Two of the oldest myths about reading are:
- We would be better people if we read more (for example here, here and here – but I could find hundreds of examples)
- We have only really read a book when we get to the last page.
Unfortunately, such attitudes, which are mainly to do with reading fiction, have been transplanted by many readers to academic reading. We kid ourselves that there is a certain kind of reading that is correct, even if we unconsciously don’t follow that “correct” reading in our academic practice. Here is a short investigation into what academic reading actually comprises.
Many institutions have help pages that try to distinguish different kinds of reading and to provide hints and training in those types. I picked an article from the University of Sheffield, which looked typical, dividing reading into three kinds:
- Scanning (looking for specific facts)
- Skimming (getting a quick overview of the content)
- Detailed reading, or focused reading
In practice there are many different kinds of reading. The above-mentioned University of Sheffield site, for example, has an interesting section on teaching yourself speed reading. (A glance at the Wikipedia entry for “speed reading” suggests that speed reading is largely what was referred to above as “scan reading”.) They also cover reading English as a second language, a topic that appears to be little covered by universities, at least on their websites.
Even if we accept there are these three types of reading, do they correspond with what academics actually do? There have been many studies of how academics read, for example, a series of large-scale studies by Carol Tenopir and her team at the University of Tennessee: one isTenopir, Volentine and King, 2012, “Scholarly Reading and the Value of Academic Library Collections” based on a survey carried out for JISC in 2011. While this article is as much about the library as about reading, it has some valuable insights into the type of reading carried out by academics. It was a large-scale study, with over 2,117 respondents, and showed the average academic reads around 39 “scholarly readings” (books, articles and other publications) per month, or 468 per year. There is a positive correlation between reading more and getting more citations for the articles you publish, by the way – perhaps it is true that the more you read, the more you succeed. But I think the reality is rather more complex.
The average academic spends 448 hours per year to carry out this reading of 468 publications. That comprises less than one hour per reading. The study does, commendably, show the time spent in browsing or searching for content before it is read in depth:
- 16 minutes for articles
- 11 minutes for books
- nine minutes for other publications.
These figures seem to comprise part of the total reading time. When asked how they find articles and books, researchers stated that more articles are found via searching than by any other method, while the biggest source of information about books comes from personal communication.
Another exceptionally useful table, table 4, summarises how long academics take to read each article or book. From the figures, 71 minutes is devoted to reading an article in detail, and 131 minutes to reading a book. While it is just about feasible that a scholar could read an article in detail in just over an hour, 131 minutes to read an academic book is remarkable. If the academic monograph is, on average, 90,000 words long, then researchers are reading the text at a rate of 41,000 words per hour – the fastest speed reading ever invented. Can it be possible for an academic to read in detail an academic book so quickly?
Perhaps “information foraging” is a clue to actual reading practice. Two academics from the University of Bath, Duggan and Payne, found that skim reading a full article resulted in better comprehension of the whole article than reading half the article in detail – so, for many purposes, skim reading is preferable. They describe a reading strategy called information foraging:
Suppose a reader monitors the rate of information gain while reading. Further, suppose that they set a threshold of acceptable gain. If readers begin to read linearly, they may continue until the rate of information gain drops below the acceptable threshold. At this point they will leave the current “patch” of text and skip to the beginning of the next patch. (Duggan and Payne, page 7)
The ”patch” may be a paragraph, or another page. Duggan and Payne measured comprehension by asking questions of the readers about what they considered important and unimportant statements in the text, as a way to assess the reader’s comprehension of the text. I think it is quite possible that academics, when they state they read in detail, are actually combining skim reading with detailed reading. That would explain my own reading practice quite well.
I don’t suggest for a moment that researchers don’t read anything in detail, only that our reading strategy may change from moment to moment as we take a view on the relative importance of the next “patch” of text for us. This strategy seems not to vary between print and digital reading. Which suggests that the dire claims by the Cassandras of digital reading, Nicholas Carr, James Evans, and others, about the terrible effects of digital content on our brains, are somewhat wide of the mark. We were reading badly, or terrifically well, depending on your viewpoint, well before digital content came along. Who knows, as a consequence of the digital environment in which we all live today, perhaps our skills at information foraging may actually be improving. If you factor in the time required to get hold of a print copy, compared to the time to find a digital version, you might see a further increase in overall reading speed. As a result of digital, we might, in other words, be reading better than ever before.