What this lesson teaches
I can justify a sample size, how to select it, when to collect data and which methods to use.
Syllabus 8834, Fieldwork, 1. Pre-fieldwork stage. Students should be able to develop a plan that: establishes the primary and secondary data needed to examine the question/hypothesis posed; identifies appropriate methods to determine sample size, select sample and collect data.
Make a guess
Students survey 100 people at one bus stop at lunchtime on a weekday. Is this better than surveying 20 people across different times and places?
- No, because a survey works best with 20 people.
- Yes, because lunchtime is when everyone is out.
- No. A bigger sample at one spot repeats the bias.
- Yes, because 100 is five times as many as 20.
Show the answer
No. A bigger sample at one spot repeats the bias.
Who you reach depends on where and when you ask. Workers and people who rarely go out are missed however many are asked at that stop.
A bigger sample reduces some uncertainty, but it cannot fix a biased way of choosing people.
The population is everyone you want to learn about. A sample is the smaller group you actually study. Your sampling frame is the list you choose from. A list of community centre visitors misses people who rarely go out. Say so.
Random sampling gives everyone on the list a known chance of being picked. Systematic sampling picks every nth person or place. Stratified sampling splits the population into groups, like age bands or areas, then samples within each.
Convenience sampling picks whoever is easiest to reach. It is quick, but may miss workers, people who cannot get around, or people who speak another language. Purposive interviews pick people with relevant experience. They give depth, but not population percentages.
Choose sample size based on how varied people are, which groups you compare and your resources. Cover different days and times. Asking 100 people at the same convenient spot is no more representative than asking 20.
- Sampling frame
- The list you choose your sample from.
- Stratified sampling
- Split into groups, then sample each.
- Bigger is not fairer
- More people from a biased method stay biased.
Step by step
Define the frame
Which streets, households, sites or times?
Choose and justify
How does it capture real variety?
Report limits
Group counts, timing and refusals.
Worked example: A better fix
A survey at one park bench at 10 am mainly reaches park users. Add chosen housing blocks, different times and an accessible interview option. This helps far more than asking 50 more people at the same bench.
Watch out for this
Equal numbers from each group always represent the population.
Equal numbers help compare groups. For population totals, you may need proportional samples.
Check your understanding
A weekday lunchtime survey misses many working residents. Which change fixes this?
- Use a longer questionnaire at lunchtime.
- Ask twice as many people at the same lunchtime.
- Add evening and weekend sampling times to the plan.
Show the answer
Add evening and weekend sampling times to the plan.
Yes. It reaches people at work during the day.
Check your understanding
Which sampling method ensures residents from each age group are included in proportion to the population?
- Stratified sampling by group.
- Systematic sampling of every tenth person passing.
- Convenience sampling.
Show the answer
Stratified sampling by group.
It divides the population into groups and samples each in proportion.