Data Analysis for Your Thesis: A Free Course Without SPSS
Take one realistic student survey from raw spreadsheet to finished results chapter. You will clean messy data, check reliability and assumptions, choose and run the right tests, and report every result in APA style, using free tools that run in your browser instead of SPSS.
What you will learn
- Turn research questions into hypotheses and choose the right statistical test for each one
- Estimate the sample size you need with a power analysis
- Clean and code raw survey data: duplicates, missing-value codes, spelling, reverse-coded items and scale scores
- Describe your sample and build a Table 1 with means, standard deviations and frequencies
- Check scale reliability with Cronbach’s alpha and test normality and outliers
- Run and interpret t-tests, ANOVA, Mann-Whitney, Kruskal-Wallis, chi-square, correlation, multiple and logistic regression and mediation
- Report every result in APA 7 style and assemble a complete results chapter
Who this course is for
Undergraduate, master’s and PhD students who must analyse quantitative data for a thesis, dissertation or research project, and anyone who wants to learn applied statistics by doing it. No previous statistics course is needed: each lesson explains the idea before you run the test. If you already know the basics, the lessons on assumptions, effect sizes and APA reporting will still sharpen your results chapter.
How the course works
Each lesson explains one step of the method in plain language and then sends you to a free tool on this site with the data already loaded, so you practise on a real analysis as you read. A short quiz at the end of every lesson checks your understanding. Create a free student account to save your progress on every device. After the last lesson, pass the final assessment with 70% or more to earn a certificate of completion that you can download as a PDF at any time and that anyone can verify online. There are no fees, no deadlines and nothing to install.
Course files
- student-survey-raw.csv: the messy survey export used in Lesson 2 (243 rows, 18 variables).
- student-survey-clean.csv: the cleaned dataset used from Lesson 3 onwards (240 students, with reverse-coded item and scale scores).
Syllabus
- 1Plan your analysis before you collect dataTurn research questions into hypotheses, identify variables and their measurement levels, choose tests and work out the sample size you need.15 min
- 2Clean and code your survey dataFind and fix duplicates, missing-value codes, inconsistent spelling, impossible values and reverse-worded items, then compute scale scores.20 min
- 3Describe your sample and build Table 1Summarise categorical variables with frequencies and continuous variables with means, standard deviations, medians and skewness.15 min
- 4Check the reliability of your scalesCompute Cronbach’s alpha, read item-total statistics and see what happens when a reverse-worded item is not recoded.15 min
- 5Check normality and outliersTest normality with Shapiro-Wilk and Q-Q plots, screen for univariate and multivariate outliers, and decide what to do about them.15 min
- 6Compare two groups: t-tests and Mann-WhitneyRun independent and paired t-tests, check Levene’s test, interpret effect sizes and use the Mann-Whitney U test for skewed outcomes.20 min
- 7Compare three or more groups: ANOVA and Kruskal-WallisRun a one-way ANOVA with post-hoc tests, understand why the omnibus test and pairwise tests can disagree, and use Kruskal-Wallis with Dunn’s tests.20 min
- 8Analyse categorical data with the chi-square testTest the association between two categorical variables, check expected counts, read adjusted residuals and report Cramér’s V.15 min
- 9Correlation and multiple regressionBuild a correlation matrix, then predict exam performance from several variables, interpret B, β and R², and check the regression assumptions.25 min
- 10Logistic regression and mediationPredict a yes/no outcome with odds ratios, then test whether social support affects performance indirectly through stress.25 min
- 11Write your results chapter in APA styleStructure the results chapter, format numbers, statistics and tables in APA 7, and assemble the course findings into a complete write-up.20 min
- ✓Final assessment and capstone12 questions on the whole course; 70% unlocks your certificate.20 min
Tools you will use
Frequently asked questions
Do I need SPSS or any other software?
No. Every exercise runs in a free tool on this site, in your browser. The tools produce the same statistics as SPSS (they are tested against SPSS-equivalent routines in SciPy, statsmodels and pingouin) and format them as APA tables.
Can I use my own thesis data instead of the course dataset?
Yes. Work through each lesson with the course data first so you can compare your output with the numbers in the text, then repeat the step with your own data. Your data never leaves your browser.
What is the course dataset?
A realistic, simulated survey of 240 university students with gender, age, faculty, weekly study hours, sleep, a four-item stress scale, a four-item social support scale, counselling use and midterm and final exam scores. It is built to show the problems and results you will meet in real thesis data.
How do I get the certificate?
Complete all eleven lessons (finishing a lesson quiz marks the lesson complete) and score at least 70% in the final assessment. The certificate page then lets you add your name, print or save it as a PDF and add it to LinkedIn.
Is this enough for my methods and results chapters?
It covers the analyses most quantitative theses use and shows how to report them. Always follow your university’s and supervisor’s guidance on which tests and formats they expect.