Likert Scale Examples: 3, 5, 7, and 10 Point Scales You Can Copy

A Likert scale looks like the easiest part of survey design: write a statement, add "strongly disagree" through "strongly agree", done. Then the design decisions start. Five points or seven? Should there be a neutral middle? Why do half the published scales run the labels in the opposite direction?

This guide gives you copy-ready Likert scale examples for the most common question types, and quick answers to those design decisions. If you want the scales generated and formatted for you, the free Likert scale generator builds 3, 5, 7, and 10 point scales you can paste straight into your survey.

The classic: 5-point agreement scale

The workhorse of survey research. Balanced, symmetric, with a genuine neutral midpoint:

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

Use it with statements, not questions: "The onboarding process was easy to follow", not "Was the onboarding process easy to follow?". Keep every statement about exactly one thing; "the app is fast and reliable" is two questions wearing one item's clothes. More wording traps like that are covered in how to write survey questions.

7-point agreement scale

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

Seven points capture more nuance and give you finer-grained data for statistical analysis. The cost is respondent effort: the difference between "somewhat agree" and "agree" is real to a methodologist and invisible to a tired respondent on a phone.

Satisfaction scale (5-point)

1. Very dissatisfied
2. Dissatisfied
3. Neither satisfied nor dissatisfied
4. Satisfied
5. Very satisfied

Note the pattern shared by all good Likert scales: balance. Two negative options, two positive options, symmetric wording. An unbalanced scale like "poor, fair, good, very good, excellent" has three positive options and quietly inflates your scores.

Frequency scale (5-point)

1. Never
2. Rarely
3. Sometimes
4. Often
5. Always

Frequency labels are vague by nature; one person's "often" is another's "sometimes". When precision matters, anchor them with numbers: "rarely (less than once a month)", "often (several times a week)". Vague quantifiers are fine for attitudes, risky for behavior you plan to report as fact.

Importance and likelihood scales

Importance (5-point): Not at all important · Slightly important · Moderately important · Important · Very important

Likelihood (5-point): Very unlikely · Unlikely · Neither likely nor unlikely · Likely · Very likely

The likelihood scale stretched to 11 points (0 to 10) is the basis of the Net Promoter Score. If you are running NPS specifically, keep the standard 0 to 10 format so your number stays comparable with everyone else's.

The midpoint question: should you offer a neutral option?

Arguments exist both ways. A midpoint gives genuinely neutral respondents an honest home; removing it forces a lean, which some researchers want. The evidence, on balance, favors keeping it: without a midpoint, truly neutral people either quit, skip, or pick a side at random, and all three are worse than measured neutrality.

Drop the midpoint (4-point or 6-point scales) only when you have a specific, defensible reason to force direction, and say so in your methods section.

Reverse-coded items, and what they are for

A reverse-coded item flips the direction of measurement: if your scale measures satisfaction, most items are phrased positively ("I would recommend this service") and one or two negatively ("I regret choosing this service"). Respondents who agree with both are not reading.

Two rules make reverse-coding work:

• Flip the scoring before analysis. A "strongly agree" on a negative item counts as a 1, not a 5, when you compute scale means. Forgetting this step is a classic thesis-week discovery.
• Use it sparingly. One or two reversed items per scale is enough to catch straightliners. A scale that zigzags every item exhausts honest respondents.

Reverse-coded items are one of several fair ways to catch low-effort responding; how to write attention checks covers the full toolkit, and the attention check generator gives you copyable checks to pair with your scales.

Mistakes that quietly skew Likert data

From scale design to analysis

Quick analysis notes your future self will thank you for. Single Likert items are ordinal: report medians and frequencies, and prefer non-parametric tests. Multi-item scales averaged into a score are commonly treated as interval data and analyzed with means, t-tests, and ANOVA. Decide before fieldwork which one you are building, because it changes how many items you need. And if you plan to compare groups on your scale scores, size the study with the statistical power calculator rather than a plain sample size estimate.

Frequently asked questions

What is a Likert scale?

A rating scale that measures attitudes by asking people to respond to statements along a symmetric range, classically five points from "strongly disagree" to "strongly agree". Named after psychologist Rensis Likert.

Is a 5-point or 7-point Likert scale better?

Five points for general audiences and mobile surveys; seven when respondents know the topic and you need finer discrimination. Both are defensible defaults, and consistency within one survey matters more than the choice itself.

Should a Likert scale have a neutral option?

Usually yes. Without a midpoint, genuinely neutral respondents skip, quit, or answer randomly. Force a direction only with a specific methodological reason.

Can I copy these scales into my survey?

Yes, that is what they are for. The Likert scale generator formats any of them (with your own question stems) ready to paste into any survey tool.

All of our calculators and generators live in the free research tools collection.