Skip to main content
All CollectionsCard Sorting
Analyzing your card sort
Analyzing your card sort

How to analyze, similarity matrix, dendrograms

Chelsea avatar
Written by Chelsea
Updated over 3 years ago

Jump to sections in this article

Once you've received responses, it's time to start analyzing.

Viewing individual participant responses

To view the card sort analytics, head into your card sort project. Click the Participants tab, to view each participant's response.

View the individual responses by clicking on the purple icon to the right of the participants name. The modal shows the category and what cards the participant sorted into it.

To view all an aggregate of the analytics, click on the Analytics tab. You'll see 5 pages across the top:

  • Cards

  • Categories

  • Merged Matrix

  • Similarity Matrix

  • Dendrogram

Cards tab

The cards tab lists each of the cards and which categories participants sorted them into.

Categories tab

The categories tab lists the categories and which cards were sorted into them. It's another way of viewing the data shown on the cards tab.

Depending what type of card sort you decided to conduct, the categories tab will look different. For example, if you conduct an open or hybrid card sort, you'll have the ability to merge categories.

If you've conducted either an open or hybrid card sort you'll want to organize and analyze your research by merging some of the categories together. Since participants created some categories, it's important to standardize them. Merging categories together helps standardize. For example, imagine participant 1 names a category "Home products" and participant 2 names a category "Home cleaning". If both categories contain similar cards, you may want to merge them into a standardized category called "Home".

Alternatively, participants might have spelling typos when creating categories. When you are reviewing you notice one category called "Grocery" and another category called "Grocry". If both categories contain similar cards, you will want to merge them to a standard category called "Grocery".

Another reason you'll want to merge categories is because categories have similar names. For example, if one participant creates a category called "Health & Wellness" while another participant creates a category called "Health" and a third participant creates a category called "Health and Wellness" as the researcher you realize that all three categories are really the same and you want to merge them together so you can analyze the cards together.

Lastly, if multiple participants create the category "payments" it will appear as multiple different categories. To combine them, you'll need to use the "merge" feature.

To merge two categories together, select the check box to the left of the category names and the Merge Category button will become active.

When you click the blue Merge Categories button, you'll be asked to name the standardized category. Enter the name and click the Merge button to complete the merging process.

The new merged category will have a blue line to the left of the category.

If you've made a mistake and want to unmerge a category, click the the checkbox and then click the Unmerge categories button. The categories will then revert to their original state.

Merged matrix

In this tab, you'll be able to view the cards and categories and how frequently participants sorted cards into categories. The darker color the box, the more participants sorted a cards into a category.

You can view the grid in frequency count (Count #) or as a percentage (Percentage %) by using the toggle on the top right of the screen.

Anything that was merged will be reflected in this grid.

Similarity matrix

The similarity matrix is a table that shows the correlations between cards. If participants frequently group two cards together in the same category, they will have a higher correlation (i.e. the cards belong in the same category).

The darker the color, the higher the correlation is. If the box is white, it means that 0 participants grouped those cards together. For example, 0% of participants sorted Hair Care & First Aid together.

If a card appears close to each other on the similarity matrix, it means that they're related to each other. The boxes directly next to each other are most related. The boxes farthest away from each other are least related.

For example, Skin Care & Makeup were grouped together 100% of the time.

If you hover over a box, it will tell you the % of participants that grouped those two items together.

Dendrogram

Dendrograms visually show us what items are related. You'll notice your cards on the left side. If those cards are related, the line connecting them will appear closer to the Y axis.

Earrings, Jacket and Shirt are highly related which is why the red line connecting them is touching the Y axis. That means they should be grouped together under the same category.

Related article: Setting up a card sort

Did this answer your question?