Stratified Sampling
A method of sampling that involves the division of a population into smaller sub-groups known as strata before sampling.
What is Stratified Sampling?
In stratified sampling, the population is divided into mutually exclusive groups based on a relevant characteristic (e.g., gender, income level). Then, independent probability samples (usually simple random samples) are drawn from within each stratum.
Why Stratified Sampling Matters
It ensures adequate representation of key subgroups, especially small minority groups, and often reduces the overall standard error compared to simple random sampling if the strata are homogenous.
Example
A researcher measuring employee satisfaction divides a company's workforce into strata by department (sales, engineering, HR) and randomly samples 10% of employees from each department.
Common Mistakes
- Creating overlapping strata where individuals could belong to multiple groups.
- Confusing stratified sampling with quota sampling, which does not use random selection within groups.