Comparison of Automatic Versus Hand Coding Techniques in Eye Tracking Software

School Name

Governor's School for Science & Mathematics

Grade Level

12th Grade

Presentation Topic

Consumer Science

Presentation Type

Mentored

Mentor

Mentor: R. Andrew Hurley, Clemson University

Oral Presentation Award

2nd Place

Abstract

In the fast paced world we live in today, researchers look for techniques that can increase productivity while decreasing research time. In the growing field of wearable eye tracking, studies traditionally take significant amounts of time, especially coding during data analysis. The time-consuming nature of hand coding can cause researchers to lessen the complexity of the study or cut the number of participants in order to meet deadlines. Tobii Pro Analyzer Software has offered a solution to this problem by introducing automatic coding, which can save a considerable amount of time and money. However, this software is in its infant stages and is still in the process of being refined. In this study, the data from 15 participants shopping for chocolate and cereal was analyzed using hand coding and automatic coding. Significance tests were performed to determine statistical differences in accuracy between the hand coded and automatic coded data. Overall, it was determined that there is no significant difference (p>0.05) between the hand coded data and automatic coded data for the chocolate or cereal planograms, which are graphical representations of the attention data of participants. These results are extremely beneficial to researchers, as there is now evidence that the automatic software works for eye tracking studies and thus should be implemented to save time and money.

Location

Wall 210

Start Date

3-25-2017 9:45 AM

Presentation Format

Oral and Written

Group Project

No

COinS
 
Mar 25th, 9:45 AM

Comparison of Automatic Versus Hand Coding Techniques in Eye Tracking Software

Wall 210

In the fast paced world we live in today, researchers look for techniques that can increase productivity while decreasing research time. In the growing field of wearable eye tracking, studies traditionally take significant amounts of time, especially coding during data analysis. The time-consuming nature of hand coding can cause researchers to lessen the complexity of the study or cut the number of participants in order to meet deadlines. Tobii Pro Analyzer Software has offered a solution to this problem by introducing automatic coding, which can save a considerable amount of time and money. However, this software is in its infant stages and is still in the process of being refined. In this study, the data from 15 participants shopping for chocolate and cereal was analyzed using hand coding and automatic coding. Significance tests were performed to determine statistical differences in accuracy between the hand coded and automatic coded data. Overall, it was determined that there is no significant difference (p>0.05) between the hand coded data and automatic coded data for the chocolate or cereal planograms, which are graphical representations of the attention data of participants. These results are extremely beneficial to researchers, as there is now evidence that the automatic software works for eye tracking studies and thus should be implemented to save time and money.