Reading from the screen: Good for education?
A conference paper by Geoff Kaufman (Carnegie Mellon University) and Mary Flanagan (Dartmouth College) called “High-low split: Divergent cognitive construal levels triggered by digital and non-digital platforms” (PDF available) has had some coverage [Science Daily] [Psychology Today] [PC Magazine]. The studies published in the paper compared levels of construal (perception and comprehension, or ‘gist’) of subjects reading the same matter from print, and a screen in an RCT (Randomised Control Trial).
There were four studies, each comparing performance in digital and print formats, as follows:
One: Completing the same survey (Behaviour Identification Form)
Seventy-seven participants (average age of 24.2 years), completing the same survey either on an iPad 2 or in print. After being randomly placed in the iPad or print groups, subjects completed survey testing their use of abstract and concrete descriptions of various behaviours and events. Care was taken to provide the same size text and page layout (using PDF for the iPad). Subjects were also asked to indicate their preferred means of reading, and use of digital devices. This first study found that subjects completing the survey on the iPad were more likely to favour concrete descriptions. In the words of the authors, “Study 1 provided initial evidence that digital mobile devices may indeed trigger a lower level of construal, compared to the level triggered by non-digital displays, for individuals processing the same information.” (p.2775).
Two: Concrete versus abstract understanding
Eighty-one participants (average age of 21.2 years) were randomly placed and asked to read the same fictional account, which was rich in detail. Again, PDF format was used. In a post-test administered by pencil and paper, in the words of the authors,
Three (A): Drawing conclusions from data presented in tabular format
Sixty participants (average age of 20.9 years), given fictitious factual information in the form of a table, each cell containing a rating about a Japanese car. As before, participants were randomly assigned to read the information on screen or in print. After two minutes, participants were required to select which of the four fictitious models was the superior one. The authors report that “A significantly higher proportion of print participants reported the correct answer (66%) compared to the digital platform condition (43%)” (p.2776).
Three (B): Orientating readers toward particular forms of engagement
One hundred and nineteen participants (average age 21.7) repeated study 3A, this time all using a PC laptop with one of three additional conditions. One third “were randomly assigned to complete a priming activity intended to activate a high cognitive construal level, a third… completed a parallel activity intended to activate a low construal level, and a third… completed neither of these activities” (p.2776). Of the participants in the first group (those stimulated toward high construal), 48% were correct. The low construal group scored 25%, the control group 30%.
Such are the studies and their findings. So, does this mean we should not present learning materials digitally to students, for fear of compromising their success? It’s important we interpret these results carefully – and alongside the rich findings of similar recent studies.
The findings of Kaufman and Flanagan are consistent with those of previous studies. In a recently accepted article entitled “Reading and studying on the screen: An overview of literature toward good learning design practice” (I’ll provide a link once it’s published), I overview much of the recent literature on reading from the screen vs reading print. My conclusions are, as relevant to Kaufman and Flanagan, as follows:
Regarding point 3 above, the comparison of text on screen with print is like comparing theatre to cinema, on the condition that the cinema version consist of a single, static camera facing the live production. Clearly, much is lost in the cinematic version! However, much more is possible in cinema. Multiple takes, varying camera angles, post-production effects (special effects and enhanced sound), and sharp editing result in a much more accessible, portable and often more powerful entertainment experience that can be enjoyed again and again. Similarly, while Shakespeare Live! may not be the same as being there, it does benefit from easy distribution and a more varied audience experience from an orchestrated series of camera perspectives. No metaphor is perfect, so I hope I'm getting my point across here!
Given that tertiary graduates will increasingly require the ability to engage with extended narrative on screen, it makes sense to expose them to on screen text and train them to properly use it. Various studies have shown that on screen reading skills can be learned. The Kaufman and Flanagan study reveals that readers do approach digital text differently; the challenge to educational designers is to encourage appropriate levels of reading, which should include leveraging the digital environment in ways that assist in student understanding.
There were four studies, each comparing performance in digital and print formats, as follows:
One: Completing the same survey (Behaviour Identification Form)
Seventy-seven participants (average age of 24.2 years), completing the same survey either on an iPad 2 or in print. After being randomly placed in the iPad or print groups, subjects completed survey testing their use of abstract and concrete descriptions of various behaviours and events. Care was taken to provide the same size text and page layout (using PDF for the iPad). Subjects were also asked to indicate their preferred means of reading, and use of digital devices. This first study found that subjects completing the survey on the iPad were more likely to favour concrete descriptions. In the words of the authors, “Study 1 provided initial evidence that digital mobile devices may indeed trigger a lower level of construal, compared to the level triggered by non-digital displays, for individuals processing the same information.” (p.2775).
Two: Concrete versus abstract understanding
Eighty-one participants (average age of 21.2 years) were randomly placed and asked to read the same fictional account, which was rich in detail. Again, PDF format was used. In a post-test administered by pencil and paper, in the words of the authors,
participants in the non-digital platform condition exhibited higher scores on the inference items… than did participants in the digital platform condition… The opposite pattern emerged for the detail-oriented questions, with digital platform participants exhibiting a better average score on these items… compared to their non-digital platform counterparts (p.2775).In other words, print readers were better at drawing out the higher level inferences of the story and digital readers were better at factual recall.
Three (A): Drawing conclusions from data presented in tabular format
Sixty participants (average age of 20.9 years), given fictitious factual information in the form of a table, each cell containing a rating about a Japanese car. As before, participants were randomly assigned to read the information on screen or in print. After two minutes, participants were required to select which of the four fictitious models was the superior one. The authors report that “A significantly higher proportion of print participants reported the correct answer (66%) compared to the digital platform condition (43%)” (p.2776).
Three (B): Orientating readers toward particular forms of engagement
One hundred and nineteen participants (average age 21.7) repeated study 3A, this time all using a PC laptop with one of three additional conditions. One third “were randomly assigned to complete a priming activity intended to activate a high cognitive construal level, a third… completed a parallel activity intended to activate a low construal level, and a third… completed neither of these activities” (p.2776). Of the participants in the first group (those stimulated toward high construal), 48% were correct. The low construal group scored 25%, the control group 30%.
Such are the studies and their findings. So, does this mean we should not present learning materials digitally to students, for fear of compromising their success? It’s important we interpret these results carefully – and alongside the rich findings of similar recent studies.
The findings of Kaufman and Flanagan are consistent with those of previous studies. In a recently accepted article entitled “Reading and studying on the screen: An overview of literature toward good learning design practice” (I’ll provide a link once it’s published), I overview much of the recent literature on reading from the screen vs reading print. My conclusions are, as relevant to Kaufman and Flanagan, as follows:
- Reading from the screen is perceived by readers as being a different genre to reading from print. Various studies show that readers approach reading from the screen differently than they do reading from print. Surface reading (or overconfidence) is common for on-screen reading. The issue is not so much the fact that text is on screen, rather the approach of the reader is not self-critical enough.
- Education design techniques geared to raise the readers’ construal with on screen text can be applied to reduce cognitive load, overconfidence, and the natural tendency toward lower construal. The final study (3B) clearly demonstrates that readers can improve their construal level if they are properly calibrated or oriented to the task at hand. While 48% is still short of the 66% correct attained by print readers, the significant gain is evidence that intervention can make a positive difference.
- Comparing text from the screen and in print is not a fair appraisal of the former’s utility in education. If all the on screen student is exposed to is the same text they could otherwise have printed, the educational opportunities of on screen have not been applied. Educational designers can – and ought to – apply on screen reading in ways that encourage effective engagement with the ideas in the text.
Regarding point 3 above, the comparison of text on screen with print is like comparing theatre to cinema, on the condition that the cinema version consist of a single, static camera facing the live production. Clearly, much is lost in the cinematic version! However, much more is possible in cinema. Multiple takes, varying camera angles, post-production effects (special effects and enhanced sound), and sharp editing result in a much more accessible, portable and often more powerful entertainment experience that can be enjoyed again and again. Similarly, while Shakespeare Live! may not be the same as being there, it does benefit from easy distribution and a more varied audience experience from an orchestrated series of camera perspectives. No metaphor is perfect, so I hope I'm getting my point across here!
Given that tertiary graduates will increasingly require the ability to engage with extended narrative on screen, it makes sense to expose them to on screen text and train them to properly use it. Various studies have shown that on screen reading skills can be learned. The Kaufman and Flanagan study reveals that readers do approach digital text differently; the challenge to educational designers is to encourage appropriate levels of reading, which should include leveraging the digital environment in ways that assist in student understanding.
Comments
Post a Comment