Moving beyond the desktop: Searching for information with limited display size

Abstract

Increasingly, users are performing more sophisticated types of tasks, like information search, across computing platforms including desktops/laptops, tablets, and smartphones. While much research has been done to improve efficiency for each of these devices in the area of information search, few investigations have taken a pragmatic approach to determining the real efficiency costs across current state of the art devices and searching paradigms (typically browser based). The study examined comparative task execution times for searching tasks under three different conditions: varying screen sizes (desktop, tablet, smartphone), varying interaction devices (mouse & keyboard and touchscreen), and varying types of search (Within Document, Known Item, and Exploratory).

In addition, the impact of pagination technique (scrolling versus paging) on Within Document searching and the impact of using a generic website versus a mobile website for Known Item searching on a smartphone were investigated. The aim was to inform current practice on user efficiency across these variables and future design in improving efficiency on the small screen device. A within subjects (n=29) design was used comparing task execution times across three devices: a desktop, a tablet (iPad) and a smartphone (iPod).

Final results suggest that the typical user should expect to spend about 38% more time completing a Within Document task on the mobile device than they would at the desktop or on the tablet. On average, a user spent 203% more time completing a Known Item task on the mobile device than they did at the desktop (125% more than on the tablet) and about 35% more time on the tablet than the desktop. For the Known Item task, at least 38% of the extra time spent using the mobile device using a generic website is due to latency. For the Exploratory task, users found that the poor support for browser based tabs (multiple open windows) on the mobile and tablet made the search process more cumbersome. In general, user perception, in terms of task load and usability, aligned well with actual task execution times across the three devices and under the varying conditions.