4. Purposeful sampling strategies
• Extreme cases
(outliers)
• Maximum variation
• Homogeneous groups
• Typical/average case
• Critical case
• Snowball
• Criterion
• Theory-based
• Confirming &
disconfirming cases
• Stratified purposeful
• Opportunistic/emergen
t
• Purposeful random
5. Extreme cases / outliers
• Cases that have gone extremely well or extremely wrong
– i.e. excellence or extreme failure
• Illuminative of other, less extreme instances
6. Maximum variation sampling
• Pick a few cases from different conditions: eg:
• very poor, middle class, very rich
• very large, medium, small (town, class, college)
• Identify shared patterns across these variations
7. Homogeneous groups
• Small group of similar cases to describe in depth
• Used to gain information about a particular subgroup (i.e.
single dads)
• Used in focus group research
8. Typical/average case
• Identified with the help of key informants or previous
research
• What criteria do you use to define typical?
• e.g. Typical science classroom/prof
9. Critical case sampling
• If it happens here, it will happen everywhere
• e.g. if conservative group adopts new technology, every
other group will
10. Snowball / chain sampling
• Ask participants: who should I talk to? Who knows a lot
about… Who does… ?
11. Criterion sampling
• Participants must meet certain criteria
• EG:
• age requirements
• weekly users of Facebook, Twitter, AND LinkedIn
• teach mathematics at x level
12. Theory based sampling
• A type of criterion sampling
• The criteria are derived from theory
• EG:
• early adopters (diffusion of innovations)
• high communication apprehension
• spatial learners
13. Confirming & disconfirming cases
• Confirming cases support your analysis
• Disconfirming cases contradict your analysis, prompt you
to look for alternate explanations
• Used later in the research process, after patterns have
emerged
• EG
• non-traditional students who do learn better when using the Smart
Pen
• non-traditional students who perform worse when using the Smart
Pen
16. Purposeful random sampling
• Used to select a small number of cases to study from a
large pool of available cases that qualify
• EG:
• typical case sampling: pick 5 out of 50 cases
17. Sample size in qualitative research
• Sample to the point of redundancy
• For IRB purposes, specify a safely large number