Stratified Sampling

Introduction: Stratified Sampling

Stratified Sampling is a statistical method in which a population is divided into smaller, distinct subgroups known as strata. Random samples are then taken from each stratum to ensure that all key groups are proportionally represented. This approach improves the accuracy and fairness of results compared to simple random sampling, especially when the population is diverse.

Background

Stratified Sampling was developed to address the limitations of purely random sampling in heterogeneous populations. When important subgroups differ significantly, random selection alone may overlook or underrepresent them. By dividing the population into meaningful strata based on shared characteristics, researchers ensure balanced representation and more reliable estimates. This method is widely used in social sciences, healthcare, and market research to produce unbiased and comprehensive results.

Key Elements / Features

• Strata Definition: Subgroups are defined by common attributes such as age, gender, income, education, or region.
• Random Selection Within Strata: A random sample is taken from each subgroup to maintain objectivity.
• Proportional or Equal Sampling: Sampling can reflect the size of each stratum (proportionate) or assign equal weight regardless of size (disproportionate).
• Improved Accuracy: Reduces sampling error by ensuring that all segments of the population are properly represented.

Applications / Examples

• Education: Selecting students from different year levels to ensure each group is included.
• Healthcare: Sampling patients by age or health condition to compare outcomes across categories.
• Market Research: Dividing consumers by income or geographic region to obtain representative opinions.
  

Example: In a survey of 1,000 employees, a company divides staff by department and randomly selects individuals from each one, ensuring all departments contribute to the sample.

Relevance / Impact

Stratified Sampling increases representativeness and statistical precision. It helps reduce bias and improves the generalisability of results across diverse populations. Although it requires detailed information about the population to define strata accurately, it remains one of the most effective sampling methods for balanced and dependable research outcomes.

See also

• Random Sampling
• Sampling Methods
• Sample Size
• Simple Random Sampling