What is a Likert scale?
A Likert scale is a rating scale that measures how strongly people agree or disagree with a statement. It's named after Rensis Likert, the American social psychologist who introduced it in 1932 as a way to measure attitudes systematically instead of guessing at them. Nearly a century later it remains the most widely used question format in survey research, from customer satisfaction to employee engagement to academic studies.
The classic format presents a statement — "The onboarding process was easy to follow" — and asks respondents to pick a position on a symmetric scale from strongly disagree to strongly agree. Because the scale is balanced around a midpoint, it captures both the direction of an attitude (positive or negative) and its intensity (mild or strong).
One distinction worth knowing: a single statement rated this way is a Likert item; a Likert scale, strictly speaking, is a set of related items whose scores are combined to measure one underlying attitude. In everyday usage the terms blur together, and that's fine — but if you're measuring something complex like engagement or trust, multiple items combined will always be more reliable than a single question.
5-point vs 7-point scales
The 5-point scale is the default for good reason: it's fast to answer, easy to understand on any device, and coarse enough that respondents don't agonize over the difference between adjacent options. For customer feedback, event surveys, and most business use, 5 points is plenty.
A 7-point scale adds somewhat disagree and somewhat agree, giving respondents room for nuance. It can detect smaller shifts in attitude, which makes it popular in academic research and in tracking studies where you'll compare waves over time. The cost is slightly higher effort per question — on a phone, seven options is where a scale starts to feel crowded.
Then there's the midpoint debate. A neutral option lets genuinely undecided people say so, but it also gives disengaged respondents an easy exit. Some researchers drop the midpoint (a 4- or 6-point "forced choice" scale) to push people off the fence. The evidence is mixed; the practical advice is simpler: keep the midpoint unless you have a specific reason to force a choice, and whatever you choose, keep it consistent across your survey and across time.
How to write good Likert items
- One concept per statement. "The course was well-organized and engaging" is two questions wearing one trench coat. If someone found it organized but dull, their answer is noise. Split double-barreled statements.
- Use balanced, neutral wording. "How amazing was our support team?" pre-loads the answer. State something neutral — "The support team resolved my issue" — and let the scale carry the judgment.
- Keep the direction consistent. If agreement means "good" in one item and "bad" in the next, respondents misread and your data suffers. Within a block, keep polarity predictable.
- Handle agreement bias carefully. People tend to agree with statements regardless of content — it's called acquiescence bias. Researchers counter it by mixing in reverse-worded items ("The interface was confusing" alongside positive statements), but reversed items are easy to misread, so use them sparingly, word them plainly, and remember to reverse their scores before analysis.
Need a starting point? Our survey question generator drafts unbiased questions for any topic, and our sample size calculator tells you how many responses you need before the percentages mean anything.
How to analyze Likert data
The single most common mistake with Likert results is collapsing them into an average and stopping there. A mean of 3.5 can describe a group that mostly picked 3s and 4s — or a polarized group split between 1s and 5s. Those are very different findings, and the average hides the difference entirely. Distributions beat averages: always look at how responses spread across the options before summarizing them.
That's why practitioners lean on top-box and bottom-box reporting. Top-box (or top-two-box) is the percentage who chose the two most positive options; bottom-box is the two most negative. "62% agree or strongly agree" is concrete, robust to how you number the scale, and instantly comparable across questions and survey waves. The calculator above reports both.
Is a mean ever acceptable? Strictly, Likert responses are ordinal — the gap between "agree" and "strongly agree" isn't guaranteed to equal the gap between "neutral" and "agree" — so purists prefer medians and frequencies. In practice, means are widely used and useful for ranking many items or tracking a single item over time, especially when you're combining several items into a scale score. Report the mean if it helps, but show the distribution alongside it.
And the standard way to show that distribution is exactly the chart this tool draws: a diverging stacked bar. Negative responses sit on red tones, positive on green, neutral in gray — so one glance tells you the balance of sentiment, and the segment widths tell you the intensity. It's the visualization you'll find in research reports and UX benchmarks alike, because it shows direction, strength, and polarization in a single compact bar.
How to present Likert results
When you report more than one item, put each statement on its own diverging bar and order the items by agreement, most positive at the top. Ordered bars turn a wall of numbers into an instant ranking: readers see your strongest and weakest areas without reading a single percentage. Add the top-box figure as a label, keep the color scheme identical across charts, and resist 3D effects and pie charts — Likert data is comparative, and bars are what make comparison easy. The PNG download above gives you a clean, presentation-ready bar for each question.
Using Likert scales in Fomr
Fomr supports Likert-style questions out of the box: rating fields for single statements and matrix questions for rating a whole block of statements on the same scale — the classic Likert layout. Build the survey, share it by link or embed, and export your counts to chart here. If you're starting from scratch, our guides on creating a satisfaction survey and building a course evaluation form both walk through Likert question design step by step — and every Fomr form is free with unlimited responses.