Outlier sampling focuses on the extremes – the end-points of the normal distribution bell-curve.
Outlier sampling studies cases that are unusual or special in some way, such as outstanding successes or notable failures. Many programs can have ‘best’ sites of implementation. Studying why these sites are different can provide insight into both what is unique to that case and also what is typical and shared with other sites.
Examples
An example of a notable failure that served as a purposive evaluation sample includes the 9/11 terrorist attacks in the United States. Congress established the National Commission on Terrorist Attacks Upon the United States, an independent, bipartisan commission, to evaluate the circumstances of the terrorist attacks, including preparedness for and the immediate response to the attacks.
Advice for choosing this method
Cross-check your case with others in the field, experts or participants to determine that the case is extreme.
Advice for using this method
Include as much descriptive detail as possible to illustrate how the outlier case is different from the normal case. Consider what can be learnt from the outlier case.
This page is a Stub (a minimal version of a page). You can help expand it. Contact Us to recommend resources or volunteer to expand the description.
'Outlier sampling' is referenced in:
Blog
Framework/Guide
- Rainbow Framework :