The Likert scale is a response scale that goes from five to seven points, asking the customer or interviewee to agree or disagree with a statement. The scale presumes that the intensity of someone's attitude or opinion is linear, going from strongly agree to strongly disagree, and assumes that attitudes can be measured.
Likert data can be a reliable source of information about people's attitudes, as long as you are aware of the tendency for people to agree with statements. For example, each of the five (or seven) responses can be counted in terms of how often a sentiment is expressed, or given a numerical value that measures the attitude being explored.
To reduce the impact of this bias, researchers can ...
- Questions should be phrased as statements: If you present customers with facts, you can learn more about different aspects of your business. The score averages can help you track customer satisfaction over time and work to improve it.
- Be sure to consider both positive and negative statements when making your evaluation, in pairs, for consistency: One-off sentiment measurements might not give you an accurate picture of customer opinion. Try measuring customer sentiment in multiple instances to see if there is agreement, disagreement, or neutrality.
When to use Likert scale
A Likert scale can measure statements of agreement from researchers across disciplines, such as UX, marketing, and customer experience.
The scale can also be used to measure:
- Agreement: Strongly agree → strongly disagree
- Frequency: Often → never
- Quality: Very good → very bad
- Likelihood: Definitely → never
- Importance: Very important → unimportant
If you want to get a more accurate measurement of everyone's responses, it's worth asking them to agree or disagree with multiple statements. You can then combine or average a person's responses.