• Journal of Information Technology Services
• /
• v.19 no.1
• /
• pp.113-127
• /
• 2020

Current web accessibility checks and automatic assessments have been pointed out that the assessment items and scores are evaluated from the developer's point of view rather than from the user's one. In addition, although most of the grades of an automatic assessment on the public web sites are excellent because they are built in accordance with the web accessibility development guidelines, not a few web sites shows relatively low grades in evaluating their usability test made by those users. Taking into account the inadequacy of these web accessibility assessments, this study has identified the differences between the grades of usability evaluations and automatic evaluations for people with low vision and analyzed the major factors affecting web accessibility usability evaluations using Repertory Grid Techniques. Also, the Hard Laddering method of the Means-End Chain theory was adopted to visualize the relationship between Attributes-Conferences-Value and a hierarchical value system analysis based on FGI(Focused Group Interview) to people with the low vision. This study proposed the measures to improve the current web accessibility automatic assessment allocation, expert evaluation criteria, and user task assessment. In particular, it is a web accessibility user evaluation model that can consider the web accessibility quality certification criteria and user review assessment by directly analyzing the user cognitive structure and value system. This study is expected to be useful as a research to enhance the quality of web accessibility assessment.