City Research Online

Identifying Web Usability Problems from Eyetracking Data

Ehmke, C. & Wilson, S. (2007). Identifying Web Usability Problems from Eyetracking Data. Paper presented at the British HCI conference 2007, 03-09-2007 - 07-09-2007, University of Lancaster, UK.


Eye-tracking research is increasingly used to supplement usability tests in both commercial and academic practice. However, while there has been research into links between eyetracking metrics and usability problems, this has so far fallen short of establishing a general correlation scheme between the two.

Consequently, practitioners are left to make subjective judgements when interpreting eye-tracking data. We address the lack of general guidance by proposing an initial correlation scheme based on data from an exploratory study which aimed to find a wide range of possible correlations between usability problems and eye-tracking patterns.

User testing of two websites was conducted and a set of diverse usability problems was extracted from the data; these were then analysed and some were correlated with users' eye-tracking patterns. In addition to this initial correlation scheme, a further finding from this study is that usability problems are connected to not just a single eyetracking pattern, but to a specific sequence of patterns. This sequence of patterns seems to arise from different coping strategies that users develop when a problem is experienced.

Publication Type: Conference or Workshop Item (Paper)
Additional Information: © Stephanie Wilson et al. The original version of this article is published in the eWiC series at
Publisher Keywords: Eye-tracking, User testing, Usability problems
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
R Medicine > RE Ophthalmology
Departments: School of Health & Psychological Sciences > Optometry & Visual Sciences
PDF - Published Version
Download (459kB) | Preview



Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login