Comparison of digital and printed text for learning

If you’re reading text for learning — where reading comprehension is critical — which is better: reading printed material, or reading the text online?

Singer, L., and Alexander, P. (2017) ‘Reading Across Mediums: Effects of Reading Digital and Print Texts on Comprehension and Calibration’, The Journal of Experimental Education, vol. 85, no. 1, pp. 155-172 [Online]. DOI: 10.1080/00220973.2016.1143794.


  1. Overview
    1. Texts on comprehension and calibration
    2. The challenges and possibilities of reading digital and print texts
    3. Lingering issues
  2. Method and measures
  3. Results and discussion
    1. Topic knowledge
    2. Medium preference
    3. Medium usage
    4. Comprehension across mediums
    5. Performance judgments and calibration
  4. Conclusions and implications


More or less, the answer is print!

Setting the scene

Technology has brought new possibilities for learning via text: new forms (such as multimedia books and longform articles), new tools (such as tablets and e-readers), and new presentations (such as photographic displays and non-linear text). We do a lot of reading online, so it is important to know if, and how, reading comprehension varies between digital and print text.

Previous research has found that:

On the other hand, many people strongly prefer digital texts. You can get texts faster and they are more portable.

There is an issue of calibration: the ‘contrast between predicted and actual performance’ (Singer and Alexander, 2017, p. 158). That is, do learners know which is best for their own reading comprehension? How well do they judge their own performance?

This study

This study involved 90 college students. On average, they were about 19 years old (so, “digital natives”). They were asked to read several texts related to their area of study interest. The texts were 450-word excerpts from textbooks and from New York Times news articles, written at about an 8th or 9th grade reading level.

They started with several basic questions, about whether they preferred digital or print for various reading situations, and if they used digital textbooks for their classes. They were told that their reading comprehension would be tested in this study.

After reading the passages, they were asked to identify the main idea, key points, and any other details they remembered, and then they were asked whether they thought they had better reading comprehension with the digital passages or the printed ones.

The findings

These students reported that overall they prefer digital text, especially for studying. They were somewhat more inclined to prefer printed text for very casual reading (like reading a magazine while on vacation). Almost 65% of them used a digital textbook for at least one class.

For the reading passages, they were able to identify the main idea about equally using digital and print text. (That is, 60% were able to identify the main idea, which I think is a real problem in itself!) However, they were better able to identify key points and other details from the printed passages. And yet, 69% thought they did best with digital text.

Digital text may be adequate for learners who only need the main idea, but they are not best suited if the learner wants to learn more deeply from the reading. The authors recommend that further study should be done into whether calibration can be improved by telling the students about the difference in their comprehension rates between the two mediums. Would they be willing to change their habits, or would they find that getting the main idea in a portable medium is sufficient for their desired outcomes?

See also

The same authors published another article (at the same time) detailing the results of their desk research as they attempted to find other articles addressing the same research questions: Singer, L., and Alexander, P. (2017) ‘Reading on Paper and Digitally: What the Past Decades of Empirical Research Reveal’, Review of Educational Research, vol. 87, no. 6, pp. 1007-1041 [Online]. DOI: 10.3102/0034654317722961.

Leave a reply