Data Analysis & Visualisation

This module develops the end-to-end data-analysis workflow a researcher needs to move from a raw dataset to a defensible visual argument. Learners progress from exploratory data analysis and cleaning to statistical summarisation, then on to the grammar of graphics and the design of figures that read correctly in both print and screen media. The emphasis is on reproducible analysis in which every figure and table is regenerable from documented code and source data. By the end, learners can take an unfamiliar dataset, interrogate it responsibly and present honest, publication-ready visuals.

Learning Outcomes

Perform exploratory data analysis using summary statistics, missing-value audits and univariate visual checks.

Clean and reshape tabular data to tidy format using Pandas (melt, pivot, merge, groupby).

Compute and interpret descriptive statistics including measures of centre, spread, skewness and correlation.

Select appropriate chart types for comparison, distribution, composition and relationship questions.

Design accessible figures using colour-blind-safe palettes, legible typography and clear axis labelling.

Produce publication-quality graphics with Matplotlib, Seaborn and Plotnine using the grammar of graphics.

Build small interactive dashboards for exploratory review using Plotly or Streamlit.

Evaluate charts for honesty, including scale manipulation, cherry-picked baselines and misleading aggregation.

Topics Covered

Exploratory data analysis workflow and the Tukey tradition

Tidy data principles and reshaping with melt, pivot and stack

Data cleaning: types, parsing dates, strings and categoricals

Missing data: patterns, imputation strategies and audit trails

Descriptive statistics: mean, median, quantiles, IQR, standard deviation

Correlation, rank correlation and visualising relationships

The grammar of graphics: data, aesthetics, geometries, scales, facets

Chart choice: comparison, distribution, composition, relationship

Matplotlib object-oriented API and Seaborn statistical plots

Plotnine (ggplot2 in Python) for layered graphics

Colour theory for data: perceptual uniformity, colour-blind-safe palettes

Typography, annotation and gridline discipline for figures

Interactive visualisation with Plotly and dashboards with Streamlit

Chart honesty: axis manipulation, baselines, dual axes, Simpson's paradox

Reproducible figure pipelines with scripted builds and version control

Skills Developed

Exploratory data analysis Data cleaning and tidying Statistical summarisation Chart selection and design Grammar-of-graphics thinking Accessible visualisation Interactive dashboarding Critical reading of charts

Suggested Resources

Wickham, H. & Grolemund, G. (2017). R for Data Science, O'Reilly Media

Wilke, C.O. (2019). Fundamentals of Data Visualization, O'Reilly Media

Tufte, E.R. (2001). The Visual Display of Quantitative Information, 2nd ed., Graphics Press

Cairo, A. (2019). How Charts Lie: Getting Smarter about Visual Information, W.W. Norton

VanderPlas, J. (2016). Python Data Science Handbook, O'Reilly Media

Waskom, M.L. (2021). Seaborn: statistical data visualization, Journal of Open Source Software

Official Pandas, Matplotlib, Seaborn and Plotly user guides

Assessment

Assessment rewards the ability to take an unfamiliar dataset through a reproducible analysis pipeline and present honest, well-designed visuals, with a mix of portfolio evidence and a timed practical paper.

Dataset investigation report with reproducible figures (2500 words incl. code)45%

Timed practical visualisation examination and MCQ paper35%

Portfolio of three chart-redesign and dashboard exercises20%

Practice bank: 199 MCQs · Timed mock & self-paced modes

Start Practice →