Gaze data and Webgazer Visualiser

In our research, we took advantage of consumer webcams and open-source software to gather data from a larger group of students, demonstrating high scalability and cost-effectiveness of this approach. We aimed to study the differences in reading behavior between students reading regular text with illustrations and the same text with highlighted important terms. The subjects, first-year university students, filled out a questionnaire and introductory test in order to determine their knowledge level. A total of 80 full-time students were directly involved in the experiment. The students were subsequently randomly divided into two equally sized groups - the experimental group, which was enrolled in a variant of the e-course containing chapters with visually distinguished important terms, and the control group, with the same subchapters but without the highlighted terms. A total of 68 students took part in both measurements; however, only 59 achieved serviceable results from the first measurement and 61 from the second one. The data was analyzed both qualitatively using visualizations and quantitatively using statistical measures on the data and test results. We haven’t observed a significant difference between the experimental and control group in neither the gaze patterns, not their test results. We did however find a significant gaze pattern differences between the higher scoring students and the failing students.The first takeaway is the feasibility to conduct larger scale eye-tracking experiments. Most of the studies in this area use very small population samples due to the time and financial constraints of the methods using specialized eye-tracking hardware. Our method is scalable, as it relies on low-end hardware and outsources the recording to the students’ devices. The second takeaway is that although eye-tracking may not be useful to fine-tune text already optimized for readability, it may be a useful tool to identify and help struggling learners.