How to Interpret Likert Scale Mean Scores in Your Project: 4-Point and 5-Point Decision Rules

Quick Summary

Likert-type questions are commonly used in final-year project questionnaires to measure respondents' level of agreement, satisfaction, perception, frequency or opinion.

For a common four-point agreement scale:

The arithmetic midpoint of the assigned scores is:

(4 + 3 + 2 + 1) ÷ 4 = 2.50

This is why many undergraduate projects using this coding system adopt 2.50 as a criterion mean.

Under such a decision rule:

• Mean of 2.50 or above may be interpreted as agreement.
• Mean below 2.50 may be interpreted as disagreement.

For a five-point scale coded from 1 to 5, the arithmetic midpoint is 3.00.

These are not universal rules for every questionnaire.

Your scale, coding system, research methodology and supervisor's approved decision rule should determine how the results are interpreted.

What Is a Likert Scale?

A Likert scale uses ordered response categories to measure attitudes, perceptions or opinions.

For example, your questionnaire may contain the statement:

Social media helps students access academic information.

A respondent might select:

• Strongly Agree
• Agree
• Disagree
• Strongly Disagree

These categories have a clear order.

Strongly Agree represents stronger agreement than Agree, which represents stronger agreement than Disagree.

However, assigning numerical values such as 4, 3, 2 and 1 does not automatically prove that the psychological distance between every response category is exactly equal.

This is one reason the method used to analyse Likert responses should be stated clearly in your research methodology.

Likert Item vs Likert Scale

This distinction is important.

A Likert item is one individual statement.

For example:

Online learning improves my academic productivity.

A Likert scale usually combines several related items intended to measure a broader construct.

For example, five statements might collectively measure:

Students' attitudes towards online learning.

Individual Likert responses are ordinal categories.

When several related items are combined into a score, researchers may use additional statistical methods depending on the design of the instrument and the analysis being conducted.

For an undergraduate project, follow the approach approved in your Chapter Three rather than selecting a method only because another project used it.

4-Point and 5-Point Likert Scales

Two common questionnaire formats are four-point and five-point scales.

4-Point Likert Scale

A common four-point agreement scale is:

This format does not include a neutral response.

Respondents are therefore asked to lean towards either agreement or disagreement.

5-Point Likert Scale

A common five-point agreement scale is:

This version provides a midpoint or neutral category.

Do not switch between a four-point and five-point scale after you have already collected your data.

The response scale and coding should be defined before analysis and described in Chapter Three.

How to Calculate a Likert Mean Score

Where your approved methodology uses mean scores, a common calculation is:

Mean = Σfx ÷ N

Where:

• Σ means sum
• f means frequency
• x means the numerical score assigned to the response
• N means total number of valid responses

The calculation gives a weighted mean for the questionnaire item.

Example: Calculating a 4-Point Likert Mean

Suppose 100 respondents answered:

Students use social media to access academic information.

The responses were:

The weighted total is:

100 + 135 + 40 + 10 = 285

Then:

Mean = 285 ÷ 100

Therefore:

Mean = 2.85

If your Chapter Three methodology established 2.50 as the criterion mean, then:

2.85 ≥ 2.50

Under that decision rule, the item would be interpreted as agreement.

You could report:

Respondents agreed that social media helps students access academic information, with a mean score of 2.85.

Why Is the Criterion Mean for a 4-Point Scale Often 2.50?

For a scale coded:

• Strongly Agree = 4
• Agree = 3
• Disagree = 2
• Strongly Disagree = 1

the arithmetic midpoint of the possible scores is:

(4 + 3 + 2 + 1) ÷ 4

= 10 ÷ 4

= 2.50

That is where the commonly used 2.50 criterion mean comes from.

It is not a value automatically produced by SPSS, nor is it a universal statistical rule.

It is an interpretation threshold adopted by some researchers using this particular coding system.

If your methodology adopts this approach, a simple decision rule may be:

State the rule in Chapter Three before applying it to results in Chapter Four.

Is 2.50 Always the Decision Point for a 4-Point Likert Scale?

No.

It depends on how your questionnaire is coded and how your methodology defines the interpretation.

For example, if you code:

• Strongly Agree = 1
• Agree = 2
• Disagree = 3
• Strongly Disagree = 4

the numerical direction has been reversed.

Although the arithmetic midpoint is still 2.50, a lower mean now represents stronger agreement.

This demonstrates why you should never copy a decision rule without checking the actual coding system.

Your Chapter Three should clearly state:

• The response categories
• Numerical codes
• Method of calculating scores
• Criterion or decision rule
• Meaning assigned to results above or below that point

The 4-Point Likert Scale Decision Rule

If your questionnaire uses:

and your approved methodology adopts the midpoint as the criterion mean, then:

Criterion mean = 2.50

A possible decision rule is:

However, the exact wording should match what the questionnaire measures.

For example:

Agreement

• Agree
• Disagree

Availability

• Available
• Not Available

Effectiveness

• Effective
• Not Effective

Extent

• High Extent
• Low Extent

Perception

• Positive perception
• Negative perception

Do not label every questionnaire item as Accepted or Rejected when those words do not fit the construct being measured.

What Is the Criterion Mean for a 5-Point Likert Scale?

For a five-point scale coded:

• Strongly Agree = 5
• Agree = 4
• Neutral = 3
• Disagree = 2
• Strongly Disagree = 1

the arithmetic midpoint is:

(5 + 4 + 3 + 2 + 1) ÷ 5

= 15 ÷ 5

= 3.00

This explains why 3.00 is often used as a reference point in a five-point scale.

However, there is an important interpretation issue.

If 3 = Neutral, then a mean of exactly 3.00 lies at the midpoint.

It should not automatically be described as agreement.

5-Point Likert Scale Decision Rule

One possible interpretation approach is:

But this is only one possible framework.

Some projects instead create intervals for all five categories.

For example, a methodology might classify scores into ranges corresponding to:

• Strongly Disagree
• Disagree
• Neutral
• Agree
• Strongly Agree

If your department requires this approach, use the ranges established in your approved methodology.

Do not mix two different interpretation methods in the same project.

Should 3.00 Count as Agreement on a 5-Point Scale?

Not automatically.

If the response coded 3 represents Neutral, then the midpoint of 3.00 should normally be recognised as the centre of the scale under that coding system.

Therefore, writing:

3.00 and above = Agree

would classify the exact neutral midpoint as agreement.

A clearer rule, where your methodology permits it, is:

• Above 3.00 = tendency towards agreement
• 3.00 = midpoint
• Below 3.00 = tendency towards disagreement

Your supervisor may adopt a different decision framework, so follow the approved method for your study.

How to Interpret a Likert Mean in Chapter Four

Suppose your result is:

Mean = 3.18

on a four-point scale with a criterion mean of 2.50.

You could write:

The respondents agreed that online learning improves access to educational resources, with a mean score of 3.18, which was above the criterion mean of 2.50.

Suppose another item has:

Mean = 2.21

You might write:

Respondents disagreed that unreliable internet access was no longer a challenge, as the item recorded a mean score of 2.21, below the criterion mean of 2.50.

The interpretation should reflect the actual wording of the statement.

Example of a Likert Results Table for Chapter Four

Table 4.3: Respondents' Views on Social Media Use for Academic Purposes

You could interpret the table as:

Table 4.3 shows that respondents agreed that social media helps students access academic information, with a mean score of 2.85. However, respondents disagreed that social media always improves concentration while studying, with a mean score of 2.05.

You do not need to repeat every number in the table.

Explain the findings that answer the research question.

Should You Include Standard Deviation?

Where your methodology requires it, standard deviation can provide useful information alongside the mean.

The mean describes the central score.

The standard deviation describes how spread out the responses are around that mean.

For example, two questionnaire items could both have:

Mean = 3.00

while having very different response patterns.

One item might have most respondents selecting 3.

Another might have half selecting 1 and half selecting 5.

The same mean does not necessarily mean the responses were distributed in the same way.

This is why your results may be more informative when the mean is considered alongside:

• Frequencies
• Percentages
• Standard deviation
• The wording of the item

where those statistics are appropriate for your methodology.

Be Careful With Negatively Worded Statements

Consider:

Social media distracts students from their academic work.

Suppose respondents strongly agree with this statement.

The item will have a high mean if your coding is:

4 = Strongly Agree

But the high score represents agreement with a negative statement.

You should therefore report what respondents actually agreed with rather than automatically describing every high mean as a positive outcome.

For example:

Respondents agreed that social media distracts students from academic work.

That is clearer than:

The item recorded a positive result.

When Should You Reverse-Code a Likert Item?

Reverse coding becomes important when positive and negative items are combined into one overall score.

Suppose your questionnaire contains:

Online learning improves my productivity.

and:

Online learning makes it difficult for me to complete academic work.

If both questions contribute to a combined score representing positive attitude towards online learning, they point in opposite directions.

For a four-point scale, reverse coding may be:

After reverse coding, higher values consistently represent the same direction.

Do not reverse-code an item simply because respondents gave an unfavourable answer.

Reverse coding should follow the design of the scale.

Do You Need to Reverse-Code When Analysing Items Separately?

Not always.

If you are simply presenting each questionnaire statement separately and explaining what respondents said, you can interpret the negatively worded item according to its actual wording.

Reverse coding is especially relevant when several positive and negative items are being combined into:

• A total score
• An average score
• A scale score
• A reliability analysis

Know why you are reverse-coding before changing the values.

Can You Calculate One Overall Mean for Several Likert Items?

Sometimes, but not automatically.

Suppose five questionnaire items were specifically designed to measure:

Students' perceived usefulness of online learning.

If those items legitimately form one scale, your approved methodology may allow you to calculate a composite score, such as a sum or average.

But do not calculate a grand mean across unrelated questions merely because they all use the same response options.

Before combining items, consider:

• Whether they measure the same construct
• How the questionnaire was designed
• Whether any items require reverse coding
• Reliability of the scale where appropriate
• Your approved methodology

If your questionnaire contains several scales, analyse them according to their intended structure.

For related guidance, see Cronbach's Alpha for Your Questionnaire: Acceptable Value and How to Report It in Chapter Three.

Mean vs Frequency and Percentage

A mean gives you one summary value.

Frequency and percentage show how respondents were distributed across the available choices.

Suppose an item has:

Mean = 2.50

That single value does not show whether:

• Nearly everyone selected the middle categories, or
• Respondents were strongly divided between high and low responses.

Therefore, do not treat the mean as the entire finding.

Where appropriate, consider it alongside frequency and percentage information.

IBM SPSS treats ordinal variables as categories with an inherent order but without an automatically measurable distance between categories. This is why the type of analysis should be chosen deliberately rather than assuming every numbered response behaves like a continuous measurement.

A Chapter Three Paragraph You Can Adapt

If your actual methodology uses a four-point agreement scale and a criterion mean of 2.50, you could write:

Questionnaire items were rated using a four-point response scale consisting of Strongly Agree, Agree, Disagree and Strongly Disagree, assigned scores of 4, 3, 2 and 1 respectively. Where mean scores were used for descriptive interpretation, a criterion mean of 2.50, representing the arithmetic midpoint of the assigned scores, was adopted. Items with mean scores of 2.50 or above were interpreted as agreement, while items with mean scores below 2.50 were interpreted as disagreement.

Only use this paragraph if it describes the method actually used in your study.

For a questionnaire measuring something other than agreement, change the interpretation terms accordingly.

Example for a 5-Point Scale

If your methodology uses:

• Strongly Agree = 5
• Agree = 4
• Neutral = 3
• Disagree = 2
• Strongly Disagree = 1

you might state:

Questionnaire responses were measured using a five-point scale ranging from Strongly Agree (5) to Strongly Disagree (1). The arithmetic midpoint of the scale was 3.00. Mean scores above the midpoint were interpreted as tending towards agreement, scores below the midpoint as tending towards disagreement, while the midpoint represented the centre of the response scale.

Again, use only the decision rule approved for your study.

How to Calculate Likert Mean in SPSS

If your questionnaire responses are already entered into SPSS, the software can calculate descriptive statistics.

For individual items, make sure:

• Each variable is coded correctly.
• Missing responses are handled appropriately.
• Reverse-coded variables have been created where necessary.
• You understand what each variable represents.

Your research methodology should determine whether means, medians, frequencies or another summary approach is appropriate.

For a complete workflow, read How to Analyse Questionnaire Data in SPSS for Chapter Four.

Common Mistakes to Avoid

Treating 2.50 as a Universal Rule

The 2.50 value comes from the midpoint of a particular four-point coding system.

It is not automatically the correct decision rule for every questionnaire.

Treating 3.00 as Agreement on a 5-Point Scale

If 3 = Neutral, the exact midpoint should not automatically be interpreted as agreement.

Forgetting Which Direction the Scale Was Coded

If higher numbers represent disagreement rather than agreement, the interpretation changes.

Always check your coding.

Mixing 4-Point and 5-Point Decision Rules

A 2.50 midpoint belongs to a 1-to-4 coding scheme.

A 3.00 midpoint belongs to a 1-to-5 coding scheme.

Do not use the wrong criterion.

Reporting Only the Mean

The mean does not show the full distribution of responses.

Where appropriate, include frequencies, percentages and measures of variability.

Combining Unrelated Items Into a Grand Mean

Questionnaire items should not be combined simply because they all use the same Likert response options.

Forgetting Negative Statements

A high mean on a negatively worded item does not automatically represent a desirable outcome.

Read the wording carefully.

Reverse-Coding Without a Reason

Do not reverse responses just because you prefer a different finding.

Reverse coding is a scoring procedure, not a way to change results.

Changing Results to Match Expectations

Report the actual responses.

A finding that contradicts your expectation can still be an important research result.

Calculate Your Likert Mean Scores

If your approved methodology requires mean scores, you can use MonoEd's free Likert Scale Mean Calculator.

Enter your actual response frequencies to calculate:

• Weighted mean
• Standard deviation
• Decision based on your selected rule
• A results table for review

Check the output against your questionnaire and your Chapter Three methodology before including it in your project.

The calculator performs the calculation. It does not decide which analysis method your research should use.

Frequently Asked Questions (FAQs)

What is the criterion mean for a 4-point Likert scale?

If the response values are coded 4, 3, 2 and 1, the arithmetic midpoint is:

(4 + 3 + 2 + 1) ÷ 4 = 2.50

Many undergraduate projects therefore use 2.50 as a criterion mean.

It is an adopted decision rule, not a universal requirement.

What is the decision rule for a 4-point Likert scale?

For a scale coded from 1 to 4 where higher values represent greater agreement, a project may adopt:

• 2.50 and above = Agree
• Below 2.50 = Disagree

Use this only when it matches your approved methodology.

What is the criterion mean for a 5-point Likert scale?

For response values coded from 1 to 5, the arithmetic midpoint is:

3.00

If 3 represents Neutral, the exact midpoint should normally be recognised as the centre of the scale rather than automatically classified as agreement.

What is the decision rule for a 5-point Likert scale?

One possible rule is:

• Above 3.00 = tendency towards agreement
• 3.00 = midpoint or neutral
• Below 3.00 = tendency towards disagreement

Your project may use a different approved interpretation framework.

Is 2.50 accepted or rejected on a 4-point Likert scale?

If your methodology explicitly states:

2.50 and above = Agree

then an item with a mean of exactly 2.50 falls into the agreement category under that rule.

This is a methodological convention established by the researcher, not a universal property of Likert data.

How do I calculate the mean of Likert responses?

Multiply each response score by the number of respondents who selected it, add the weighted scores and divide by the total number of valid responses.

For example:

Mean = Σfx ÷ N

Should I use mean or percentage for Likert-scale questions?

They provide different information.

Percentages show how respondents are distributed across response categories.

A mean provides a single numerical summary where your methodology treats the coded responses in a way that permits mean-score interpretation.

Your research design should determine which summaries you report.

Can I calculate an overall mean for my questionnaire?

Only when the items being combined legitimately form a scale or construct and your methodology supports combining them.

Do not average unrelated questionnaire questions simply to produce one overall number.

What does a high Likert mean score mean?

It depends on the coding and wording.

If higher values represent stronger agreement, a higher mean indicates responses tending towards the higher agreement categories.

However, if the statement itself is negative, high agreement may represent an unfavourable finding.

Should I reverse-code negative Likert questions?

Reverse coding may be required when positive and negative items are combined into a common scale and higher scores need to represent the same direction.

It may not be necessary when each item is simply being interpreted separately.

Can I use Likert mean scores in SPSS?

SPSS can calculate means from numeric variables.

However, individual Likert responses are ordinal categories, so whether mean-based interpretation is appropriate should follow your research methodology rather than being determined solely by what the software can calculate.

Final Checklist

Before interpreting your Likert results, confirm that:

• You know whether the questionnaire uses a 4-point or 5-point scale.
• The numerical coding is correct.
• Your decision rule is stated in Chapter Three.
• You know why 2.50 or 3.00 was selected.
• Negatively worded items are interpreted correctly.
• Reverse-coded items were handled correctly where necessary.
• Items are combined only when there is a methodological reason.
• Your interpretation matches the wording of each statement.
• You have not changed responses to obtain a preferred result.
• Your Chapter Four results use the same method described in Chapter Three.

The most important thing is consistency.

Choose and justify your analysis method before interpreting the results, then apply it consistently to the actual responses you collected.