A survey response format presenting statements with a symmetrical agree-to-disagree rating range, typically five or seven points. In UX research, Likert scales measure attitudes, satisfaction, and perceived usability — they're easy to administer at scale but require careful question framing to avoid acquiescence bias.
Common contexts
- Measuring perceived ease-of-use after each task in a usability study to complement behavioral observations
- Tracking satisfaction trends across quarterly surveys to detect sentiment shifts before they appear in churn data
- Comparing Likert responses across two design variants after an A/B test to add attitudinal data to behavioral metrics
Use when
Use Likert scales when you need quantifiable attitude data across a large sample — pair them with a follow-up open text field so you can explain what's driving the scores, not just measure them.
Avoid when
Don't use Likert scales as a substitute for observational research — they measure what users say they feel, not what they actually experience, and those two things regularly diverge.
Acquiescence bias is so consistent in Likert data that positively and negatively worded versions of the same question almost always produce different scores — always check which direction you've framed your statements.
Real-world examples
- Google's HEART framework combines Likert-based attitudinal surveys (happiness scores) with behavioural metrics like task success rate, giving product teams a multi-dimensional view of user experience quality.
- SurveyMonkey's default satisfaction templates use 5-point Likert scales because research shows they offer the best balance of granularity and cognitive load for general audiences.
- Apple's App Store uses a 5-point star-rating scale (a Likert variant) across 2 million apps, generating a dataset Apple's editorial team uses to surface high-quality apps in editorial picks.