A Likert scale survey question asks respondents how strongly they agree, disagree, or feel about a statement. It uses a fixed set of ordered answer options. It is one of the most common ways to measure attitudes, opinions, and satisfaction, because it turns a feeling into a number you can compare.
Writing a good one takes more than adding five radio buttons. Wording, answer labels, and order all shift the results. In this blog, we cover how to write, structure, and analyze Likert scale survey questions that measure what you intend.
What is a Likert scale survey question?
A Likert scale survey question is a closed-ended question that asks respondents to rate agreement, frequency, importance, or satisfaction on an ordered scale. A typical range runs from “Strongly disagree” to “Strongly agree.” Each answer option carries a numeric value, so you can measure attitudes quantitatively.
Organizational psychologist Rensis Likert introduced the technique in his 1932 paper, A Technique for the Measurement of Attitudes. He scored each response to a statement and combined those scores into an overall attitude measure.
His core idea still holds. An attitude is too broad to capture with one question, so you measure it with several statements. You will see this approach in:
- Social science and academic research
- Market research and brand tracking
- Customer satisfaction and product feedback surveys
- Employee engagement and training evaluations
Likert scale vs Likert item vs rating scale
A Likert item is a single statement with ordered answers. A Likert scale is a set of related items whose scores you combine. People use the terms interchangeably, and that habit causes analysis mistakes. For the wider topic, see our guide on what a Likert scale is. Boone and Boone note in Analyzing Likert Data that Likert-type data and Likert scale data call for different analysis procedures.
| Term | What it is | Example | How to report it |
|---|---|---|---|
| Likert item | One statement with ordered agreement options | “Ted’s Pizza has excellent customer service.” | Frequency distribution, median, mode |
| Likert scale | Four or more related items, scored and combined | Five statements that together measure opinion leadership | Combined score, plus a reliability check |
| Rating scale | A numeric or star rating with no agreement statement | “Rate your visit from 1 to 5 stars” | Mean and distribution |
| Semantic differential | Opposite adjective pairs at each end | “Cheap … Expensive” | Position between the two poles |
Which types of Likert scale survey questions can you ask?
Likert scale survey questions most often measure six things: agreement, frequency, importance, satisfaction, quality, and likelihood. The thing you measure decides which answer labels you use.
- Agreement: Strongly disagree, Disagree, Neither agree nor disagree, Agree, Strongly agree. Example: “The checkout process was easy.”
- Frequency: Never, Rarely, Sometimes, Often, Always. Example: “How often does your manager give you useful feedback?”
- Importance: Not at all important, Slightly important, Moderately important, Very important, Extremely important.
- Satisfaction: Very dissatisfied, Dissatisfied, Neutral, Satisfied, Very satisfied.
- Quality: Very poor, Poor, Fair, Good, Excellent.
- Likelihood: Very unlikely, Unlikely, Neither likely nor unlikely, Likely, Very likely.
Match the verb in your question to the labels. A question that asks “How often” cannot take agreement answers.
Step-by-step guide to writing a Likert scale survey question
You write a strong Likert scale survey question by defining one attitude and stating it in one plain sentence. Then you pair it with balanced labels. These steps keep each item clean.
- Define the attitude.
Decide what you are measuring, such as service quality or willingness to share product information.
- Write one idea per statement.
“The staff were friendly and fast” is two questions. Split it, as Pew Research Center advises for double-barreled questions.
- Use plain, specific words.
Name the thing being rated, such as “delivery time” instead of “the experience.”
- Match the labels to the statement.
Use agreement labels for statements and frequency labels for behavior.
- Balance the options.
Give the positive and negative sides the same number of points, with similar intensity words on each end.
- Mix the direction.
Add a few negatively worded statements so respondents read each item instead of ticking one column.
- Pilot the draft.
Ask a small group to explain each item in their own words, then fix any that they read differently.
Likert scale survey question examples
Good Likert scale survey question examples pair one clear statement with matching labels. The three groups below show how the wording changes with the goal. For more ready-to-use sets, see these Likert scale examples.
Customer feedback statements
Customer feedback statements work best when each one covers a single service detail. Use agreement labels from Strongly disagree to Strongly agree.
- “Ted’s Pizza has excellent customer service.”
- “You get a lot of food for the price at Ted’s Pizza.”
- “The online order form was easy to complete.”
Attitude and opinion leader statements
Attitude statements describe the respondent, not a brand. Together, these three items could form part of an opinion leader or early adopter scale:
- “People come to me for information on new products.”
- “I enjoy sharing information with others.”
- “I consider myself knowledgeable on a variety of issues.”
One statement tells you little. The combined score tells you where a person sits on the trait.
Workplace and frequency statements
Workplace surveys often switch to frequency labels, because behavior is easier to report than feelings. Use Never to Always.
- “How often does your team share progress updates?”
- “How often do you receive feedback you can act on?”
How do you choose the right Likert scale response options?
You choose response options by deciding how many points to offer. You also decide whether the scale is unipolar or bipolar, and whether to include a midpoint. These three choices shape the data more than anything else.
How many scale points should you use?
Five and seven points are the most common, and nine appears in some research. An odd number of points gives respondents a neutral middle. An even number removes it and pushes them to pick a side.
- Use five points for short mobile surveys and broad audiences.
- Use seven points when you need finer separation between respondents.
Our guide to 5-point vs 7-point Likert scales compares both in detail. Label the middle point “Neither agree nor disagree” rather than “No opinion.” The first signals a neutral view. The second mixes neutrality with having no view at all.
Should the scale be unipolar or bipolar?
A bipolar scale runs between two opposites, such as “Extremely dissatisfied” to “Extremely satisfied.” A unipolar scale measures one quality from none to a lot. An example is “Not at all helpful” to “Extremely helpful.”
- Choose bipolar when the topic has a true opposite, like satisfaction.
- Choose unipolar when you measure amount, like helpfulness or frequency.
The unipolar vs bipolar Likert scale guide shows both formats with sample questions.
How do answer order and wording bias Likert responses?
Agreement statements invite acquiescence bias, the tendency to agree no matter what a statement says. Showing the agree side first may also inflate scores, so you should test the order. Our guide to acquiescence bias explains the pattern in depth.
Pew Research Center reports that less educated and less informed respondents tend to agree more. The effect is stronger when an interviewer is present than in self-administered surveys.
Four practical fixes reduce it:
- Alternate the anchor order. Flip the agree and disagree sides within a survey wave, then compare scores during analysis.
- Mix positive and negative statements. Respondents who agree with everything then show up as inconsistent.
- Try a forced choice. Pew recommends asking respondents to pick between two statements as an alternative to agree or disagree formats.
- Keep wording neutral. Remove loaded adjectives so the statement does not signal the “right” answer.
Why measure one attitude with several statements?
One statement gives an unstable reading, so researchers measure an attitude with several statements and combine the scores. A person can misread a single item or answer it on a bad day. Averaging across items cancels much of that noise.
For a new scale, write far more items than you need. Then narrow the pool through data analysis, brainstorming, focus groups, or a literature review.
To check that your final items hang together, calculate Cronbach’s alpha. This statistic shows how consistently a set of items measures the same thing. The authors of Making Sense of Cronbach’s Alpha report that acceptable values range from 0.70 to 0.95. They also note that a very short item set lowers alpha, while a value above 0.90 may signal redundant items.
How do you analyze Likert scale survey data?
You analyze Likert scale survey data by first deciding whether you have single items or a combined scale. Single items produce ordinal data. The answers have an order, but the gaps between them are not guaranteed to be equal. Our guide to ordinal data covers the statistics in more depth.
| Data you have | Safe summaries | Watch out for |
|---|---|---|
| One Likert item | Frequency distribution, median, mode, top-2-box percentage | Reporting a mean as if the gaps were equal |
| Combined multi-item scale | Total or average score, plus standard deviation | Skipping the reliability check |
| Group comparison | Side-by-side distributions of the same item | Comparing items with different labels |
Top-2-box means the share of respondents who chose the two most positive options, such as Agree and Strongly agree. It gives a single number that executives read quickly.
Common mistakes with Likert scale survey questions
Most bad Likert data comes from a small set of avoidable mistakes in the statement, the labels, or the layout. Check every draft against this list.
- Double-barreled statements. One item asks about two things, so the answer is unreadable.
- Leading wording. “Our excellent support team resolved your issue” pushes respondents toward agreement.
- Unbalanced labels. Three positive options and one negative option skew the results upward.
- Switching scale direction. Reversing the order between questions without warning causes accidental wrong clicks.
- Too many points. A nine-point scale on a phone raises effort and rarely adds usable detail.
- Number-only labels. Digits without words mean different things to different people.
- Long grids. Twenty statements in one matrix invite straight-lining, where respondents pick the same column down the page.
How to build a Likert scale survey question in QuestionPro
The QuestionPro Likert scale question type lets you add a scale to any survey. You drag in the question, then enter your statements and answer options. You pick the number of points, such as five or seven, and choose unipolar or bipolar labels to match your statement.
Three setup details matter:
- Templates: Start from a Likert scale survey template instead of a blank page, then edit the wording to fit your topic.
- Editable labels: Apply the balance and wording rules above directly to every answer option.
- Mixed question types: Combine Likert items with matrix and other question types in the same survey software project, so one survey can measure several attitudes.
What makes a Likert scale survey question worth asking
A Likert scale survey question earns its place when it measures one clear attitude with words your respondents read the same way. The answer options are the smallest part of that job.
The better work happens before and after the survey goes out:
- Define the attitude first.
- Test the order and wording for bias.
- Check that your items agree with each other.
- Report results in a way the data type allows.
Treat the scale as a measuring instrument, and it will give you numbers you can defend.
Frequently Asked Questions (FAQs)
Use at least three or four related statements per attitude so you can test whether they agree with each other. Then drop weak items that lower Cronbach’s alpha. A single statement cannot show whether respondents read it the same way.
It produces quantitative numbers, but it measures subjective views. Pair a Likert item with an optional open-ended follow-up, such as “What is the main reason for your rating?” The numbers then come with context you can act on.
Yes, when space allows. Fully labeled points leave less room for interpretation than endpoint-only labels or bare numbers. Respondents then agree on what the middle options mean, which makes group comparisons more reliable.
Pew Research Center uses ordered scales, such as excellent to poor. It cautions that agree/disagree statements can inflate agreement, particularly in interviewer-led surveys. Pew often recommends forced-choice wording as the safer alternative for opinion questions.
Offer “Don’t know” or “Not applicable” as a separate option outside the scale. Placing it at the midpoint blends two different meanings: a neutral view and no view at all. That muddies your average and your neutral percentage.
