Stratified Sampling
When researchers use random sampling, they cannot choose important subpopulations on purpose. Stratified sampling, on the other hand, enables researchers to include key subpopulations in a sample by dividing the whole population into several categories. For example, when anthropologists need to select men and women of all age groups, they use stratified sampling. First, researchers divide a population into men and women, then into four age groups: 15-29, 30-44, 45-59, and over 59. This process produces 8 subgroups of the whole population. Next, researchers randomly select people from each of the eight categories. Stratified sampling is effective when each subgroup has unique characteristics of its own, and researchers study these patterns.
Additionally, there is a similar method called Quota Sampling. In this method, researchers choose individuals from each category for convenience instead of randomly. Although this flexible selection process is not unbiased, quota sampling reflects population characteristics at less cost and time compared to strict stratified sampling.
This page was created by a Minnesota State University, Mankato student. Last updated 11/14/04.