Evaluation of Eye-Movement Metrics ina Software DebbugingTask using GP3 Eye Tracker

Teaching different programming subjects is an increasing challenge nowadays as because of growing user demands, the latest paradigms and technologies must be taught. Students in higher education now are of the age who were born into the digital world; however, the success of fulfilling programming courses is lagging behind. Humancomputer interface-based research has emerged in numerous fields of science recently, which could lead to the revolutionising of education. These interfaces could also help professors as a support system in transferring knowledge in a more efficient way besides supporting students acquiring adequate home learning methods. In this study, the applicability of eye movement tracking systems in respect of a programming task is examined, in which during the exploration and correction of the errors of an incorrectly functioning algorithm, the eye movement parameters are observed, recorded and evaluated. The test subjects participating in the research were divided into two groups according to some of their characteristic parameters, where the first group during debugging rather applied minor modifications and the more common technique of compile and run, which otherwise is also the most characteristic feature of students studying programming, while the members of the other group increased emphasis on analysing. In the statistic evaluation of the research, the parameters characteristic of the eye movement tracking of the two groups, as well as the efficiency of these groups were analysed. Based on the results, regarding the efficiency and the number of fixations, a significant difference could be shown between the two groups (U=22.5, Z=-2.236, p=0.025 (2-tailed), r=0.48) and (t(20)=2.507, p=0.021 (2-tailed), d=1.106), while concerning the duration of fixation and the saccade length, the difference shown was infinitesimal (t(20)=0.544, p=0.592 (2tailed), d=0.26) and (t(19.992)=1.965, p=0.063 (2-tailed), d=0.79).