Likert scale: examples, points and scoring
A Likert scale asks people how much they agree with a statement, from “strongly disagree” to “strongly agree”. Here is how to choose the number of points, write statements that measure one thing, and read the results honestly.
The standard labels
| Points | Labels, low to high |
|---|---|
| 5 | Strongly disagree · Disagree · Neither agree nor disagree · Agree · Strongly agree |
| 7 | Strongly disagree · Disagree · Somewhat disagree · Neither agree nor disagree · Somewhat agree · Agree · Strongly agree |
| 4 (forced choice) | Strongly disagree · Disagree · Agree · Strongly agree |
Other symmetric sets work the same way: frequency (Never → Always), importance (Not at all important → Extremely important), satisfaction (Very dissatisfied → Very satisfied). Label every point, not just the ends. Fully labelled scales are easier to answer consistently.
How to score a Likert scale
Code the answers 1 to 5 (or 1 to 7). For a multi-item scale, reverse-code negatively worded items (on 5 points, 6 minus the score) and sum or average them per person. For individual items, the distribution tells the real story.
Interactive: why the average can lie
Set how many people picked each point. Watch the mean against the median and the “top-2 box”. Then press Split opinion.
Mean
3.43
Treats the gaps between points as equal — a debated assumption.
Median
4
Agree. The safer summary for ordinal data.
Top-2 box
60%
Share choosing the two most positive points. Bottom-2: 27%. Mode: 4.
30 responses. The mean and median agree when answers cluster; they drift apart when answers pile up at both ends.
Example Likert scale questions
Customer experience
- The checkout process was easy to complete.
- My order arrived when promised.
- I found the information I needed on the website.
Employee engagement
- I have the tools I need to do my job well.
- I feel comfortable raising concerns with my manager.
- I understand how my work contributes to the team’s goals.
Course and training feedback
- The session objectives were clear.
- The pace was right for me.
- I can apply what I learned this week.
Product and app usability
- I could complete my task without help.
- The wording on the screens was clear.
- I would use this feature again.
Writing good Likert statements
- One idea per statement. “The staff were friendly and knowledgeable” is two statements.
- Avoid double negatives. Disagreeing with “I am not unhappy” is a puzzle, not a measurement.
- Mix positive and negative wording carefully in long scales to catch people ticking straight down the column, then reverse-code.
- Be specific. “Support replied within a day” beats “Support was good”.
Build a Likert grid in the survey maker: add a Likert scale field, type one statement per line and choose 5 or 7 points. On phones it shows one statement at a time with numbered buttons, so no sideways scrolling.
Questions people ask
What is a Likert scale?
A set of statements that respondents rate on a symmetric agree–disagree scale, usually 5 or 7 points. It was introduced by psychologist Rensis Likert in 1932. Strictly, the “scale” is the sum of several such items; a single one is a Likert-type item.
Should I use a 5-point or 7-point Likert scale?
Five points is easier on phones and for general audiences. Seven points gives slightly finer distinctions and suits experienced respondents. Research generally finds reliability levels off beyond about seven points.
Should a Likert scale have a neutral midpoint?
Include one when “no opinion” is a genuine answer. Removing it forces a choice, which can inflate agreement or annoy people who truly are neutral. If you drop it, consider a separate “Don’t know” option.
How do you analyse Likert scale data?
For single items, report the distribution, median and top-2-box percentage. Many researchers also report means, especially for multi-item scales summed into a score, but be aware this treats the points as evenly spaced.
What’s the difference between a Likert scale and a rating scale?
A Likert item measures agreement with a statement (strongly disagree → strongly agree). A rating scale asks for a judgement such as quality (poor → excellent) or stars. Both are ordinal.