Likert scale question · Rating

Likert scale questions

Ask how much people agree with a statement, from Strongly disagree to Strongly agree, with a label on every point. Not applicable and Don’t know stay out of the mean, and you get the share who agree, the mean and the median.

Available onFree & up· every plan
The Likert scale question as respondents see it on a laptop and a phone

Live demo

Answer it. Your answer joins the results.

This is the question your respondents get. Pick a point, or Not applicable or Don’t know, and the results beside it recalculate with yours counted.

How much do you agree or disagree? The onboarding process was easy to follow.

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

In action

The Likert scale question in action

Likert scale in action · One respondent agrees and moves up to Strongly agree with the arrow keys, two others pick Not applicable and Don’t know, and the results update.
  1. 0:00A respondent sees the statement and five labelled points, from Strongly disagree to Strongly agree.
  2. 0:03They click Agree, then press the right arrow to move to Strongly agree.
  3. 0:06On a phone, a second respondent sees the points stacked as full-width rows and taps Not applicable.
  4. 0:09A third presses Don’t know. The scale dims and can’t be picked until they press it again.
  5. 0:12All three join the results: the bars move, and N/A and Don’t know join the base but not the mean.

Every feature

Everything the Likert scale question can do

A labelled agree scale, with Not applicable and Don’t know handled properly, logic on the range, and labels that follow the answer into your data.

On the respondent’s screenWhat people answering see and use.

LiveFree

A label on every point

Give each point its own label, from Strongly disagree to Strongly agree. On a wide screen each label sits under its point, and the number stays on the button.

LiveFree

Stacks on a phone

The scale is laid out by the question’s own width, not the screen’s. When it’s narrow, as on a phone, a labelled scale turns into full-width rows with each label beside its number.

LiveFree

End and middle labels

Or label just the ends, and the middle, in a row under the points.

On an even number of points the middle label isn’t tied to a point.

LiveFree

Agree first or disagree first

Show Strongly agree first with one switch. Only the screen changes: Agree is still stored as 4, so your data and charts read the same either way.

LiveFree

Keyboard and screen readers

The scale is one Tab stop. The arrow keys move along the points and Not applicable and wrap at the ends, and each point is announced with its number and label.

Try it in the demo above: Tab to the scale, then use the arrow keys.
LiveFree

Required, soft required or optional

Required by default, and Not applicable or Don’t know count as an answer. Or warn once before a skip with your own message, or make it optional.

LiveFree

Ask why, under the scale

Add a comment box under the scale, optional or required, to hear the reason behind the answer.

LiveFree

Translated labels

The question, every point label, the end labels, Not applicable and the non-answer labels translate per survey language.

Not applicable and Don’t knowAnswers that aren’t a point on the scale.

LiveFree

A Not applicable choice

Turn on Offer a Not applicable choice and an N/A button sits under the points, in the same group, so the arrow keys reach it. Give it your own label.

LiveFree

Don’t know and Prefer not to answer

Non-answer options add pills under the scale. Picking one dims the scale and takes it out of the tab order. Add Refused or your own, each with a code you choose.

LiveFree

In the base, out of the mean

Not applicable and Don’t know count in the number who answered and show as their own rows. The mean, median and Top-2-Box use only the answers on the scale.

LiveFree

Checked on the server

Answers are checked again when they’re saved. A point outside the scale, N/A when it’s off, an unknown code, or a point sent with a Don’t know is refused.

Logic, piping and scoringRoute and personalise on agreement.

LiveFree

Logic on agreement

Skip, show or hide later questions when the answer is at least, at most, or between two points, such as Agree or above.

LiveFree

Quotas on a range

Build quota cells on the same conditions, “is between” included.

LiveFree

Pipe the label

Insert the answer into a later question as its label, so the follow-up reads the way the respondent answered.

LiveFree

Score a negative statement

There is no reverse-score switch. To score a negatively worded statement the other way, add a computed variable that reads the scale as a number, such as 6 minus the answer.

Reversing the order on screen doesn’t change how an answer is scored.

In the builder and your dataSet-up, imports, cleaning and labels.

LiveFree

Presets from 3 to 11 points

Pick 3, 5, 7 or 10 points and the scale runs from 1 to that number, pick 0–10 for 11 points, or type your own range.

Set the range before the labels: each label stays with its point, and a point a new range leaves out loses its label.

LiveFree

Live preview

The editor previews the real scale with your labels, order and N/A as you type.

Non-answer pills show in the survey preview, not in the editor’s preview.

LiveFree

Import from Excel, Word or PDF

A SCALE or LIKERT row in a sheet comes in as a Rating Scale with its range. From Word, PDF or text, a Likert with statements comes in as a grid.

In a sheet, the options cell comes in as the point labels. A single Qualtrics agree or disagree item comes in as a labelled Rating Scale, and a Qualtrics battery as a grid. Word and PDF imports use AI credits.

LiveFree

Correct an answer

Change a respondent’s point, or set it to Not applicable or Don’t know, when cleaning data.

LiveFree

Labels that travel

Your point labels follow the answer: into Results, crosstab rows, dashboards, decks, and the SPSS and Excel value labels.

Crosstabs are on Solo+, SPSS export on Professional+.

Not part of the Likert scale: stars and faces (see Rating scale), a slider (use Slider), a 0–10 likelihood to recommend (use NPS), attention checks, and a reverse-score switch.

Statement batteries

Several statements, one shared scale

To ask about several statements on the same agree scale, use a grid: one row per statement, one column per point. Each statement is still reported, cut and exported on its own.

  • One scale for every statement. Type or paste the statements and the scale points once. Shuffle the statements, add a Not applicable column, and choose a table, a list, an accordion or a carousel, with its own layout on phones.
  • Results per statement. Results shows a heat-map of answers per statement and point. In crosstabs each statement is its own row with a Mean and Top-2-Box (Solo+), and tab books add a summary of Top-2-Box across every statement (Research Team).
    The Surveti Results card for the battery as a heat-map: a row for each statement (The setup guide was clear, I knew who to ask for help, I was productive in my first week) with its base n, a column for each point from Strongly disagree to Strongly agree and a Not applicable column, each cell the share of that statement’s answers, darker for more, then the statement’s Mean and Top-2
    Results as a heat-map: each cell is the share of a statement’s answers on that point, darker for more. Not applicable has its own column and stays out of each statement’s Mean and Top-2.
  • In your deck. A report slide shows a Top-2-Box bar for each statement, so you can see which ones people agree with most.
  • One column per statement. Each statement exports as its own column, the question’s column name plus _1, _2 and on, and in SPSS as an ordinal variable with every point labelled.

Top-2-Box by statement

Report deck slide, from the PDF export · share who agree

A report slide of Top-2-Box by statement: The setup guide was clear 69%, I knew who to ask for help 52%, I was productive in my first week 77%.
Each bar is the share of that statement’s scale answers that were Agree or Strongly agree. Try the battery in the demo above.
See everything a grid can do

Build it

Add a Likert scale question

  1. 1

    Add the question

    In the builder, add a question and choose Rating Scale as its type, with the Numbers style. Write the statement into the question, such as “How much do you agree or disagree? The onboarding process was easy to follow.”

  2. 2

    Pick the number of points

    Click 5-point (or 7-point) and the scale runs from 1 to 5. Set the range before you type labels: each label stays with its point, so a new range that leaves a point out drops its label.

  3. 3

    Label every point

    Open Per-point labels (labeled Likert) and type Strongly disagree, Disagree, Neither agree nor disagree, Agree and Strongly agree, from 1 up. Keep agreement at the high end, so Top-2-Box means agree.

  4. 4

    Add Not applicable and Don’t know

    Tick Offer a “Not applicable” choice and give it a label. Then turn on Non-answer options and keep Don’t know, and Prefer not to answer if you want it. Add N/A in one place only, not as a non-answer as well.

  5. 5

    Choose the order on screen

    To show Strongly agree first, tick Reverse point order on screen (numeric). Only the screen changes: Agree is still stored as 4.

  6. 6

    Set the rules and preview

    In Rules, keep it required, or make it soft required or optional. Preview on desktop and phone: on a narrow screen the points stack as full-width rows with their labels.

The Rating Scale editor for the onboarding statement: the Numbers style, the 5-point preset, Min Value 1 and Max Value 5, the low, middle and high labels left empty, Offer a “Not applicable” choice ticked with “Not applicable”, Non-answer options with Don’t know coded DK, Per-point labels open with Strongly disagree, Disagree, Neither agree nor disagree, Agree and Strongly agree, and the preview of the labelled row with its Not applicable choice
EditorPoints, a label on each, and N/A in one place.
The Non-answer options box ticked, with the hint “Kept out of averages; shown in tables.”, a Don’t know row coded DK and the Add option link
Non-answersCodes you control, kept out of averages.
The builder’s phone preview of the onboarding statement: the five points stacked as full-width rows, each number with its label beside it, then the Not applicable and Don’t know pills
Phone previewStacked rows on a narrow screen.

The data you get

One ordinal column, with every point labelled

The answer exports as the point number in one column. Your labels come along as value labels, and Not applicable and Don’t know get codes of their own that never clash with a point.

respondent_idQ1
47762a85-0001-4000-8000-000000000001DK
47762a85-0002-4000-8000-0000000000023
47762a85-0003-4000-8000-0000000000034
47762a85-0004-4000-8000-0000000000042
47762a85-0005-4000-8000-0000000000052
47762a85-0006-4000-8000-0000000000065
47762a85-0007-4000-8000-0000000000071
47762a85-0008-4000-8000-0000000000082

Trimmed to the question’s column. Points export as 1 to 5, Not applicable as NA and Don’t know as DK. Every export also carries the standard respondent columns (status, timing, weight and more).

One column per question

The column takes your export tag, or the question’s position, Q and its number. Each cell holds the point number, 1 to 5.

Labels as value labels

In SPSS and in the Excel Value Labels sheet, every point carries its label, so 4 reads as Agree. CSV and JSON carry the codes only.

Ordinal in SPSS

The variable comes measured as ordinal, ready for medians and non-parametric tests, with no setting to change first.

Not applicable and Don’t know

CSV writes NA and DK. SPSS codes them 97 and 96, outside the scale, labelled, and declared missing when you ask for it.

A battery: one column per statement

A grid of statements exports one column for each, the question’s column name plus _1, _2 and on, each an ordinal SPSS variable with every point labelled.

Weighting

Weighted exports carry a weight column, and the SPSS syntax sets WEIGHT BY weight. Weighting is on Research Team.

Charts & analysis

Agreement, means and tests from the same question

Every Likert question is charted for you: a bar for each labelled point, the mean, median and Top-2-Box, and a diverging Likert bar on dashboards and in decks. Crosstab it against any other question and test the differences.

The Surveti Results card for the onboarding statement: Mean 3.7, Median 4 and Top-2 box 63% over the 106 answers on the scale, 120 responses, a bar for each of the 5 labelled points from Strongly disagree to Strongly agree in scale order, and Not applicable and Don’t know in a box of their own, counted in the base
ResultsEvery point by its label, in scale order, with N/A and Don’t know shown and counted.

Where it charts

  • ResultsLabelled bars, mean, median, Top-2-Box
  • DashboardA Likert tile, plus Top-2-Box, Mean and Base tiles
  • Report deckA Likert slide in the PDF export
  • CrosstabsIn the rows or the banner
  • Tab booksBoth boxes and the mean; median and SD if added
  • Client portalResults, crosstabs, dashboard, report
A Surveti dashboard: the onboarding statement as a diverging Likert bar, disagreement left of the neutral point and agreement right of it, each point a share of the 106 answers on the scale, a legend of the 5 labels, and Not applicable and Don’t know listed under the bar as counted in the base, above Top-2-Box, Mean and Base tiles
DashboardA Likert tile of the shares of scale answers, with Top-2-Box, Mean and Base tiles.
A report slide from the PDF export: the onboarding statement with n = 120 and a Top-2-Box of 63%, the same diverging bar, each label’s share in the legend, and Not applicable and Don’t know listed under the bar
Report deckThe same Likert bar on a report slide, from the PDF export.

The onboarding process was easy to follow, by ageColumn %, letters at 95% and 99%

Crosstabs · Solo+Letters · Professional+
The Surveti Crosstabs view of the onboarding statement by age group: a row per labelled point, Not applicable and Don’t know rows, then Top-2-Box and Mean rows, in column percentages with the base for each age band, significance letters beside the cells that differ, and the note that the top and bottom box leave Not applicable and non-answers out
Point rows are shares of everyone in the column, Not applicable and Don’t know included. Top-2-Box and the mean use the scale answers only. A letter marks a figure significantly higher than the column with that letter: Top-2-Box and the mean are higher for 35–54 than for 18–34, and higher again for 55+ than for both. Letters need Professional+. The table itself needs Solo+; on Solo it runs without letters.

Analysis that applies

  • Crosstabs

    Solo+

    Put the question in the rows, one row per labelled point, with Top-2-Box and the mean underneath. Not applicable and Don’t know get their own rows and stay out of the mean.

    Crosstabs

  • Cut other questions by agreement

    Solo+

    Use the Likert question as the banner, one column per point, to see how people who agree answer everything else. Group points into your own nets, such as Agree or Strongly agree. A net you add counts everyone who answered, Not applicable and Don’t know included.

    Crosstabs

  • A means test and a proportions test

    Professional+

    The mean row is tested with Welch’s t-test on effective bases, and the point rows and Top-2-Box with a test of proportions, in the same table.

    Significance testing

  • Tab books

    Research Team

    A table per question with the points, Top and Bottom box (from one to three points) and the mean, plus the median and standard deviation if you add them.

  • Weighting

    Research Team

    Weights carry through to the mean and counts in Results, crosstabs, dashboards, decks and exports.

  • Trend the mean

    Track the mean of the same statement across waves.

How Surveti compares

Likert scales across the common survey tools

Likert scale capabilities in Surveti, SurveyMonkey, Qualtrics, Typeform and Alchemer, as of 28 October 2026
CapabilitySurvetiSurveyMonkeyQualtricsTypeformAlchemer
A Likert battery on the free plan (or entry plan)yesRating Scale and grid on every planpartial15, 3Matrix not named for Basic; Multiple Choice worksyes39Matrix table on the free accountyes52Matrix on Freeno76, 63No free plan
Agree/disagree preset scales (5- and 7-point)partialAgreement preset on single choice; typed on a Rating Scale, pasted on a gridpartial21, 7Answer Genius; not on Matrixpartial38Suggested Choices; point counts not statedunclear44, 47No presets describedyes75Rating Agree lists of 3, 5 and 7
Statement battery with one shared scaleyesA grid; a row can have its own scaleyes2Matrix/Rating Scaleyes26Up to 100 statementsyes44Matrixyes54, 56Radio Button Grid
Randomise statementsyesStatements and points, every planpaid2, 12, 14Tier unclearyes26Statement and scale-point randomisationno44Can’t randomise a Matrixyes62Rows and columns
Reverse-coded items (per statement)partialA computed variable; no per-statement switchunclear2, 3, 7Weights per column; no switch documentedpartial32, 40Recode per question, not per statementunclear44, 45Nothing documentedunclear72, 56Reporting values per column
N/A or “Don’t know” kept out of scoresyesOut of the mean and Top-2-Box, on one statement or a batteryyes2, 4N/A column with weights onyes38Exclude From Analysisunclear44, 42N/A not mentioned for Matrixpartial55, 74, 68Out only if its value isn’t numeric
Per-statement results (mean per statement)paidMean per statement in crosstabs · Solo+; Results shows countsyes2Weighted average per statementyes28, 29Statistics Tablepartial46Counts and percentages onlyyes55, 68, 60Row averages need numeric values
Stacked bar chart of the batterynoResults heat-map; Top-2-Box bars in decksyes22Stacked baryes31, 30Stack Chart; pages disagreeno46Grouped bars or a tablepartial60Per row, not the whole grid
Diverging bar chartpartialOne statement, on dashboards and in decks; not a batteryunclear22, 13Not mentionedunclear30, 31Not mentionedunclear46Not mentionedunclear60, 57Not in the chart lists
Top / bottom box (top-2-box)yesResults, dashboards, crosstabs; both boxes in tab booksunclear7, 23No built-in setting foundyes28, 29, 41Statistics Tableunclear46Not mentionedyes57, 58Answer Grouping / Top and Bottom Box
Export of the battery: one column per statementyesA numeric column per statement, every planpaid11, 15, 14Spans columns; export not on Basicyes34, 26, 39Broken across columnsunclear53Layout not statedyes55, 77One column per row
SPSS export of the battery with value labelspaidEvery point labelled, ordinal · Professional+paid10Values follow the weightsunclear34Value labels not describedno53CSV and XLSX onlypaid65, 72, 77Full Access only
Mobile layout of the battery documentedyesList, chips, accordion, carousel or segmented barpartial2May need sideways scrollingyes26Accordionunclear44Not describedyes55One row at a time
As of 28 Oct 2026, from each vendor’s public help pages and pricing.yespartialpaidunclearno

Partial means partly built in or possible only with a documented workaround; paid means built in on paid tiers or licences only; unclear means the vendor’s docs don’t settle it. Google Forms, QuestionPro and SurveySparrow were also reviewed. Google Forms has a linear scale from 0 or 1 up to 10 with a label on each end, and grids with one answer per row. QuestionPro sets a custom or inverse scale per question on its higher licences, and can give any single-choice question top and bottom scores. SurveySparrow’s group rating runs from 3 to 10 points.

Sources (46)
  1. [2] SurveyMonkey: Matrix/Rating Scale. https://help.surveymonkey.com/en/surveymonkey/create/matrix-rating-scale/ · accessed 28 Oct 2026
  2. [3] SurveyMonkey: Likert scales. https://help.surveymonkey.com/en/surveymonkey/create/likert-scales/ · accessed 28 Oct 2026
  3. [4] SurveyMonkey: Not applicable option. https://help.surveymonkey.com/en/surveymonkey/create/not-applicable-option/ · accessed 28 Oct 2026
  4. [7] SurveyMonkey: Likert scale guide. https://www.surveymonkey.com/learn/survey-best-practices/likert-scale/ · accessed 28 Oct 2026
  5. [10] SurveyMonkey: SPSS exports. https://help.surveymonkey.com/en/analyze/spss-exports · accessed 28 Oct 2026
  6. [11] SurveyMonkey: XLS exports. https://help.surveymonkey.com/en/surveymonkey/analyze/xls-exports/ · accessed 28 Oct 2026
  7. [12] SurveyMonkey: Randomizing answers. https://help.surveymonkey.com/en/create/randomizing-answers · accessed 28 Oct 2026
  8. [13] SurveyMonkey: Custom charts. https://help.surveymonkey.com/en/surveymonkey/analyze/custom-charts/ · accessed 28 Oct 2026
  9. [14] SurveyMonkey: Individual pricing. https://www.surveymonkey.com/pricing/individual/ · accessed 28 Oct 2026
  10. [15] SurveyMonkey: Free vs paid plans. https://www.surveymonkey.com/product/free-vs-paid-plans/ · accessed 28 Oct 2026
  11. [21] SurveyMonkey: Answer Genius. https://help.surveymonkey.com/en/surveymonkey/create/answer-genius/ · accessed 28 Oct 2026
  12. [22] SurveyMonkey: Chart types. https://help.surveymonkey.com/en/surveymonkey/analyze/chart-types/ · accessed 28 Oct 2026
  13. [23] SurveyMonkey: Combining answer choices. https://help.surveymonkey.com/en/surveymonkey/analyze/combining-hiding-answer-choices/ · accessed 28 Oct 2026
  14. [26] Qualtrics: Matrix table. https://www.qualtrics.com/support/survey-platform/survey-module/editing-questions/question-types-guide/standard-content/matrix-table/ · accessed 28 Oct 2026
  15. [28] Qualtrics: Results statistics table. https://www.qualtrics.com/support/survey-platform/reports-module/results-section/visualizations/tables/statistics-table/ · accessed 28 Oct 2026
  16. [29] Qualtrics: Reports statistics table. https://www.qualtrics.com/support/survey-platform/reports-module/reports-section/reports-visualizations/table-visualizations/statistics-table-visualization/ · accessed 28 Oct 2026
  17. [30] Qualtrics: Results bar chart. https://www.qualtrics.com/support/survey-platform/reports-module/results-section/visualizations/charts/bar-chart/ · accessed 28 Oct 2026
  18. [31] Qualtrics: Reports bar chart. https://www.qualtrics.com/support/survey-platform/reports-module/reports-section/reports-visualizations/chart-visualizations/bar-chart-visualization/ · accessed 28 Oct 2026
  19. [32] Qualtrics: Recode values. https://www.qualtrics.com/support/survey-platform/survey-module/question-options/recode-values/ · accessed 28 Oct 2026
  20. [34] Qualtrics: Export formats. https://www.qualtrics.com/support/survey-platform/data-and-analysis-module/data/download-data/export-formats/ · accessed 28 Oct 2026
  21. [38] Qualtrics: Formatting answer choices. https://www.qualtrics.com/support/survey-platform/survey-module/editing-questions/formatting-answer-choices/ · accessed 28 Oct 2026
  22. [39] Qualtrics: Free accounts. https://www.qualtrics.com/support/survey-platform/managing-your-account/trial-accounts/ · accessed 28 Oct 2026
  23. [40] Qualtrics Community: matrix reverse coding (search summary). https://community.qualtrics.com/survey-platform-54/matrix-tables-reverse-coding-27259 · accessed 28 Oct 2026
  24. [41] Qualtrics Community: top/bottom box for a Likert matrix (search summary). https://community.qualtrics.com/survey-platform-54/grouping-answer-choices-from-a-likert-matrix-in-a-qualtrics-report-33002 · accessed 28 Oct 2026
  25. [42] Typeform: Opinion Scale question. https://help.typeform.com/hc/en-us/articles/360052363932-Opinion-Scale-question · accessed 28 Oct 2026
  26. [44] Typeform: Matrix question. https://help.typeform.com/hc/en-us/articles/360054731432-Matrix-question · accessed 28 Oct 2026
  27. [45] Typeform: Questions that don’t work with logic. https://help.typeform.com/hc/en-us/articles/5821097529620-Why-doesn-t-my-question-work-with-logic · accessed 28 Oct 2026
  28. [46] Typeform: Results summary. https://help.typeform.com/hc/en-us/articles/360029577071-Your-Results-summary · accessed 28 Oct 2026
  29. [47] Typeform: Question types. https://help.typeform.com/hc/en-us/articles/360051789692-Question-types · accessed 28 Oct 2026
  30. [52] Typeform: Free plan. https://help.typeform.com/hc/en-us/articles/360032972852-Free-plan · accessed 28 Oct 2026
  31. [53] Typeform: Export responses. https://help.typeform.com/hc/en-us/articles/360029253572 · accessed 28 Oct 2026
  32. [54] Alchemer: Rating (Likert Scale). https://help.alchemer.com/help/rating · accessed 28 Oct 2026
  33. [55] Alchemer: Radio Button Grid. https://help.alchemer.com/help/radio-buttons-grid · accessed 28 Oct 2026
  34. [56] Alchemer: Likert via a Radio Button Grid. https://help.alchemer.com/help/best-practices-use-a-radio-button-grid-to-combine-rating-statements · accessed 28 Oct 2026
  35. [57] Alchemer: Top box reporting. https://help.alchemer.com/help/top-box-or-top-2-box-reporting · accessed 28 Oct 2026
  36. [58] Alchemer: Answer grouping. https://help.alchemer.com/help/answer-grouping-recoding · accessed 28 Oct 2026
  37. [60] Alchemer: How each question reports. https://help.alchemer.com/help/how-does-each-question-report · accessed 28 Oct 2026
  38. [62] Alchemer: Randomization. https://help.alchemer.com/help/randomization · accessed 28 Oct 2026
  39. [63] Alchemer: Question type guide. https://help.alchemer.com/help/question-type-guide · accessed 28 Oct 2026
  40. [65] Alchemer: SPSS exports. https://help.alchemer.com/help/spss · accessed 28 Oct 2026
  41. [68] Alchemer: Standard Report statistics. https://help.alchemer.com/help/standard-report-available-statistics · accessed 28 Oct 2026
  42. [72] Alchemer: Reporting values. https://help.alchemer.com/help/reporting-values · accessed 28 Oct 2026
  43. [74] Alchemer: Answer option special settings. https://help.alchemer.com/help/answer-option-special-settings · accessed 28 Oct 2026
  44. [75] Alchemer: Library of common answer options. https://help.alchemer.com/help/using-the-library-of-common-answer-options · accessed 28 Oct 2026
  45. [76] Alchemer: Plans and pricing. https://www.alchemer.com/plans/ · accessed 28 Oct 2026
  46. [77] Alchemer: Compare licenses (license grid). https://www.alchemer.com/plans-pricing/compare-licenses/ · accessed 28 Oct 2026

What’s different here

N/A and Don’t know, handled the same everywhere

On a Likert scale question they count in the base, show as their own rows and stay out of the mean and Top-2-Box in Results, crosstabs and tab books, and out of the median wherever it’s shown, whatever their code. In SPSS each gets its own labelled code outside the scale.

A means test in the crosstab

The mean row gets Welch’s t-test and the point rows a test of proportions, side by side. Crosstabs are on Solo+, letters on Professional+.

Ordinal in SPSS, labels included

A Likert scale exports as an ordinal variable with every point labelled, and a battery as one ordinal variable per statement. SPSS export is on Professional+.

Logic on a range

Show or skip on “is between”, as well as at least and at most, so “Agree or above” or “between Disagree and Agree” is one condition. Question-level logic is on every plan; the survey-level skip and branching panel is on Professional+.

A diverging chart, ready made

Dashboards and report decks draw a single Likert statement as a diverging bar, disagreement to the left of neutral and agreement to the right, with its Top-2-Box.

Mean, median and spread together

Results shows the mean, median and Top-2-Box beside every labelled point, and tab books add the standard deviation and Bottom-2-Box.

When to use it

Use a Likert scale to measure agreement with a statement

When to use it

  • You’re measuring agreement, attitude or opinion about a clear statement.
  • You want a balanced scale people understand at a glance.
  • You’re comparing several statements on the same scale. Ask them as a grid.
  • You need a standard format to compare against earlier waves or norms.

When to reach for something else

  • You’re asking how satisfied, how likely or how good, not how much they agree. Use a rating scale.
  • You want the likelihood to recommend on 0 to 10. Use NPS.
  • The statement holds two ideas, like “fast and friendly”. Split it into two statements.
  • You’re asking for a category or a choice. Use single choice.

Sample wording

“How much do you agree or disagree? The onboarding process was easy to follow.” Strongly disagree · Disagree · Neither agree nor disagree · Agree · Strongly agree · Not applicable · Don’t know

Best practices

  • Keep the scale symmetric: as many agree points as disagree points around the middle.
  • Label every point, so the steps read as evenly spaced.
  • Put agreement at the high end, so Top-2-Box and the chart read agreement as positive.
  • Offer Not applicable when the statement may not apply, and Don’t know when people may have no view, so neither is forced onto the middle point.
  • Write one idea per statement.
  • Mix positive and negative statements to catch agreeing by habit, and plan how you’ll score the negative ones before you launch.

Accessibility

Usable without the mouse or the screen

Labels read aloud

Each point is announced with its number and label, such as “4, Agree”, and whether it’s chosen. Not applicable is announced as one more choice in the same group.

One Tab stop, arrow keys

Tab lands on the scale once. The arrow keys move along the points and Not applicable, choosing as they go, and wrap at the ends.

Room to tap

Every point is at least 44px tall. On a narrow screen a labelled scale stacks into full-width rows, so labels never shrink to fit.

Don’t know, out of the way

Pressing Don’t know dims the scale and takes it out of the tab order, so the answer can’t be half one and half the other.

Use it in

Templates and solutions that use Likert scales

Template

Employee Engagement Survey

“I feel valued and appreciated at work.” and six more statements, rated 1 to 5.

Template

Course Evaluation

“The course content was relevant and well-organized.” and four more statements, rated 1 to 5.

Solution

Academic research

Attitude scales with SPSS-ready ordinal data.

Solution

Customer feedback

Track agreement with what customers say about you.

Frequently asked questions

What’s the difference between a Likert scale and a rating scale?

A Likert scale asks how much someone agrees with a statement, on a balanced scale from disagree to agree. A rating scale measures an amount, such as satisfaction, quality or likelihood. In Surveti both are a Rating Scale question: label every point with agree wording and you have a Likert scale. For a 0 to 10 likelihood to recommend, use NPS.

How many points should a Likert scale have?

Five and seven are the most common. Five is quick and easy to label; seven gives finer steps. Keep it symmetric. The builder has presets for 3, 5, 7, 10 and 11 points.

How do I add Not applicable and Don’t know?

Tick Offer a Not applicable choice, and turn on Non-answer options for Don’t know or Prefer not to answer. Both count in the base and show as their own rows, and neither is included in the mean, median or Top-2-Box. Add N/A in one of the two places, not both.

What is Top-2-Box?

The share of answers in the top two points: Agree and Strongly agree on a 5-point agree scale. Surveti works it out from the answers on the scale, so in the example above 63% of the 106 people who gave a point agreed. Results shows it beside the mean and median, and crosstabs test it between groups.

Should I mix positive and negative statements?

Often, yes: it helps catch people who agree by habit. Surveti has no reverse-score switch, so a negative statement’s Top-2-Box still counts agreement. To score it the other way, add a computed variable such as 6 minus the answer.

What does Likert data look like in an export?

One column holding the point number, 1 to 5, with NA and DK for Not applicable and Don’t know. SPSS (Professional+) makes it an ordinal variable with every point labelled and codes 97 and 96 for N/A and Don’t know. A battery exports one column per statement.

Can I build a Likert scale as a single choice or a grid?

Yes. A grid is the way to ask several statements on one scale. A single choice with numbered agree answers also gets a mean and Top-2-Box. For one statement, the Rating Scale gives you the stacked phone layout and a label on every point.

How do I compare agreement across groups?

Use Crosstabs (Solo+) to put the question against any banner, such as age or region. On Professional+, letters mark differences in the mean, Top-2-Box and each point. On Research Team, weights carry into the same tables.

Add a likert scale question in minutes.

Start free — no credit card. 50 question types, logic, and analysis in one place.