Closed card sorting tests content fit within a proposed structure
Aliases: closed card sort · predefined categories · category fit
What it is
Closed card sorting provides categories and asks participants to place content, testing category coverage, boundaries, and label interpretation. It evaluates content–category fit, not hierarchical navigation, entry salience, or task findability.
Why it happens
Provided categories reduce generation effort but constrain choice. Agreement can reflect clear labels or categories so broad that everything fits. Disagreement may arise from polysemy, overlap, or unfamiliar wording. Without a no-fit option, the task manufactures assignments.
Studying it
Analyze each card's category distribution, confusion pairs, no-fit responses, and reasons. Randomize cards, manage category-order effects, and compare segments. Use findings to hypothesize a structure and tree tasks to test findability.
Where it stops holding
A closed sort cannot establish the optimal structure, only fit among offered choices. Participants may learn categories during sorting, hiding first-use difficulty. Use open or hybrid sorting when novel organization is the goal.
Applying it
- Offer no-fit and uncertain choices with explanation.
- Inspect confusion pairs and coverage, not overall agreement alone.
- Separate label ambiguity, content polysemy, and structural overlap.
- Revise the candidate tree and validate it with realistic finding tasks.