How to Analyse Questionnaire Data in SPSS for Chapter Four: Step-by-Step Guide

Quick Summary

SPSS can help you organise, analyse and present questionnaire responses for Chapter Four of your final-year project.

A typical workflow is:

1. Review your research questions and approved analysis method.
2. Create a variable for each questionnaire item.
3. Code responses consistently.
4. Enter or import one respondent per row.
5. Check the data for errors and missing responses.
6. Run frequencies and percentages for demographic or categorical variables.
7. Calculate mean and standard deviation for Likert-scale items where appropriate.
8. Create combined scores only when the questionnaire and research method justify doing so.
9. Test hypotheses using the statistical test specified by your research design.
10. Present clean tables and interpret the results according to your research questions.

Do not start by clicking random SPSS options.

Start with this question:

What analysis does each research question or hypothesis actually require?

Your Chapter Four analysis should be consistent with the method stated in Chapter Three.

What You Need Before Analysing Questionnaire Data in SPSS

Before opening SPSS, prepare:

• Your final questionnaire
• Completed questionnaires or exported online responses
• Your research questions
• Your hypotheses, if applicable
• Your approved Chapter Three methodology
• Your measurement scales
• A coding plan or codebook

Your codebook explains how each questionnaire response will appear in SPSS.

For example:

If you choose this coding system, use it consistently throughout the dataset.

Do not code some questions as 4 = Strongly Agree and others as 1 = Strongly Agree unless the scale was intentionally designed that way and you account for it during analysis.

Understand the SPSS Layout

For basic questionnaire analysis, two parts of the SPSS Data Editor are especially important.

Variable View

Variable View describes the variables in your dataset.

You can define information such as:

• Variable name
• Variable type
• Variable label
• Value labels
• Missing values
• Measurement level

For example, instead of using a long variable name such as:

Social media helps students access academic information

you might use:

q1

and place the full questionnaire statement in the Label field.

Data View

Data View contains the actual responses.

For a typical questionnaire dataset:

• One row represents one respondent.
• One column represents one variable or questionnaire item.

For example:

The numbers only make sense when your coding system is clearly defined.

Step 1: Create Your Variables in SPSS

Open SPSS and select Variable View.

Create one variable for each piece of information you intend to analyse.

For example:

Use short and understandable variable names.

Examples include:

• age
• gender
• level
• dept
• q1
• q2
• q3

Then use the Label field to provide the full meaning of the variable.

This makes your SPSS output easier to understand later.

Step 2: Add Value Labels

If a variable uses coded categories, define what each code means.

For example, suppose you use:

• 1 = Male
• 2 = Female

In Variable View:

1. Find the variable.
2. Click its cell under Values.
3. Open the Value Labels window.
4. Enter 1 as the value and Male as the label.
5. Click Add.
6. Enter 2 and Female.
7. Click Add.
8. Click OK.

For a four-point Likert scale, you may define:

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

IBM SPSS allows numeric values to be associated with descriptive value labels, making coded questionnaire data easier to read.

Step 3: Set the Measurement Level

SPSS allows variables to be marked as nominal, ordinal or scale.

A simple guide is:

Do not choose a statistical test simply because SPSS labels a variable as nominal, ordinal or scale.

The appropriate analysis depends on:

• Your research question
• How the variable was measured
• Your research design
• Statistical assumptions
• Your approved methodology

Follow your supervisor's approved analysis plan where one has already been established.

Step 4: Enter Your Questionnaire Responses

After defining your variables, switch to Data View.

Enter each respondent's answers using the codes in your codebook.

For example, suppose Respondent 1 selected:

• Female
• 400 level
• Strongly Agree for Question 1
• Agree for Question 2

Using the example coding system, the row might contain:

Do not repeatedly type the full words Strongly Agree if your variable has already been coded numerically.

Keep the original questionnaires or exported responses safely so that you can check suspicious entries later.

Can You Import Questionnaire Responses From Excel or Google Forms?

Yes.

If your questionnaire was completed online, you do not necessarily have to type every response manually.

For Google Forms, you can export the responses to a spreadsheet and prepare the file for SPSS.

Before importing, check that:

• Each row represents one respondent.
• Each column represents one variable.
• Column names are clear.
• Blank responses are identified.
• Response categories are consistent.
• Multiple-choice responses are coded correctly.
• No unnecessary spreadsheet headings are included.

After importing the data into SPSS, verify it before beginning your analysis.

Do not assume that an imported dataset is automatically clean.

Step 5: Check and Clean Your Data

Do not start analysing immediately after entering or importing your responses.

First check the data.

Look for:

• Missing responses
• Duplicate respondents
• Impossible values
• Incorrect coding
• Responses in the wrong columns
• Ineligible participants
• Accidental text entries in numeric variables
• Unusual values that may be data-entry errors

For example, if your Likert scale only allows:

1, 2, 3, 4

and you find:

6

that entry needs investigation.

Check the original response before making a correction.

Do not guess what the respondent intended.

Check Frequencies to Find Coding Errors

One useful way to spot problems is to run a frequency table before conducting the main analysis.

For example, if gender should contain only 1 and 2, a frequency table can reveal an accidental value such as 12.

You can then return to the original questionnaire and verify the correct response.

Step 6: Handle Missing Responses Carefully

A respondent may leave one or more questions unanswered.

A missing response is not automatically zero.

Do not replace a missing answer with:

• 0
• The mean
• Your preferred response
• Another respondent's answer

unless your approved methodology specifically requires an appropriate missing-data procedure.

SPSS can distinguish missing data from valid responses.

IBM's documentation notes that user-defined missing values can be specified for variables and are excluded from many calculations.

If you are unsure how missing responses should be handled in your study, discuss them with your supervisor before analysis.

Step 7: Run Frequencies and Percentages

Frequencies and percentages are commonly useful for describing categorical questionnaire variables.

Examples include:

• Gender
• Department
• Academic level
• Age group
• Marital status
• Employment category
• Years-of-experience category

In many SPSS versions, you can obtain frequency tables through:

Analyze → Descriptive Statistics → Frequencies

Move the variables you want to analyse into the variable box and run the analysis.

SPSS can then show the number and percentage of respondents in each category.

Example of a Frequency and Percentage Table

Suppose 100 people completed your questionnaire.

Your Chapter Four table might look like this:

Table 4.1: Distribution of Respondents by Gender

A simple interpretation would be:

Table 4.1 shows that 58 respondents, representing 58.0%, were female, while 42 respondents, representing 42.0%, were male.

That is usually enough for a simple demographic table.

Do not turn a straightforward frequency table into a long paragraph that merely repeats every number.

Step 8: Analyse Likert-Scale Questionnaire Items

Many Nigerian undergraduate projects use questionnaires containing statements such as:

Social media improves students' access to academic information.

Respondents may choose:

• Strongly Agree
• Agree
• Disagree
• Strongly Disagree

One common coding system is:

Depending on your approved methodology, you may report:

• Frequency
• Percentage
• Mean
• Standard deviation

How to Calculate Mean and Standard Deviation in SPSS

In many SPSS versions:

Analyze → Descriptive Statistics → Descriptives

Select the questionnaire items you want to analyse.

Under the available statistics or options, you can request statistics such as:

• Mean
• Standard deviation
• Minimum
• Maximum

Then run the analysis.

Do not calculate a mean simply because another student's project did so.

Use the method stated in your research methodology.

What Does a Mean Score of 2.50 Mean?

For a four-point scale coded:

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

the arithmetic midpoint of the scale is:

(4 + 3 + 2 + 1) ÷ 4 = 2.50

This is why some projects use 2.50 as a criterion or decision mean.

For example, an approved decision rule might classify:

• Mean of 2.50 or above as agreement
• Mean below 2.50 as disagreement

However, 2.50 is not a universal rule for every questionnaire.

Your decision point depends on:

• Your response scale
• How the responses were coded
• Your research method
• How your supervisor expects the items to be interpreted

Do not automatically copy 2.50 into Chapter Three or Chapter Four just because another project used it.

Read our guide on how to interpret Likert scale mean scores before applying a decision rule.

Step 9: Reverse-Code Negative Questionnaire Items When Necessary

Suppose a scale contains these two statements:

Social media improves my academic productivity.

and:

Social media prevents me from concentrating on my academic work.

The second statement points in the opposite direction.

If several items are being combined into a single score where a larger number must consistently represent the same underlying direction, negatively worded items may need reverse coding.

For a four-point scale:

In SPSS, one common approach is:

Transform → Recode into Different Variables

Then:

1. Select the item.
2. Create a new variable, such as q2r.
3. Open Old and New Values.
4. Recode 4 as 1.
5. Recode 3 as 2.
6. Recode 2 as 3.
7. Recode 1 as 4.
8. Save the new variable.

Using a new variable preserves the original response and makes your analysis easier to audit.

Do not reverse-code a statement merely because respondents gave an answer you do not like.

Reverse coding should follow the design and scoring rules of the scale.

Step 10: Decide Whether Questionnaire Items Should Be Combined

This is an important step that students sometimes miss.

Suppose Questions 1 to 5 are intended to measure:

Students' perceived usefulness of social media for learning.

Your methodology may require those items to be combined into a single scale or score.

But you should not simply average unrelated questions because they all use the same Likert response options.

Items should only be combined when there is a reasonable methodological basis for treating them as measurements of the same construct.

That basis may come from:

• An established questionnaire
• Your questionnaire design
• Previous research
• Your approved methodology
• Reliability or scale analysis where appropriate

If your supervisor has approved analysing every questionnaire statement separately, follow that method instead.

Step 11: Check Reliability Where Appropriate

If several questionnaire items are intended to measure the same underlying construct, your research method may require you to assess whether those items form a sufficiently consistent scale.

Cronbach's alpha is one reliability measure commonly available in SPSS.

A typical SPSS path is:

Analyze → Scale → Reliability Analysis

You can place the relevant items into the analysis and select the appropriate reliability model.

However, do not calculate Cronbach's alpha simply because it appears in another project.

Whether reliability analysis is necessary, how it should be conducted and how the result should be interpreted depend on your questionnaire and research design.

In some projects, reliability testing is discussed in Chapter Three rather than introduced for the first time in Chapter Four.

Follow the structure approved by your supervisor.

Step 12: Choose a Statistical Test That Matches Your Hypothesis

If your project contains hypotheses, descriptive statistics such as percentages and means may not be enough.

Depending on the research question and variables involved, possible tests include:

• Chi-square test
• Correlation
• Independent-samples t-test
• Paired-samples t-test
• ANOVA
• Regression
• Non-parametric alternatives

Do not choose a test because:

• Another student used it.
• It looks easy in SPSS.
• You already know how to click the menu.
• It gives a significant result.

Your test should follow from:

• Your research question
• Your hypothesis
• Your variable types
• Your research design
• The assumptions of the statistical test
• Your approved Chapter Three methodology

Step 13: Run a Chi-Square Test When Appropriate

A chi-square test of independence can be used in appropriate situations to examine an association between categorical variables.

For example, you might investigate whether two categorical variables are associated.

In many SPSS versions:

Analyze → Descriptive Statistics → Crosstabs

Then:

1. Place one categorical variable in Row.
2. Place the other in Column.
3. Click Statistics.
4. Select Chi-square.
5. Continue.
6. Open Cells.
7. Select the counts and percentages required for your analysis.
8. Run the procedure.

Do not stop after seeing the p-value.

Check the output and make sure the assumptions of the test are reasonable for your data.

If SPSS reports problems involving expected cell counts or your table contains very small groups, discuss the appropriate interpretation or alternative method with your supervisor.

Step 14: Understand the p-Value Before Writing Your Decision

Many projects test hypotheses using a significance level such as:

α = 0.05

If 0.05 is the significance level approved in your methodology, a common decision framework is:

• p ≤ 0.05: reject the null hypothesis
• p > 0.05: fail to reject the null hypothesis

But do not write:

The null hypothesis is accepted.

when the correct statistical interpretation is usually that you failed to reject it based on the available evidence.

Also avoid treating statistical significance as proof that an effect is large, important or practically meaningful.

Report what your analysis actually supports.

Step 15: Write the Results in Chapter Four

SPSS output is not your finished Chapter Four.

You need to convert the relevant output into clean tables and explanations that answer your research questions.

Example: Frequency Result

Table 4.1 shows that 58 respondents, representing 58.0% of the sample, were female, while 42 respondents, representing 42.0%, were male.

Example: Mean Result

Suppose your approved method uses a criterion mean of 2.50.

You might write:

The respondents agreed that social media improves access to academic information, with a mean score of 3.18, which was above the criterion mean of 2.50.

Only write this if those are your actual results and the decision rule was established in your methodology.

Example: Hypothesis Test

Suppose the approved significance level is 0.05 and your analysis produces p = 0.021.

You might write:

The test produced a p-value of 0.021, which was below the 0.05 significance level. The null hypothesis was therefore rejected.

You should also report the relevant test statistic and other information required by your department or discipline.

Do not invent values to complete a Chapter Four table.

Use the actual values produced from your real dataset.

How to Structure Chapter Four

The exact structure varies by institution, but a questionnaire-based Chapter Four may contain sections such as:

4.1 Introduction

Briefly explain what the chapter presents.

For example:

This chapter presents the analysis of data collected for the study. The results are organised according to the research questions and hypotheses.

4.2 Response Rate

If relevant, show:

• Number of questionnaires distributed
• Number returned
• Number valid for analysis
• Response rate

4.3 Demographic Characteristics

Present relevant characteristics such as:

• Gender
• Age group
• Department
• Level of study

Do not include demographic variables that have no relevance simply to make the chapter longer.

4.4 Analysis of Research Questions

Present the tables and results that answer each research question.

A useful structure is:

Research Question One

Then:

1. Table
2. Result
3. Brief interpretation

Repeat for the remaining research questions.

4.5 Test of Hypotheses

If your project includes hypotheses, present each hypothesis and its approved statistical test.

4.6 Summary of Findings

Where your department requires it, briefly summarise the major findings before moving to the discussion chapter.

Follow your institution's approved project format.

How to Export SPSS Results

You do not need to retype every number from your SPSS output.

Depending on your SPSS version, output can be copied or exported into formats suitable for further formatting.

Before placing a table in your project:

• Remove unnecessary columns.
• Use clear headings.
• Check decimal places.
• Remove irrelevant SPSS terminology.
• Make sure values have not changed during copying.
• Follow your department's table format.

Raw SPSS screenshots are usually less readable than properly formatted project tables.

Use screenshots only where your department specifically requires them.

Common SPSS Mistakes Students Make

Analysing Before Cleaning the Data

An incorrect entry can affect frequencies, means and statistical tests.

Check the data first.

Using the Wrong Codes

If 4 = Strongly Agree, maintain that meaning throughout the dataset unless you intentionally reverse-code an item.

Forgetting What Each Variable Means

Use clear variable labels so that you do not confuse q12, q13 and q14 later.

Treating Every Likert Item as One Scale

Questions that happen to use the same response options do not automatically measure the same construct.

Using 2.50 Without Explaining It

If your methodology uses a criterion mean, show how the decision point was obtained and use it consistently.

Running a Statistical Test That Does Not Match the Hypothesis

Your research question determines the analysis.

The SPSS menu does not.

Reporting Only the p-Value

Depending on your discipline and approved reporting style, other information such as the test statistic, degrees of freedom, sample size or effect size may also be relevant.

Changing Responses to Obtain a Preferred Result

Never modify genuine responses simply because your hypothesis was not supported.

A result that differs from your expectation is still a research result.

Copying Raw SPSS Output Directly Into Chapter Four

Clean your tables and explain what the important results mean.

Inventing Missing Data

If a respondent did not answer a question, do not create an answer for them.

Handle missing values according to your research method.

Use the Right Tool for Your Analysis

Not every questionnaire project requires complex statistical software.

If your analysis only requires frequency and percentage, you can use MonoEd's Simple Percentage Calculator.

If your approved method requires mean scores for Likert-scale items, you can use the Likert Scale Mean Calculator.

Use SPSS or another approved statistical package when your study requires more detailed analysis.

The important thing is not which software looks more advanced.

It is whether the analysis correctly answers your research questions.

Frequently Asked Questions (FAQs)

How do I analyse questionnaire data in SPSS for Chapter Four?

Create and code your variables, enter or import the questionnaire responses, clean the dataset, then run the descriptive or inferential statistics specified in your research methodology.

Afterward, convert the relevant output into clear tables and interpret each result according to your research questions and hypotheses.

How do I analyse Likert-scale questionnaire data in SPSS?

First code the response categories consistently.

Depending on your approved methodology, you may use frequencies and percentages for individual responses or calculate statistics such as the mean and standard deviation.

If several items are intended to form one scale, determine whether they should legitimately be combined before calculating a total or average score.

How do I calculate frequency and percentage in SPSS?

In many SPSS versions, go to:

Analyze → Descriptive Statistics → Frequencies

Select the required variables and run the procedure.

The output can show how many respondents selected each category and the corresponding percentages.

How do I calculate mean and standard deviation in SPSS?

One common route is:

Analyze → Descriptive Statistics → Descriptives

Select the appropriate numeric variables and request the mean and standard deviation.

Only use these statistics when they fit your measurement approach and approved methodology.

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

If your scale is coded 4, 3, 2, 1, its arithmetic midpoint is 2.50.

Some projects therefore use 2.50 as a criterion mean.

However, this should not be treated as a universal rule. Follow the decision rule established in your methodology.

Do I need Cronbach's alpha before analysing questionnaire data?

Not every questionnaire requires Cronbach's alpha.

It may be appropriate when several items are intended to measure the same construct and your methodology requires an assessment of their reliability.

Follow the method approved for your particular questionnaire.

Can I use SPSS for only 30 respondents?

SPSS can analyse a dataset containing 30 respondents.

Whether 30 respondents are sufficient for your research is a separate question that depends on your population, sampling method, research design and intended statistical analysis.

The fact that SPSS can calculate a result does not mean the sample size is appropriate.

Can I enter questionnaire data directly from Google Forms?

Yes.

You can export responses from Google Forms, clean and organise the spreadsheet, then import the prepared data into SPSS.

Always verify the imported variables and values before analysis.

What should I do if SPSS shows missing values?

Check the original questionnaire or form response first.

Determine whether the value is genuinely missing or whether an entry error occurred.

Do not guess the missing response.

Handle missing data according to the procedure approved for your study.

Which statistical test should I use for my questionnaire?

There is no single test for every questionnaire.

The appropriate test depends on your research question, hypothesis, variables, research design and statistical assumptions.

Do not choose chi-square, correlation, t-test or ANOVA simply because another project used it.

Can I use another student's SPSS output for my Chapter Four?

No.

Your analysis should come from the actual data collected for your own study.

Using another person's output would not represent your respondents, questionnaire or research findings.

Can AI analyse my questionnaire data for me?

AI can help explain statistical concepts, organise your workflow or help you understand output, but the analysis must use your real data and an appropriate method.

You should verify the calculations and be able to explain the method and results during your project defence.

Final Checklist Before Writing Chapter Four

Before you begin writing your results, confirm that:

• Your data comes from your actual respondents.
• Each variable is coded correctly.
• Missing responses have been identified.
• Reverse-coded items have been handled correctly where applicable.
• Your analysis matches Chapter Three.
• Your tables answer the research questions.
• Statistical tests match the hypotheses.
• Your decision rules were established before interpreting the results.
• You have not altered responses to obtain preferred findings.
• Every number reported in Chapter Four can be traced back to your actual analysis.

SPSS can perform calculations quickly, but it cannot decide whether your research design is appropriate.

Your research questions, methodology and actual data should determine the analysis.