Semantic Differential Analysis

When performing UX research methods, there are several choices and paths you can go down. In my last blog I talked about a method called the System Usability Scale. In short, that method is a great way to measure functionality and usability within a system. The method I’ll be going over today is called the Semantic Differential Analysis.

Semantic Differential Analysis

According to UX Methods: A Quick Guide to User Experience Research Methods, this UX method measures emotional feedback from users, which differs from measuring functionality. The SDA method measures attitudes and feelings on a product by comparing opposing adjectives on a scale. This survey is an attempt to quantify user feelings based on patterns or inconsistencies. This method can be deployed before and after a re-design.

How does it work?

First, you need to define 5 – 10 opposite adjective pairs for your product and draw a line between each pair. Try to select words that relate back to each other. For example:
Easy <—–> Hard
Confusing <——> Clear

Next, have the participants use the system or product prototype. After this activity, survey the users. Have a user mark the points on each line between the two opposing words that best represents their attitude. Below is an example from the book, Essentials of Marketing Research: Putting Research into Practice. The question they researched was to evaluate customers’ last shopping experience.


Make sure to survey at least four other users to reveal patterns. Finally, you can chart the results together. Collect all results and display on a single chart. This is the best way to observe for patterns. In the below example, scores from two rival stores were displayed against each other for evaluation.


The Semantic Differential Analysis Scale is a great way to quantify user attitudes and feelings. Before or after a re-design of a product or service, you can deploy this survey to customers. If most responses appear to have a negative pattern or overall user feeling, than you have a lot of work to do! But the good thing is, generally, you should be able to tell what area you need improvement on. If consistent responses show that your product was dirty but was reliable, than you know exactly where you need to improve.

If you choose to use this method, good luck on your endeavors!