The curriculum as
a map, not a line.
Every topic of the course sits in a conceptual domain: data, relationships, inference, survey statistics, mathematical foundations. Each one links to the module that teaches it and the labs that let you explore it. Start anywhere.
- Interactive lab
- Course module
Data & measurement
What a variable is and how it can be summarised.
- 01
Measurement levels
Nominal, ordinal, scale, and why level is not the same as role.
- 02
Index construction
Reverse scoring, Cronbach's alpha, cutting the index.
- –
Distributions
Shape, centre and spread; theoretical families and empirical histograms.
- 09
Descriptive statistics · Variance
Deviations become squares, squares become variance.
Relationships
How two or three variables move together.
- 03
Crosstabs
Row vs column percentages, expected counts, χ², Cramér's V.
- 06
Correlation
Pearson vs Spearman and what r cannot tell you.
- 04
Lazarsfeld paradigm
X → Y when you control for Z; Simpson's paradox.
- –
Regression
Predict with a line; minimise the squared residuals; check the model.
Inference
From a sample to a claim, with its uncertainty.
- –
Sampling & standard error
Draw repeated samples; watch the sampling distribution form.
- –
Confidence intervals
Coverage as a thing you can count.
- 05
p-values
Null distribution, tail areas, sample size, power.
- 07
t-tests
One-sample, independent, paired: choose and read.
- 08
ANOVA
SS total = SS between + SS within; F, η², p.
Survey statistics
Designs, weights and design-based standard errors.
- –
Sampling designs
Strata, clusters and PPS; who could have been selected.
- –
Inclusion probabilities & weights
From π to design weights, nonresponse and post-stratification.
- –
Calibration & raking
Match known population margins; watch the weights move.
- –
Complex survey inference
Taylor linearisation, replicate weights, design effects.
Mathematical foundations
The tools behind the methods.
- –
Probability
Events as areas; conditional probability and Bayes.
- –
Expected value
The balance point of a distribution and the law of large numbers.
- –
Functions & rates of change
Slope as a thing you drag; area under a curve.
- –
Optimization
Why the minimum of SSE is where the derivative is zero.
- –
Vectors, matrices · PCA
Least squares as a projection; rotate the data to find the most variance.
Modules follow the Understand → Try → SPSS → Test rhythm. Labs compute every value live from the data in front of you.