The selection method determines whose experiences your sample can represent.
A sample is the people or sites selected for a study. The target population is the whole group your question concerns. A sampling frame is the list or arrangement used to select them. If your question concerns all residents, a list of park visitors will miss non-users.
Convenience sampling chooses whoever is easiest to reach, such as people passing one entrance. Quota sampling sets target numbers for groups, then selects people non-randomly within them. Both can be practical, but the interviewer's choices or survey location may favour some experiences over others. This is sampling bias.
Simple random sampling uses a random process so every eligible member of the frame has an equal chance of selection. Stratified random sampling first divides the population into relevant groups, then selects randomly within each group.
Suppose 60% of residents live near the park and 40% live farther away. A proportional stratified sample of 50 could include 30 nearby and 20 farther-away residents. Selection within both groups must still be random. Choosing the easiest people to reach until these totals are filled would instead be quota sampling.
Every method has limits. The frame may omit a group, selected people may not respond, or answers may be inaccurate. A single survey time can also miss certain users. Asking more people at the same entrance may repeat the same bias. Match the improvement to the problem: a wider resident list, for example, can help include non-users.
Step by step
Define the population
If the question concerns residents, park visitors alone are not the whole group.
Choose the method
Explain how it includes relevant groups and fits the time, people and equipment available.
Record who is missing
Note non-response, times not covered and groups absent from the sampling frame.
Worked example: A specific repair
A lunchtime entrance survey misses residents at work and people who never visit. Repeating it at another time helps time coverage, but still misses non-users. A resident-based sampling frame addresses that separate problem more directly.
Watch out for this
Quota sampling is random because it includes more than one group.
Group coverage and random selection are different features. Quotas set totals; stratified random sampling also uses random selection within the groups.
Check your understanding
Students divide a resident list into two distance groups and use a random-number generator within each. What method is this?
- Convenience sampling
- Quota sampling only
- Stratified random sampling