Course
Data Analytics: Statistics and Visualisation
Describe data with the right numbers, test what the numbers claim, and draw charts that tell the truth, each lesson with code you can run.
- Lessons
- 16
- Reading time
- 4 h 6 min
- Level
- Beginner
- Taught in
- English
- Format
- Articles with runnable code
Course
What you will learn
Sixteen article lessons in four parts. Descriptive statistics: types of data, mean, median and mode, spread, percentiles and outliers. Distributions: shape and skew, the normal distribution, binomial, Poisson and exponential, sampling and the central limit theorem. Relationships and inference: correlation, association between categories, confidence intervals and hypothesis tests. Then visualisation: choosing a chart, drawing it with matplotlib and seaborn, designing it honestly, and an analysis from question to report. Every lesson builds its own small dataset, so the code runs as written.
- Descriptive statistics
- Distributions
- Relationships and inference
- Visualisation
Before you start
A computer that runs Python. Each lesson says what it builds on.
Syllabus
Syllabus
Every lesson explains one idea, then runs it in code and shows the output. The first lesson is open to everyone.
01 · Descriptive statistics
- 01Types of dataNumerical and categorical, discrete and continuous, nominal and ordinal: what each type allows you to compute and plot.Free preview17 min
- 02Mean, median and modeThree answers to “what is typical?”: how each is computed, how outliers and skew pull them apart, and which one to report.14 min
- 03Measures of spreadRange, variance, standard deviation and the interquartile range: how far values sit from the centre and why n − 1 appears.15 min
- 04Percentiles, z-scores and outliersRank values with percentiles and z-scores, read a five-number summary and box plot, and decide what to do with an outlier.14 min
02 · Distributions
- 05Distributions, histograms and skewSee the shape of a variable with histograms and density plots: bins, modality, skewness, kurtosis and log scales.14 min
- 06The normal distributionThe bell curve, the 68–95–99.7 rule, standardising values, and checking normality with Q–Q plots and tests.15 min
- 07Binomial, Poisson and other distributionsCounts of successes, events per hour, waiting times and skewed amounts: the distributions behind everyday data, with SciPy.15 min
- 08Sampling and the central limit theoremPopulations and samples, sampling bias, the standard error, and why sample means turn normal whatever the data looks like.15 min
03 · Relationships and inference
- 09CorrelationCovariance, Pearson and Spearman correlation, correlation matrices, and the many ways correlation is not causation.17 min
- 10Association between categoriesContingency tables, row and column percentages, the chi-square test and Cramér’s V, and Simpson’s paradox.15 min
- 11Confidence intervalsReport a range, not just a number: intervals for a mean and a proportion, the bootstrap, and what “95% confident” really means.16 min
- 12Hypothesis testingNull and alternative, p-values, t-tests and their non-parametric cousins, errors of both kinds, effect sizes and multiple testing.16 min
04 · Visualisation
- 13Choosing a chartMatch the chart to the question: comparison, trend, distribution, relationship or composition, and when a table beats a chart.15 min
- 14Charts with matplotlib and seabornBar, line, scatter, histogram, box, violin and heatmap charts, small multiples with facets, and styling a figure for a report.17 min
- 15Designing honest chartsAxes that start where they should, colour that carries meaning, clutter removed, annotations that make the point, and charts that mislead.15 min
- 16From question to report, end to endOne dataset, one business question: describe, compare, test, chart and write a short report a manager can act on.16 min
Feedback
Help us improve this course
Suggest a lesson, point out something unclear, or report a mistake in the code. A person reads every message and replies by email.