How unit 5 is examined
This unit has one topic, case studies, where the data analytics tools and methods of Units 1-4 are applied to a real problem; it has not been asked in recent papers.
Case studies
<span style="display:inline-block;padding:.16em .6em;border:1.5px solid currentColor;border-radius:999px;font-size:.68em;font-weight:700;letter-spacing:.06em;text-transform:uppercase;opacity:.75">Not asked since 2022</span>
Definition. <mark>A data analytics case study applies the analytics framework (collect, pre-process, explore, model, visualise, interpret) to one real problem to reach a decision.</mark>
Key points.
- A case study begins by stating the business problem and the data available, such as sales, sensor or customer records.
- The data is cleaned by handling missing values and outliers, then explored with statistics, correlation and charts.
- A model such as regression or a probability-based method is built in R, MATLAB or Python, and its accuracy is checked.
- The results are shown as visualisations and turned into a recommendation, for example forecasting sales or predicting customer churn.
Last-minute revision
- Case study means the analytics framework applied end to end to one real problem.
- Steps: problem, data collection, pre-processing, exploration, modelling, visualisation, decision.
- Pre-processing handles missing values and outliers before any model is built.
- Regression predicts a value; correlation measures the strength of a relationship.
- R, MATLAB and Python are the tools used to run the analysis.
- Typical examples are sales forecasting and customer churn prediction.
- The last step is always a business recommendation, not only a chart.
Memory hooks
- P-D-C-E-M-V-D: Problem, Data, Clean, Explore, Model, Visualise, Decide.
- Case study equals the whole pipeline on one real problem.
- Clean before you model.
Coverage checklist
- Case studies: no past questions in the supplied papers; covered by the definition and four key points above.