It is similar to stratified sampling but has key differences.
Key Concepts
Clusters vs. Strata:
Both techniques involve dividing populations into groups.
In Cluster Sampling, groups are called clusters.
Clusters are often naturally occurring, rather than manually divided.
How it Works
Cluster Formation: Population is divided into clusters naturally based on certain factors like location (e.g., California and Washington DC).
Selection Process:
Randomly select a few clusters.
Sample every member from the selected clusters.
Difference from Stratified Sampling
In stratified sampling:
Strata are formed.
A few members are sampled from every group.
In cluster sampling:
All members from a few selected clusters are sampled.
Visual Representation
Consider black circles as clusters with various members.
Clusters may be divided by time or space for convenience.
Randomly select clusters for complete member sampling.
Application Example
Students at Tables:
Tables represent clusters in a classroom.
Instead of sampling a few students randomly across the room, select entire tables and sample all students at those tables.
Sampling Process:
Randomly sample six students using cluster sampling.
Choose two tables (clusters) out of ten.
Sample all students from the selected tables.
Conclusion
Cluster sampling is beneficial when clusters are split by time or space, making it logistically easier to sample all members from some clusters rather than a few members from each cluster across the entire population.