How To Use Spss For Data Analysis In Your Final Year Project

How to Use SPSS for Data Analysis in Your Final Year Project

SPSS, which stands for Statistical Package for the Social Sciences, is the software most Nigerian supervisors expect you to use when your final year project involves numbers. If you collected questionnaire responses and now need to turn them into tables, percentages and test results for Chapter Four, SPSS is the tool that does that work. The good news is that you do not need to write code or be good at mathematics to use it. SPSS works through menus and dialog boxes, so you point, click and read the output.

This guide walks you through the whole process in order, from getting SPSS on your computer to reporting your results. Follow the steps in sequence and you will go from a pile of completed questionnaires to a clean set of results your supervisor can trust. Before you start, make sure you already know how many respondents you need. If you have not settled that, read our guide on how to determine sample size for your final year project first, because the number of questionnaires you analyse in SPSS depends on it.

What Is SPSS and Why Do Supervisors Recommend It?

IBM SPSS Statistics is a software application for statistical analysis, data management and data visualisation. It was originally developed in 1968 and has become the standard tool in social sciences, education, business, health sciences and psychology departments around the world. Most Nigerian universities teach SPSS in research methodology courses, and many provide student licences through their computer laboratories or libraries, which means your supervisor probably knows SPSS and can guide you through it.

There are three reasons SPSS suits final year students so well. First, it needs no programming. Unlike R or Python, you do not write a single line of code, because every analysis is done through dropdown menus. Second, it produces output in a structured format that is easy to copy into your project, with tables and charts your examiner will recognise. Third, its results are trusted. SPSS has been validated by decades of academic use, so reviewers and examiners accept its output without question. If your project is quantitative, which means it works with numbers and tests relationships between variables, SPSS is almost certainly the right tool. If your project is qualitative, which means it works with words and meanings, read our guide on qualitative vs quantitative research for final year projects to confirm which approach fits your study before you open SPSS at all.

How to Get Access to SPSS

Before you pay for anything, check your school first. Most universities and polytechnics in Nigeria have an institutional licence, which means you can use SPSS for free in the school computer laboratory or library. Ask your department or the ICT unit where the licensed computers are and what the opening hours are. Some schools also give students a licence to install on their own laptop, so ask about that too.

If your school does not have a licence, you can download SPSS Statistics directly from the IBM website, and IBM offers a free trial that is enough to practise with for a few weeks. Plan your analysis around that window, so you do your learning and your real analysis before the trial ends. Whichever route you take, install SPSS and open it once before your data is ready, just to confirm it works. Do not discover on the day of your analysis that the software needs an update or a licence code you do not have.

Understand the Two Main Views in SPSS

When SPSS opens, you will see two tabs at the bottom of the screen: Data View and Variable View. Understanding the difference between them is essential before you enter a single number.

Data View is your spreadsheet. Each row is one person or one record, for example one student who answered your questionnaire. Each column is one piece of information about that person, such as their gender, their age or their score on a scale. If you surveyed 200 students, Data View will have 200 rows when you finish entering your data.

Variable View is where you describe each column before you fill it with data. Here you tell SPSS the name of the variable, its type, its label, the values its codes stand for, and its measurement level. Setting this up properly is the most important preparation you will do, because mistakes here cause errors in every analysis that follows.

Measurement Levels: Nominal, Ordinal and Scale

In Variable View, every variable must have a measurement level, and getting this right decides which analyses SPSS will let you run. There are three levels.

• Nominal means categories with no order. Gender is nominal, because Male and Female are categories and neither one ranks above the other. State of origin and department are also nominal.
• Ordinal means categories with an order. Class of degree is ordinal, because First Class ranks above Second Class Upper, which ranks above Second Class Lower, and so on. A Likert scale response, from Strongly Disagree to Strongly Agree, is usually treated as ordinal, although many researchers treat summed Likert scores as scale.
• Scale means a real number you can do maths with. Age in years, CGPA, weekly study hours and test scores are scale variables. You can calculate a mean and a standard deviation for them, which you cannot do meaningfully for nominal categories.

If you set a variable to the wrong level, SPSS may refuse to run your test or, worse, run it and give you a meaningless result. Check each variable in Variable View before you move on.

How to Use SPSS for Data Analysis in 10 Steps

Follow these steps in order. They take you from raw questionnaire responses to results you can write up in Chapter Four.

Step 1: Prepare Your Data for Entry

Before you touch SPSS, number your questionnaires from 1 to your total, for example 1 to 200, and keep them in that order. Numbering lets you trace any strange value in SPSS back to the exact questionnaire it came from. Decide your coding scheme on paper first. For example, for gender, Male is 1 and Female is 2. For a Yes or No question, Yes is 1 and No is 2. For Likert scale items, Strongly Disagree is 1, Disagree is 2, Undecided is 3, Agree is 4 and Strongly Agree is 5. Write the scheme down and keep it beside you while you work.

Step 2: Import Your Data Into SPSS

If you collected responses with Google Forms, Qualtrics or another online survey tool, export the responses as a CSV or Excel file. Whenever possible, export in SPSS format with the .sav extension, because that saves a lot of time by automatically setting up the variables with the correct labels and coded values. If you have a CSV or Excel file, import it through File, then Import Data, then Excel or CSV. SPSS can import both, but if your dataset is in CSV format, it is recommended that you first save it as an Excel workbook and then import it. If you used paper questionnaires, type the data directly into Data View, one row per respondent. Save your file immediately after importing, using File, then Save As, to store it in SPSS format. SPSS is not cloud based and does not autosave, so save often as you work.

Step 3: Set Up Your Variables in Variable View

Switch to Variable View and define each variable. Give it a short name with no spaces, such as Gender, StudyHours or CGPA. Set its type, usually Numeric for coded numbers. Write a clear label, such as "Gender of respondent", so your output tables are easy to read. Under Values, define what each code means, for example 1 is Male and 2 is Female. Finally, set the Measure to Nominal, Ordinal or Scale as explained above. For our running example of a survey of 200 Nigerian university students looking at study habits, side hustles and CGPA, you would create variables called State, Gender, StudyHours, CGPA and SideHustle, with SideHustle coded 1 for Yes and 2 for No.

Step 4: Code Your Categorical Variables

String variables, which hold text, cannot be used in most statistical analyses, so your categorical variables need numeric codes. If your data arrived as text, such as "Male" and "Female" written out, use the Automatic Recode function under Transform to convert them to numbers. Manual coding gives you more control over the scheme, but Automatic Recode is faster when you have many categories. Check after recoding that every category received the code you intended, because a miscoded value will silently distort your results.

Step 5: Clean Your Data Before You Analyse It

A dataset may contain incomplete responses, irrelevant answers, responses from outside your target population, or outliers, and these issues can seriously affect your analysis and threaten the validity of your findings. Before you run any test, run Frequencies on each variable through Analyze, then Descriptive Statistics, then Frequencies, and scan the output for anything that looks wrong. Look for values outside the possible range, such as an age of 250 or a Likert response of 9. Look for missing values and decide how to handle them, either by excluding those cases or by following your supervisor's guidance. Check for duplicate rows. Fix errors now, while you can still find the original questionnaire, because cleaning after analysis means redoing every test.

Step 6: Check Reliability With Cronbach's Alpha

If you used multi item Likert scale questions, you need to check their internal consistency before calculating a total score. Run a reliability analysis in SPSS and look at Cronbach's alpha, which tells you whether the items in your scale hang together and measure the same thing. Many textbooks and supervisors treat a value of 0.70 and above as acceptable reliability. If your alpha is far below that, check whether any item needs reverse coding, which means flipping its scores so that agreement points in the same direction as the other items, and run the analysis again. Do this before you calculate scale scores, because a score built from an unreliable scale is not trustworthy.

Step 7: Calculate Your Scale Scores

If your scales are reliable, the next step is to calculate a total score for each respondent. This is usually done in one of two ways, by summing the items or by calculating their average. Different scales may require different methods, so check the scale's manual for guidance. If the manual is not available, using the average is generally the safer option. In SPSS, use Transform, then Compute Variable, to create the new score. Give the new variable a clear name and label, such as MotivationScore, so you can find it easily in later steps.

Step 8: Run Descriptive Statistics

Always start your analysis with descriptive statistics. They give you a snapshot of your data and help you spot errors or outliers you missed during cleaning. For categorical variables, such as gender or department, create frequency tables that show counts and percentages. For continuous variables, such as scale scores, use numerical measures such as minimum, maximum, mean and standard deviation. In SPSS, go to Analyze, then Descriptive Statistics, then Descriptives for the numbers, and Frequencies for the categories. Always include descriptive statistics in your results chapter. They provide the foundation for everything that follows, and your examiner will expect to see them before any test result.

Step 9: Run the Right Inferential Test

Descriptive statistics summarise your data, but they cannot tell you whether a relationship is real or just chance. That is the job of inferential statistics, which let you draw conclusions about a larger population from your sample. The test you choose depends on your research question and on the type of variables involved. Here are the tests final year students use most often, with the menu path for each one.

• Correlation, for a relationship between two scale variables. To check whether study hours relate to CGPA, use Analyze, then Correlate, then Bivariate, and select both variables. SPSS gives you a correlation coefficient between minus 1 and plus 1. A value near 1 means a strong positive relationship, a value near minus 1 means a strong negative relationship, and a value near zero means little or no relationship.
• Independent samples t-test, for comparing two groups. To compare the mean CGPA of male and female students, use Analyze, then Compare Means, then Independent Samples T Test. This test compares the means of two separate groups on one scale variable.
• One-way ANOVA, for comparing three or more groups. To compare mean scores across three or more groups, such as students from different departments or levels, use Analyze, then Compare Means, then One-Way ANOVA.
• Chi-square test, for two categorical variables. To test whether gender is related to side hustle participation, use Analyze, then Descriptive Statistics, then Crosstabs, and request the chi-square statistic. This test works with counts in categories, not with means.
• Linear regression, for predicting one variable from others. To predict CGPA from study hours and other factors, use Analyze, then Regression, then Linear. Regression tells you how much each predictor contributes and how well the set of predictors explains your outcome.

Before running an inferential test, it is good practice to verify that your data meets the assumptions of the procedure. SPSS includes tests to check these assumptions, such as the Kolmogorov Smirnov test for normality under Analyze, then Descriptive Statistics, then Explore, and Levene's test for equality of variances, which appears automatically in t-test and ANOVA output. If you are unsure which test fits your question, ask your supervisor before you run it, because choosing the wrong test is one of the most common reasons results get rejected at defense.

Step 10: Report Your Results in Chapter Four

After completing your analysis, you need to report the results in a standard format, usually APA style, depending on your academic requirements. SPSS generates a large amount of output, but you do not need to report everything. Report only the key information: your descriptive statistics, the test you ran, the test statistic, the degrees of freedom where relevant, and the significance level, which is the p value. Most Nigerian projects test hypotheses at the 0.05 level of significance, which means a p value below 0.05 is treated as statistically significant, while a p value of 0.05 or above is not. Present each result in a clean table, then write a paragraph under it that interprets the table in plain language, stating what the numbers mean for your research question. That interpretation is what earns marks, not the table alone. The results you report here become the evidence for the recommendations you will write in Chapter Five, so if you have not planned those yet, read our guide on how to write recommendations for your final year project to see how findings turn into action.

Common Mistakes Students Make in SPSS

• Setting the wrong measurement level. If SPSS treats a scale variable as nominal, or a categorical variable as scale, your test may fail or produce nonsense. Check Variable View before you analyse.
• Skipping data cleaning. One mistyped value can shift a mean or break a test. Run Frequencies on every variable and fix errors before you test anything.
• Computing scale scores before checking reliability. If Cronbach's alpha is low, your total score is not trustworthy. Check reliability first, reverse code where needed, then compute the score.
• Reporting all of the SPSS output. Your examiner does not need every table SPSS produced. Report the key numbers only, and interpret each one in your own words.
• Copying output tables without interpretation. A table with no explanation earns few marks. Every table in Chapter Four needs a paragraph that says what it shows and what it means for your study.
• Using string variables in tests. Text values cannot be analysed by most procedures. Code your categories as numbers before you run any test.
• Forgetting to save your work. SPSS does not autosave. Save your .sav file after importing, after cleaning, and after every major step, and keep a backup copy on a flash drive or in your email.
• Choosing a test because it looks impressive. Pick the test that matches your research question and your variable types, not the most complicated one. A simple, correctly chosen test beats a fancy, wrong one every time.

What to Do When SPSS Gives an Error

Most SPSS errors come from one of three causes, and all three are easy to fix. First, a variable may be the wrong type or measurement level, which stops a procedure from running. Go to Variable View, correct the type and the measure, and try again. Second, your data may contain a value the procedure cannot handle, such as text in a numeric column or a blank where a number is expected. Run Frequencies to find the offending case, fix it in Data View, and rerun the analysis. Third, you may have selected the wrong variables for the procedure, such as putting a categorical variable where SPSS expects a scale variable. Check the dialog box, move the variables to the correct boxes, and run it again. When an error persists, read the message carefully, because SPSS usually tells you exactly which variable or value caused the problem. If you still cannot fix it, take a screenshot of the error and the dialog box to your supervisor or to the ICT staff in your school laboratory, who will usually spot the problem in seconds.

Frequently Asked Questions

Do I need to know statistics before using SPSS?

You need a basic understanding of what each test does and when to use it, but you do not need to calculate anything by hand. SPSS does the computation. Your job is to set up the data correctly, choose the right test for your research question, and interpret the output accurately. If your statistics knowledge is thin, study the meaning of mean, standard deviation, correlation and p value before your analysis, and confirm your choice of test with your supervisor.

Can I use SPSS for a qualitative project?

SPSS is built for numbers, so it is the wrong tool for a purely qualitative project that works with interview transcripts and themes. If your study mixes both approaches, use SPSS for the quantitative part, such as questionnaire data, and analyse the interview data separately by hand or with a qualitative tool. Confirm your approach in your methodology chapter so your examiner can see why each tool was chosen.

How long does SPSS data analysis take for a final year project?

For a typical survey of 100 to 300 respondents, a careful first timer can finish the full workflow in a few days, including data entry, cleaning, descriptive statistics, tests and reporting. Data entry usually takes the longest, especially for paper questionnaires. You can shorten the work by collecting responses online, so the data arrives already in a spreadsheet you can import, and by cleaning as you go instead of leaving it all to the end.

Is SPSS free for students?

SPSS is commercial software, but most Nigerian universities hold an institutional licence that lets students use it free of charge in the school computer laboratory or library. IBM also offers a free trial for a few weeks, which is enough to complete a final year analysis if you plan your work in advance. Check with your department or ICT unit first, before you consider paying for a personal licence.

Conclusion

SPSS turns the most feared part of a final year project into a manageable, step by step task. You bring your data in, set up your variables, clean the dataset, check reliability where needed, describe your data, run the test that matches your question, and report the key results with a clear interpretation. None of that needs programming, and none of it needs advanced mathematics, because the software does the calculation while you do the thinking. The thinking is the part that earns marks: choosing the right test, reading the output correctly, and explaining what the numbers mean for your research question. Start early, save often, clean before you test, and confirm your choice of test with your supervisor. Do that, and Chapter Four stops being a source of fear and becomes the chapter where your hard work finally speaks for itself.

Further reading: For the full step by step workflow this guide is based on, see the DataDoing guide on steps in data analysis for a research project using SPSS, the Help In Writing guide on SPSS data analysis for research students, and the Teachers Institute guide on essential steps in data analysis using SPSS.