from paper has a small, statistically significant benefit on reading performance.
One likely reason: Readers using screens tend to think they're processing and understanding texts better then they actually are.
Furthermore, according to a study presented here at the annual conference of the American Educational Research Association by University of North Dakota assistant education professor Virginia Clinton, readers using paper saw better performance without having to expend more time or effort.
"Reading from screens had a negative effect on reading performance relative to paper," according to Clinton's study, titled "Reading From Paper Compared to Screens: A Systematic Review and Meta-Analysis," published this year in the Journal of Research in Reading.
"There is legitimate concern that reading on paper may be better in terms of performance and efficiency."
For years, schools have worried about the effects of digital reading on student comprehension, even as they've flooded classrooms with digital devices and instructional software. Those concerns have been heightened by recurring findings that students tend to score lower in English/language arts on state standardized tests when they take the exams on computers, at least in the first couple years of online test administration.
In addition, previous meta-analyses have generally—although not always—found benefits for paper reading. Clinton's paper, for example, describes a 1992 review that found reading from screens to be slower and more fatiguing than reading from paper; a 2008 review that found no reliable differences in achievement scores on computer-based and paper-and-pencil exams; and a 2018 meta-analysis that found better reader comprehension when using paper.
To update those findings and take into account possible technological advances, Clinton's review included 33 independent studies involving 2,799 participants. All the studies used an experimental design with random assignment. Some of the participants were children and some were adults, but all already had foundational reading skills. All the studies were published in English, and none included participants with learning disabilities or visual impairments.
The studies' examination of reading performance focused on comprehension, or how well participants understood what they were reading. That includes both remembering what the text said, and drawing connections based on the text.
For both literal and inferential comprehension, the advantages from reading on paper were found to be significant.
It's also important to note, through, that such benefits from reading on paper were limited to expository texts. For narrative texts, generally regarded as easier to read and requiring less background knowledge to understand, no significant difference between paper and screens were found. That finding is consistent with the popular view that screen reading is most appropriate for "light pleasure reading," Clinton wrote.
The findings regarding reading performance were similar for adults and for children.
And the issue of "calibration accuracy," or being able to predict how well you understand what you just read, was particularly interesting, the paper said. "Calibration accuracy is better when reading text from paper compared to screens," according to the study.
"Readers may be processing texts from screens less efficiently ... as they think they are understanding the text better than they are."