Open card sorting and factor analysis: a usability case study

Purpose ‐ The purpose of this paper is to describe the process, analysis, results, and implications of a card sorting usability study conducted during the planning stages of a web site redesign project at the University of Illinois at Chicago. Design/methodology/approach ‐ The methodology was based on recommendations from usability literature. An open card sort was conducted with 14 students and one faculty member using 93 cards labeled with content from the library's web site. The subjects were asked to "think aloud" and explain their rationale for sorting the cards. The researchers used statistical analysis software to run a factor analysis on the results. Findings ‐ The researchers extracted 11 categories of cards that loaded together and 27 cards that did not fit a category. The categories showed evidence of clustering by shared words, format, and process or task. Cards that did not load were standalone categories, or were redundant or meaningless to the subjects. Research limitations/implications ‐ The open card sort methodology and large number of cards resulted in cumbersome data that required specialized analysis. The qualitative data were critical to the interpretation of the quantitative data. Practical implications ‐ Libraries can use the process and analysis as a model for their own card sort usability studies. Results can be used to inform the naming of content and the creation of library web site architecture. Originality/value ‐ This study is unique in its use of the open card sort technique and factor analysis of the results. The results illustrate users' perceptions of library terminology and web site structure.