Stratified and cluster sampling

This lesson focuses on stratified and cluster sampling. Mathematics provides tools for modeling and analyzing complex systems.

Statistical inference uses samples to draw conclusions about populations. A population includes every member of the group being studied; a sample is a subset. Random sampling gives every member equal chance of selection—reduces bias. Types of probability samples: Simple random sample (each member equally likely), Stratified (divide into subgroups/strata, sample from each), Cluster (divide into groups, randomly select entire groups), Systematic (every nth member). Bias occurs when the sample systematically misrepresents the population: voluntary response bias (only motivated people respond), convenience bias (nearby/accessible sample), nonresponse bias (some groups systematically don't respond). Experimental design: a controlled experiment randomly assigns subjects to treatment/control groups to establish causation. Observational studies can show correlation but not causation—confounding variables may explain the relationship.

As you engage with this material, consider both the theoretical foundations and practical applications. Think critically about how this concept builds on prior knowledge and where you might apply it beyond the classroom.

Challenge yourself to go beyond memorization—seek to understand the "why" behind the processes and principles.