What is Sampling Methods?
Sampling methods divide into probability (every member of the population has a known, non-zero chance of selection) and non-probability (selection is judgmental or convenient). Probability methods — simple random, systematic, stratified, cluster — support inferential statistics and confidence intervals. Non-probability methods — convenience, judgment, quota, snowball — are faster and cheaper but cannot be generalized to the population without assumptions. The choice depends on whether the decision needs population-level estimates or directional reads.
How Sampling Methods actually works
The framework breaks down into the following moving parts. Knowing what each piece is — and what it is not — is what separates a B-grade answer from an A-grade answer in a written assignment.
- Simple random — every member equally likely; needs a complete sampling frame
- Systematic — every kth element from an ordered list
- Stratified — random sample within demographic strata, then weight
- Cluster — sample groups (cities, schools), then everyone within
- Convenience — whoever is easy to reach (mall intercept, employee panel)
- Judgment / quota — recruit to fill specific cells (e.g., 50 men, 50 women)
A worked example: Gallup election polling
Gallup's presidential polls use stratified random sampling on the US adult population, with strata defined by region, age, race, gender, and education. Each completed interview is weighted to bring the sample back into national proportion. The 2016 polls famously under-sampled non-college whites, demonstrating that even probability designs can fail when the sampling frame or weights miss a key stratum. The fix in 2020 was to add education as an explicit stratum.
Don't lose marks for these
- Confusing convenience samples with random samples
- Reporting a margin of error on a non-probability sample (mathematically meaningless)
- Failing to weight a stratified sample back to population proportions
How to use this on the exam
Score-maximizing moves
- Justify probability vs non-probability based on the decision's precision needs
- Specify the sampling frame and acknowledge its limits
- Compute or quote the right sample size for the desired confidence level
When to use Sampling Methods (and when not to)
Use Sampling Methods when your assignment asks you to analyze, structure, or recommend — and when you have at least two data points to populate every cell of the framework. Skip it when the question is asking for a numerical answer or a single recommendation, since Sampling Methods is a structuring tool, not a calculator.