SendForm.net

What Is a Likert Scale? Examples, Questions and How to Use One

A Likert scale asks how strongly someone agrees with a statement, using a fixed set of ordered answers. This guide shows how to choose the number of points, word the labels and build one in a form.

Sophie ParkerPublished ยท 7 min read
Illustration of a hand selecting a response on a five-point Likert scale in an online survey, representing different levels of agreement.
Key takeaways The short version, before the details.
  1. A 5 point scale suits most surveys. A 7 point scale suits engaged respondents who can make finer distinctions.
  2. Keep the neutral middle unless you have a clear reason to force a lean.
  3. Ask about one idea per statement, and balance the labels so both sides carry equal weight.

What is a Likert scale?

A Likert scale measures an attitude. You give people a statement, and they choose how far they agree from a fixed, ordered list of answers. Psychologist Rensis Likert introduced the method in 1932, and it is still a standard tool in survey research.

Here is one statement with the classic five answers:

"The checkout process was easy to use."

  1. 1Strongly disagree
  2. 2Disagree
  3. 3Neither agree nor disagree
  4. 4Agree
  5. 5Strongly agree

Likert item or Likert scale?

Strictly speaking, one statement is a Likert item. The scale is a set of related items whose scores are combined to measure one topic. In everyday use, people call a single question a Likert scale too.

5 point Likert scale or 7 point: which is better?

Use a 5 point Likert scale unless you have a reason not to. It is quick to read and works for general audiences. A 7 point scale adds finer steps, which helps when respondents know the topic well and hold nuanced views.

More points are not automatically better. If people cannot tell two neighbouring answers apart, the extra step adds effort without adding meaning.

Compare5 point7 point
Best forGeneral audiences and short surveysEngaged or expert respondents
EffortLowerHigher, with more labels to read
DetailClear direction and strengthFiner differences in strength
LabellingEasy to word every pointHarder to keep the steps distinct

A 7 point agreement scale usually adds a softer step on each side:

  1. 1Strongly disagree
  2. 2Disagree
  3. 3Somewhat disagree
  4. 4Neither agree nor disagree
  5. 5Somewhat agree
  6. 6Agree
  7. 7Strongly agree

Whichever you pick, keep it consistent across a survey. Switching between 5 and 7 points makes results harder to compare and slows people down.

When should you drop the neutral option?

Keep the middle answer by default. A neutral opinion is often real, and removing it pushes those people to pick a side they do not hold.

Drop it, using a 4 or 6 point scale, only when both of these are true:

  • You need a lean in one direction to make a decision.
  • Every respondent has enough experience to form a view.

Neutral and "not applicable" are different answers. Someone who never contacted support has no opinion of it, so they should not land on the midpoint. Add a separate option for them, or show that question only to people it applies to.

Likert-type scale examples you can copy

Agreement is only one format. When a scale measures satisfaction, frequency or likelihood instead, it is usually called a Likert-type scale. The structure stays the same: ordered, balanced answers.

Agreement scale examples

  1. 1Strongly disagree
  2. 2Disagree
  3. 3Neither agree nor disagree
  4. 4Agree
  5. 5Strongly agree
  • The instructions were clear.
  • My manager gives me useful feedback.
  • The price is fair for what I get.

Satisfaction scale examples

  1. 1Very dissatisfied
  2. 2Dissatisfied
  3. 3Neither satisfied nor dissatisfied
  4. 4Satisfied
  5. 5Very satisfied
  • How satisfied are you with the delivery time?
  • How satisfied are you with the support you received?

Frequency scale examples

  1. 1Never
  2. 2Rarely
  3. 3Sometimes
  4. 4Often
  5. 5Always
  • How often do you use the mobile app?
  • How often do you finish your work on time?

Likelihood scale examples

  1. 1Very unlikely
  2. 2Unlikely
  3. 3Neither likely nor unlikely
  4. 4Likely
  5. 5Very likely
  • How likely are you to renew next year?
  • How likely are you to attend the next event?

What mistakes ruin Likert scale questions?

Most bad Likert scale questions fail because of the wording, not the number of points. Pew Research Center's guide to writing survey questions is a good reference for the wider rules. These are the mistakes to check first:

MistakeWeak versionBetter version
Double-barrelled statementThe staff were fast and friendly.Split it: The staff were fast. The staff were friendly.
Unbalanced labelsPoor, Good, Very good, Excellent, OutstandingVery poor, Poor, Fair, Good, Very good
Leading wordingOur award-winning support was helpful.The support I received was helpful.
Vague frequencyI regularly use this feature.How often do you use this feature?
Double negativeThe app is not difficult to use.The app is easy to use.

Also watch for two habits that creep in over a long survey:

  • Too many statements. Long grids tire people, and shorter forms get more submissions. Keep only the items you will act on.
  • Changing direction. Keep the negative end on the same side for every question, so nobody misreads the scale.

How do you create a Likert scale in SendForm?

SendForm's online form builder gives you two ways. Linear scale is the quickest and gives you an average. Multiple choice shows a written label on every answer.

Both start the same way: create a form, choose Hosted form as the form type, and open the Builder tab.

Build it with the linear scale question

  1. Add a Linear scale question from the Rating group.
  2. Type your statement as the question label.
  3. Set From to 1 and To to 5. Use 7 for a 7 point scale, or 4 or 6 to drop the middle.
  4. Fill in the end labels, such as Strongly disagree for Low label and Strongly agree for High label.
  5. Switch on Required if every respondent must answer.

A linear scale can start at 0 or 1 and end anywhere from 2 to 10. Respondents see numbered cells, with your labels at the two ends only.

Build it with a multiple choice question

In the builder, Multiple choice is a single-answer question shown as radio buttons, so each person picks one option. Use it when every point needs a written label.

  1. Add a Multiple choice question from the Choice group.
  2. Use Paste options to add your labels in one go, from lowest to highest.
  3. Leave the "Other" option off.

Leave shuffle off

The Shuffle choice order setting on the Options tab applies to Multiple choice questions. It would scramble your ordered labels, so keep it off on any form that uses this method.

How to read Likert results in the live summary

Copy the link from the Share tab and send it out. As responses arrive, open the Summary tab. Use Refresh to pull in the latest answers.

  • Linear scale: the summary shows the average, the number of ratings, a count for each point and your low and high labels.
  • Multiple choice: up to six options appear as a donut chart with a legend. Seven options appear as bars sorted by count, so read them by label, not by position.

Look at the spread before the average. An average of 3 can mean everyone feels neutral, or that half love it and half hate it. The bars show which one you have.

Each question is summarised on its own, so SendForm does not add several statements into one score. To do that, export your responses to CSV, which is a paid feature, and combine them in a spreadsheet. Sullivan and Artino's paper on Likert-type data explains how to analyse the results.

Online survey form with a five point Likert scale beside a bar chart of the collected answers
SendForm online forms

Need answers you can measure?

Add a linear scale with your own labels, share one link, and see each answer counted in the summary.

Build your surveyFree to start

Frequently asked questions

A Likert scale is a survey format that asks how strongly you agree with a statement, using a fixed, ordered set of answers such as strongly disagree to strongly agree. Strictly, the scale is several related statements whose answers are combined into one score.

Neither is always better. A 5 point Likert scale is quicker to answer, so it suits general audiences and short surveys. A 7 point scale gives more room for fine distinctions and suits engaged respondents. Choose one, then use it consistently across related questions.

Usually yes, because a middle opinion is a real answer many people hold. Remove it only when you need respondents to lean one way and all of them know enough to do so. If some people cannot answer at all, add a separate not applicable option.

A single Likert item is ordinal: the answers have a clear order, but the gaps between them are not proven equal. Scores summed across several items are often treated as interval data. For one question, report the distribution or median, and treat any average as a rough guide.

A rating scale is the broad category: any question that places something on a range, such as stars or 0 to 10. A Likert scale is a specific type that measures agreement with statements. Likert-type items borrow its format for satisfaction, frequency or likelihood.