Rating scales for surveys: definitions, examples, and differences

Rating scales offer survey respondents flexibility, providing a fast method of giving feedback without burdening customers. But, knowing which rating scale to use for certain questions is vital for their success.

In this article, we’ll explore survey rating scales, question types, and tips for using them effectively.

A survey rating scale is a typical response format to a close-ended, multiple-choice question. Respondents will select from a series of options to express how they feel or what they think in response to a close-ended question.

There are several survey questions that use rating scales:

Close-ended questions, like those that researchers use when exploring the topics above, all utilize survey rating scales. 

Related reading: What is an interval scale and how to use it in your survey questions

Rating scales can come in all shapes and sizes. Some rating scales make use of words, others use numbers, and some even use symbols like stars or smiley faces.

Here are a couple of survey rating scale examples showing how to ask rating scale questions.

In a Net Promoter Score® (NPS) survey question, businesses aim to better understand customer loyalty. They achieve this through NPS surveys and by using a rating scale of 0-10.

Another common rating scale is a worded one spanning from ‘Strongly agree’ to ‘Strongly disagree’ with a certain statement. Below, you can see this type of rating scale in action.

One of the core differences that sets rating scales apart is their use of numbers, words, or both. Using rating scales with different formats can allow people to express their opinions or ideas in distinct ways.

Some rating scales replace words with symbols to enhance their readability and make responding more intuitive. For example, instead of Strongly Agree/Disagree, researchers could replace these options with smiley faces representing enjoyment levels.

Numbered rating scales are a simple but effective way of collecting data from survey respondents. The exact numbers you use on a scale can vary, often ranging between 0-10, 1-5, or 1-100 in some cases.

A numbered rating scale is a popular option for many close-ended questions. You’ll find numbered scales in the vast majority of surveys.

You can also make numbered rating scales interactive with slider rating scales. However, it’s important to note that this can create usability issues, especially on mobile.

Worded rating scales are ordinal scales that allow survey respondents to express their opinions using relative options.

While these question responses do use words, they are still inherently tied to numeric values. For example, the five options—Strongly agree to Strongly disagree—align with a 1-5 scale, simplifying numerical analysis.

Unipolar and bipolar rating scales are two different formats for structuring worded rating scales.

A unipolar rating scale starts from zero and moves in a single direction. This form of rating scale would be useful when measuring the degree of someone’s interest, satisfaction, or frequency. For example, “How often do you use our services?” could use a unipolar rating scale of “Never” to “Every day.”

A bipolar rating scale, on the other hand, is when your rating scale has two extremes. Respondents will then select where they fall on the scale of completely one thing to completely another. Opposing variable questions, like “Strongly agree/Disagree” scales, are bipolar with two distinct ends.

It's important to consider whether your rating scales should be unipolar or bipolar. This decision will help generate precise and useful data for analysis.

Related reading: 3 types of business-changing insights you can get from surveys

The Likert scale is a rating system that measures respondents’ opinions, thoughts, or emotional responses to a topic or statement. Typically, the Likert scale uses a 5 or 7-point ordinal scale.

The Likert scale can measure customer likelihood, satisfaction, and the importance they assign to a topic. 

Another approach is the Guttman scale, where questions follow a cumulative pattern so that agreeing with harder items automatically implies agreement with easier ones. For instance, if someone can run a marathon, a Guttman scale assumes they can also run a 5K. Learn more about how to use Guttman scales in survey research to discover if this hierarchical approach fits your measurement needs.

The Likert scale is an extremely useful rating scale to measure customer opinions. For example, a company may use the Likert scale to determine the effectiveness of its customer support team. They offer respondents the opportunity to respond from “Extremely responsive” to “Not responsive at all.”