upa - home page JUS - Journal of usability studies
An international peer-reviewed journal

Card Sorting: Current Practices and Beyond

Jed Wood and Larry Wood

Journal of Usability Studies, Volume 4, Issue 1, November 2008, pp. 1-6

Article Contents


Data Analyses

Current practice:

The ability to analyze card sorting data is a "touchy-feely" qualitative experience that can't be described beyond "eyeballing" and "looking for trends." Eyeballing the data is very tedious, time-consuming, and idiosyncratic, which results in a procedure or method that is not reliable. It works well when there are only five participants and a small set of items, but not so well with 30+ participants and a hundred items. While this general method can be useful as an auxiliary aid, we prefer to begin with the more formal cluster analysis and resulting tree diagram.

Recommendation:

Cluster analysis is a statistical procedure developed to analyze similarities and differences in the ways people categorize sets of domain items. The results can be used to produce a hierarchical tree diagram (formally called a dendrogram), an example of which is shown in Figure 1. While there are a number of cluster algorithms (Romesburg, 1984), we prefer a simple one based on the frequency with which content items are placed in the same category across all participants in a card sorting project. Essentially, the tree represents an average of the groups or categories produced by each of the card sort participants. It works on the principle of the similarity of each item to every other item, as sorted by the participants.

As shown in Figure 1, the alternate shaded groups represent the highest degree of similarity (i.e., items in groups of the same shade are more similar than items in different groups). An advantage of a hierarchical tree is that if larger groups than those appearing in the tree are desired for a Web site, two adjacent groups can simply be combined, because adjacent groups are more similar than non-adjacent ones.

Figure 1. Sample hierarchical tree diagram

Figure 1. Sample hierarchical tree diagram

Previous | Next