SPSS is the most commonly used statistical software in Ugandan universities. It is also the tool that terrifies the most students. The interface looks complicated, the menus are overwhelming, and one wrong click can produce meaningless results. This guide will change that.
Getting Started: The SPSS Interface
When you open SPSS, you see two main views: Data View and Variable View. Data View looks like an Excel spreadsheet — this is where your actual data lives. Variable View is where you define what each column means: variable names, types, labels, and measurement levels.
Before entering any data, always start in Variable View. Name your variables clearly (no spaces), add labels for clarity, set the correct type (numeric for numbers, string for text), and specify the measurement level (nominal, ordinal, or scale).
Entering and Cleaning Your Data
Data entry in SPSS is straightforward — type directly into the Data View cells. But before analysis, you must clean your data. Check for missing values, outliers, and impossible entries. Use Transform > Recode Into Same Variables to fix coding errors.
Always create a backup of your raw data file before making any changes. Save your work frequently and use descriptive filenames like "SurveyData_Cleaned.sav."
Step 1: Descriptive Statistics
Every analysis starts with understanding your data. Go to Analyze > Descriptive Statistics > Frequencies for categorical variables. For continuous variables, use Analyze > Descriptive Statistics > Descriptives or Explore.
Look at the mean, standard deviation, minimum, and maximum values. Do they make sense? If your age variable has a maximum of 999, you have a data entry error.
Step 2: Check Reliability (Cronbach's Alpha)
If you are using a questionnaire with multiple items measuring the same concept, you need to check reliability. Go to Analyze > Scale > Reliability Analysis. Move your related items into the Items box.
Cronbach's Alpha should be above 0.70 for acceptable reliability. If it is below 0.60, your scale is unreliable and needs revision. Remove problematic items one at a time and re-run until you reach acceptable levels.
Step 3: Normality Testing
Many statistical tests assume your data is normally distributed. Check this using Analyze > Descriptive Statistics > Explore. Look at the Shapiro-Wilk test for samples under 50, or Kolmogorov-Smirnov for larger samples.
If your p-value is above 0.05, your data is normally distributed. If below 0.05, you have non-normal data and should use non-parametric tests instead.
Step 4: Correlation Analysis
To test relationships between variables, go to Analyze > Correlate > Bivariate. Select your variables, check Pearson for normally distributed data or Spearman for non-normal data.
Look at the correlation coefficient (r) and significance (p-value). A significant positive correlation means as one variable increases, so does the other. Report both the coefficient and the p-value in your results.
Step 5: Regression Analysis
Regression tells you how much one variable predicts another. Go to Analyze > Regression > Linear. Your dependent variable goes in the Dependent box, independent variables in the Independent(s) box.
Look at the R-squared value — it tells you what percentage of variance in your dependent variable is explained by your independent variables. Check that your p-values are significant and your VIF values are below 10 (no multicollinearity).
Exporting Your Results
SPSS output can be exported to Word or PDF using File > Export. Choose the format you need. For your dissertation, copy relevant tables into your document and format them according to your university's guidelines.
"SPSS is not scary. It is just a tool. Learn the basics, practice with your own data, and you will be analyzing like a pro in no time." — CareerCraft UG