Experiments for Statisticians
Design of experiments is the planning of studies so that cause and effect can be estimated cleanly. It relies on randomisation, replication and blocking, and uses layouts such as completely randomised, factorial and block designs analysed by ANOVA. The same ideas power A/B tests, clinical trials and agricultural studies.
Stage: Design and Advanced Methods · Part of the Statistician roadmap · Last reviewed 19 November 2026
What to learn in Experiments
- Randomise treatment assignment and explain why it removes confounding
- Analyse a one-way and two-way ANOVA and interpret the F-test
- Use blocking to reduce noise from a known nuisance variable
- Plan a factorial design and read interaction effects
- Size an A/B test and avoid peeking at results early
Experiments interview questions
Why does randomisation matter in an experiment?
Random assignment makes treatment groups comparable on average for both measured and unmeasured factors, so a difference in outcomes can be attributed to the treatment rather than to confounders. Without it, observational differences may reflect who chose or received the treatment.
What is an interaction effect in a factorial design?
An interaction means the effect of one factor depends on the level of another. For example, a teaching method may help beginners but not experts. Factorial designs can detect this, while changing one factor at a time cannot.
Where Experiments fits in the Statistician roadmap
Experiments belongs to the Design and Advanced Methods stage (advanced level). Real studies.
Other units in this stage
Back to the full Statistician roadmap