The Structure of EZSort
EZSort is actually comprised of 2 separate applications: USort and EZCalc. USort provides an interface to sort cards or to enter sorted card results from a physical card sort. EZCalc analyses multiple users’ card sort data and provides a visual output of the results. Combined they provide a complete tool to perform a card sort and extract meaningful conclusions from the results.
EZSort Problems
For a program aimed at improving usability, EZSort has a sizable learning curve and a fairly counterintuitive interface. The main fault is its fractured organization as a series of screens and interfaces, as opposed to a unified application. Needing to perform a card sort with USort and then analyze the results in EZCalc introduces room for error and confusion in file management.
Separately, the two applications have their own functional shortcomings and they lack a unified feel. USort’s interface feels too outmoded to use in an actual user test. Thus, testing must be performed on paper or via another electronic interface and the results must be entered manually by the tester. Each step in the sort process must be performed sequentially making edits difficult. Also, there is a 100 card limit for a card sort. EZCalc is fairly straightforward to use and supports three types of cluster analysis: complete, single, and average. The hierarchical tree diagrams produced can dynamically show different groupings based on distance threshold limits. However, there are few options and adjusting the way the diagrams look is impossible. Also, the only way to get any precise numeric results is from the matrix view, which is not very compelling otherwise.
Room for Improvement in EZSort
There is certainly room for improvement in a card sorting application. A number of other card sorting projects are in various stagers of development. Some, like WebSort, allow subjects to perform card sorts over the internet, alleviating the need for on-site testing and providing a way to survey users nationally or internationally. Other improvements are mainly focused on improving the usability of the application itself and expanding the visualization and analysis tools. Many of these applications look promising and it is likely that a new industry standard may emerge in the next few years.