A co-occurrence matrix is a powerful way of displaying and analysing results from card sorts.
In the example below, the numbers in the top and bottom rows and far left and far right columns, are the numbers of the cards sorted by respondent 1, using a different criterion for each sort (e.g. in our card sorts example, the criteria could be ‘whether it has a handle’, ‘what drink you’d have in it’, ‘what it’s made of’ etc.)
The numbers in the inner rows and columns show the number of times respondent 1 grouped two cards in the same category, i.e. the cards co-occurred. Card 1 and card 2 were grouped together in the same category twice, card 1 was grouped with card 3 four times, cards 1 and 4 co-occurred twice, and so on.
One important advantage of using co-occurrence matrices to analyse card sorts is that the cards are coded using numerals, and don’t involve words. So you could analyse card sorts done by someone speaking a language you didn’t understand, or by someone who couldn’t speak at all.
Co-occurrence matrices also allow you to pool results across respondents. You can simply add up the co-occurrences used by each respondent to produce a pooled matrix, as in the example below.
The pooled co-occurrence matrix shows you the extent of agreement or disagreement between respondents about similarities and differences between cards. There was most agreement between respondents that cards 1 and 3 were similar (17 co-occurrences), but little agreement about the similarities between cards 3 and 6 (4 co-occurrences).
You could of course produce co-occurrence matrices to compare responses between two respondents, or between sub-groups of respondents e.g. males and females, or between medical researchers and clinicians.
Copyleft Hyde & Rugg 2021