Stratified Sampling
Definition
Stratified sampling randomly selects cases from each previously defined subgroup.
Stratified sampling divides a population into defined strata and draws a random sample from each stratum.
Simply explained
You first divide the population into groups, such as regions. Then you randomly draw samples from each group.
Why it matters in practice
The term helps to select, develop, evaluate AI systems in a well-founded manner and realistically assess their technical risks.
Example from business
A quality audit draws proportional cases from each region and product group.
Limits and pitfalls
- Incorrectly defined strata, missing weighting, and data leakage in train-test splits distort results.
Origin & history
The method is an established part of sampling theory and improves representation or precision in heterogeneous populations.
Use in mid-sized companies
- Possible application: A quality audit draws proportional cases from each region and product group.
Frequently asked questions
What does stratified sampling mean?
Stratified sampling divides a population into defined strata and draws a random sample from each stratum. What should be considered? Incorrectly defined strata, missing weighting, and data leakage in train-test splits distort results.
Sources
- Sampling Techniques · Cochran, Sampling Techniques
Source checked on 10.10.2026. Editorial application examples are hypothetical; no evidence of real customer results.
Editorially reviewed on: