Benefits of card sorting
If the participants of a card sorting exercise are representative of the user population for whom the application is being designed, then the result of the exercise - i.e. a taxonomy of some kind - will reflect a significant aspect of the user's conceptual model with regard to that application.
Allowing a group of participants to perform the same card sorting exercise provides valuable data over and above what it is generated by a single participant performing the exercise. This is because a group result will tend to iron out potential idiosyncrasies caused by individual bias or preference. If needed, statistical techniques - such as cluster analysis - can be used to combine the different "solutions" reached by the group.
Look out for...
When someone sorts a set of cards they are categorising based upon what is written on the cards. So be careful when you create these terms. It is all too easy to bias the potential groupings that will be found by the user, either deliberately or accidentally, by the choice of terms.
- Ensure that each term is as clear and unambiguous as possible
- Ensure that you have included all the items you need to categorize
- Shuffle or randomize cards prior to each participant session
- Script a set of instructions so that all participants have the same understanding of the process
- Avoid placing participants under unnecessary time pressure. Make sure they can contact you easily to ask questions or when they have finished. For example, be in the same room, but off to one side and out of the user's line of sight
- Be sensitive to a participant's wishes - some people feel more comfortable vocalising their decisions and talking things through with you. That's OK.
- Provide additional blank cards for people to write group names
- Provide rubber bands so that people can gather groups of cards together.
|
