Sampling Methods and Bias

Four main sampling methods

1. Simple Random Sampling

Every individual has an equal chance of being selected. Like drawing names from a hat.

• Pro: Unbiased, easy to understand.
• Con: May miss important subgroups.

2. Stratified Sampling

Divide the population into strata (subgroups) based on a characteristic (age, gender, etc.), then randomly sample from each stratum proportionally.

• Pro: Ensures all subgroups are represented.
• Con: Requires knowledge of population structure.

3. Cluster Sampling

Divide the population into clusters (geographic areas, classrooms, etc.), randomly select some clusters, then sample everyone in the chosen clusters.

• Pro: Practical when the population is spread out.
• Con: Higher sampling error if clusters are not representative.

4. Systematic Sampling

Select every $k$th individual from a list (e.g., every 10th person). Start from a random point.

• Pro: Simple to implement.
• Con: Can be biased if the list has a periodic pattern.

Sampling bias

Bias occurs when the sample systematically differs from the population. Common sources:

• Selection bias: Some groups are under/overrepresented.
• Response bias: People don't answer truthfully.
• Non-response bias: People who don't respond differ from those who do.

Try it in the visualization

A grid of 1000 dots represents the population. Each sampling method highlights different dots. Switch between methods to see how each selects differently.