The Effect of Screen Size on Readability Using Three Different Portable Devices

Conclusions

As we noted in our discussion of the results, error rates were not statistically significantly different for either the one page or multiple pages experiments. Therefore, our hypothesis that error rates would increase as screen size decreased was rejected. Performance times were statistically significant for the one page experiment, but not for the multiple pages experiment when all subjects were considered. Therefore our results do not support our hypothesis that a reader using a smaller device will process small amounts of information, one page or less, quickly and easily, independent of the screen size, and that as the need for scrolling or paging increases, because of the small screen size, frustration levels will rise and performance times will increase. We were also unable to correlate overall difficulty or frustration to scrolling or paging, although we did find inverse relation to screen size. Our users did have a statistically significantly stronger preference for scrolling or paging on the Rocket Book, but since this is not the largest device, this result does not support our hypothesis.

If the outlier data is removed, then we did show a statistically significant difference in performance times between the Palm Pilot and the laptop for the multiple pages experiment. This result is reflected in an alternate refined theory.

Refine the theory

Since our primary results do not support the hypothesis our theory could be rewritten as follows. For small amounts of text, one page or less, users cannot read as quickly on a small device as they can on a larger one. As the number of pages increases, and scrolling or paging is required, the size of the device does not affect performance speed.

However, if the outlier data is removed, and the alternate one-way ANOVA is used, the theory would be rewritten as follows. For small amounts of text, one page or less, users cannot read as quickly on a small device as they can on a larger one. As the need for scrolling or paging increases, performance times will be statistically significantly different for smaller devices.

Suggestions for future researchers

A primary consideration for future researchers, is to increase the number of subjects performing the experiment in order to reduce the effect of outlying data. Another suggestion is to repeat this experiment, but change the task for the multiple pages experiment. The current task, "How many papers were presented at an 'ACM CHI Conference'? (Hint: How many times does ACM CHI appear in the reference list)." could be rephrased to create an unambiguous question. A researcher might then be able to correlate error rates to the type of device and screen size. Another possibility would be to alter the text of the documents used in the multiple pages experiment to remove the possibility of ambiguity in the question.

A further suggestion would be to make the subject group either more homogeneous, or more completely representative of the public at large. Again, the second option would require an increase in the number of subjects performing the experiment. Having either a more homogeneous group of subjects, or a more inclusive group of subjects would remove the possibility of skewed results.

Impact for practitioners

Since the Rocket Book had a statistically significant difference in its user satisfaction rating for paging, designers of similar devices might want to examine what it is about the Rocket Book's paging that appeals to users. Some possibilities to consider are the placement and size of the up/down buttons, because users preferred the paging capabilities of the Rocket Book over the Palm Pilot and Laptop. Subjects also rated the Palm Pilot highest for portability, and lowest for font readability. This suggests that designers of the smallest devices should work on making the text presented on the screen clear and legible. Decreasing the number of pixels per character and the contrast of the screen would be helpful on these devices. Some other possible improvements might be increasing the resolution of the screen, or slightly increasing the size of the font or the space between lines to make the font more readable.

Department of Computer Sciences: Direct questions and comments to the student editorial team