Card Sorting-Technical

Tasks set
The following four test processes assist in optimising the information structure of a web site and in providing users with easy access by means of a conclusive naming and categorisation. Ideally, these tests are conducted before or in combination with a card sorting.

Comprehension test
Goal: these tests investigate whether users understand terms correctly, e.g. on a web site, and whether their expectations match the information that is actually available there.Therefore

• the respondents are confronted with a series of terms which they have to explain as they understand them;
• terms misunderstood or unclear are recorded and discussed;
• users are provided with a scale that helps them in assessing how suitable a term is for describing a topic.

Process phase: Requirements Analysis- before a card sorting.
Sample: 20 to 30 participants (quantitative evaluation).
Result:

• Frequency of incomprehensible terms.
• Problems related to these terms and proposals as to their improvement.
• Average significance and values (if scales are employed).

Naming test
Goal: a naming test allows of developing comprehensible generic terms for topics or rather groups of subsumed concepts. Participants have to develop names for topics or categories of terms that are conclusive from their perspective. The categories are either given or established by the respondents in a prior card sorting.
Process phase: Requirements Analysis- subsequent to a card sorting.
Sample: 10 to 15 participants (qualitative evaluation).
Result: Proposals of names.


Prioritisation test
Goal: the prioritisation test determines the perceived relevance of topics. Respondents have to evaluate the importance of categories on the basis of a scale or a ranking. The categories are either given or established by the participants in a prior card sorting.
Process phase: Requirements Analysis - subsequent to a card sorting.
Sample: 20 to 30 respondents (quantitative evaluation).
Result: Ranking of the importance of categories based on mean values.


Classification
Goal: Verify by means of a process whether existing terms are suited to describe subsumed contents and to determine into which categories new contents should be placed. Participants are confronted with a series of topics that they have to organise into given categories. These categories are only described by their actual names, not in terms of their content. In order to test whether respondents choose the right path to the information, the classification may extend to several levels.
Process phase: Requirements Analysis
Sample: 20 to 30 participants (quantitative evaluation)
Result:

• Frequency of the respondents' guessing the correct content and misjudgements with regard to a category.
• Frequencies of classifying new contents.

Timing
Within the context of a card sorting, 3 to 4 weeks per test.