Selection bias arises when a sample is not representative of the broader population, leading to skewed conclusions.
Bias may be unintentional.
Examples in Action
- Surveying only engaged users
- Testing changes on narrow segments
- Analysing incomplete datasets
- Excluding low-intent traffic
Typical Outcomes / Results
- Misleading insights
- Poor generalisation of results
- Increased decision risk
- Improved accuracy when addressed
This definition reflects standard statistical bias concepts.