Exploratory study of web navigation strategies for users with physical disabilities
J. Eduardo Pérez, Myriam Arrue, Xabier Valencia, Lourdes Moreno · 2014 · Proceedings of the 11th Web for All Conference (W4A) · doi:10.1145/2596695.2596715
Summary
This exploratory study examined the web navigation strategies of 11 users with upper-body physical impairments, revealing substantial heterogeneity in how people with similar disability types interact with websites. Participants used a wide range of assistive input devices including joysticks, oversized trackballs, reconfigured mice, numeric keypads, touchpads, head pointers, on-screen keyboards, and keyboards with covers or handsticks. They performed two tasks on the Discapnet website (a Spanish disability information site conforming to WCAG 1.0 AA): five minutes of free navigation and a ten-minute targeted search task. The researchers analysed video recordings to capture detailed interaction data including pointing times, click failures, scrolling strategies, and back/forward navigation methods. A Barrier Walkthrough evaluation of the test website identified missing internal links and skip links as the most critical barriers for users with physical disabilities.
Key findings
The study revealed dramatically different navigation strategies even among users with the same type of disability. No participant used the Tab key to navigate links — all preferred moving a cursor with their input device, despite this being substantially slower for some. Pointing times varied enormously: one head pointer user (U11) averaged 59 seconds to point to a link in free navigation, while a reconfigured mouse user (U01) managed near-instant pointing. Keyboard and head pointer users consistently used keyboard shortcuts for scrolling and back navigation, taking negligible time, while joystick, trackball, and touchpad users clicked scroll bar buttons, sometimes taking over 33 seconds for a single scroll action. Joystick and trackball users had poor accuracy near targets due to overshooting, while keyboard and head pointer users were more accurate but slower to traverse distances. One participant had to abandon the search task due to physical exhaustion. The input device used, rather than disability type or expertise level, emerged as the primary factor determining navigation strategy and performance.
Relevance
This study challenges the common practice of treating "motor impairment" as a single user category with uniform accessibility needs. The vast differences in navigation strategies — driven primarily by input device rather than disability type or expertise — demonstrate that one-size-fits-all adaptations for physical disabilities are insufficient. The findings have direct implications for web design: larger click targets benefit joystick and trackball users who struggle with precision, while navigation bars with closely grouped links benefit head pointer users whose pointing time depends on cursor travel distance. The observation that keyboard shortcuts existed on the test website but no participant discovered them argues for making shortcut availability more visible. The study also provides evidence that adaptation should be personalised based on input device characteristics, not just disability category — a principle that remains underappreciated in current accessibility guidelines.
Tags: motor accessibility · physical disability · assistive technology · web navigation · input devices · adaptive interfaces · user research
Standards referenced: WCAG 2.0 · WAI-ARIA