Free explainer2 min read
Sampling methods
Understand the model, recognise it in a stem, separate the look-alikes, then apply it.
Start with the mental model
Why does the same jar of sweets yield a different 'sample' depending on how you scoop it? Sampling is how you grab a fair handful from the jar. For example, imagine a jar of coloured sweets: if every sweet has the same chance of being picked, that's simple random; if you make sure each colour is represented, that's stratified; if you scoop whole pre-packed bags instead of individual sweets, that's cluster. In practice, this means the method is named by HOW the selection happens, not by how 'random' or fair the result looks.
01Core model
• Probability: simple random (equal chance), systematic (every kth), stratified (within subgroups), cluster (whole groups), multistage.
• Non-probability: convenience, quota, purposive, snowball.
• Sampling frame = list of the population.
• For example, a study of people who inject drugs has no accessible sampling frame, so it must use snowball sampling, reaching new participants through referrals from existing ones.
• In practice, this means the sampling frame's existence (or absence) is often what decides which method is even possible.
• Large national morbidity surveys, such as the Adult Psychiatric Morbidity Survey, use stratified sampling to guarantee proportionate representation of population subgroups.
• A study's power is its ability to detect a true difference if one really exists, given its sample; a larger sample size generally increases power.
02Memory anchor
Stratified = sample within layers; Cluster = scoop whole groups; Snowball = friend-of-a-friend.