Sampling
Studying this course? Keep your progress in one place.
Key facts
- The population is the whole group; a sample is the part you observe; a census observes everyone.
- The sampling frame is the list the sample is drawn from.
- In a simple random sample every member, and every group of size , is equally likely to be chosen.
- Opportunity sampling takes whoever is available, so it is often biased.
- Different samples give different results (sampling variation), so a sample statistic is only an estimate.
Population and sample
The population is the whole group of interest. A sample is the part you actually observe, drawn from a sampling frame.
A census observes every member, but may be too costly or slow, or may destroy the items tested (such as testing how long bulbs last).
The list from which a sample is drawn is the sampling frame. Using the sample to say something about the population is an informal inference: the sample statistic is an estimate, not the true population value.
- 1
Population
the whole group of interest
- 2
Sampling frame
the list the sample is drawn from
- 3
Sample
the part you actually observe
- 4
Informal inference
the sample statistic estimates the population value
Population
- The whole group of people or items
- Not the numbers you record
Sample
- The part actually observed
- Drawn from the sampling frame
Census
- Observes every member
- Costly, slow, may destroy items
A college has 1200 students. Ten students are chosen from the college register and asked about travel. What is the sampling frame?
Simple random sampling
In a simple random sample of size , every member has the same chance of selection and every possible group of is equally likely.
A simple random sample is free from bias in the selection, and every member has the same chance.
Its disadvantages: it needs a complete list of the population, may be slow and costly for a large population, and a random sample can still be unrepresentative by chance.
- 1
Number the frame
give every member of the sampling frame a number
- 2
Generate random numbers
use a calculator, computer or random number table
- 3
Match members
pick the members with those numbers
- 4
Reject
discard repeats and numbers outside the range
- 5
Continue
carry on until n different members are chosen
Advantages
- Free from bias in selection
- Every member has the same chance
Disadvantages
- Needs a complete list
- Slow and costly for a large population
- May still be unrepresentative by chance
A random number is repeated while choosing a sample. What should you do?
Opportunity sampling
Opportunity sampling takes the members available when you collect the data. It is quick and cheap but often biased.
In opportunity sampling (convenience sampling) you take whoever is available, for example the first 30 people to walk past.
Those present at that time and place may differ from the rest of the population. A shopping centre on a Saturday morning, for example, misses people working then.
- 1
Take whoever is available
for example the first 30 people to walk past
- 2
Sample is not chosen at random
not every member can be chosen
- 3
May differ from the population
people at work on a Saturday are missed
- 4
Biased estimate
the sample may not represent the population
| Simple random | Opportunity | |
|---|---|---|
| Needs a sampling frame | Yes | No |
| Speed and cost | Slower, costlier | Quick, cheap |
| Every member can be chosen | Yes | No |
| Risk of bias | Low | Often biased |
Simple random
- Needs a sampling frame:
- Yes
- Speed and cost:
- Slower, costlier
- Every member can be chosen:
- Yes
- Risk of bias:
- Low
Opportunity
- Needs a sampling frame:
- No
- Speed and cost:
- Quick, cheap
- Every member can be chosen:
- No
- Risk of bias:
- Often biased
Which is the best reason an opportunity sample may be biased?
Different samples differ
Two samples from the same population usually differ. This sampling variation means one sample gives only an estimate.
This sampling variation is why a single sample gives only an estimate. Larger samples are generally more reliable.
One sample may suggest a majority in favour and another may not, so be cautious about strong claims from one sample.
Worked example
Of 24 000 voters, one simple random sample of 200 gives 124 supporters. A second sample of 200 gives 110. Estimate the number of supporters from each.
- 1
First sample: proportion .
- 2
Estimate: .
- 3
Second sample: proportion .
- 4
Estimate: .
Two random samples from the same population give different means. What is the best explanation?
Critiquing a technique
Ask whether there is a frame, who could be left out, whether the sample is big enough and whether it is practical. Always answer in context.
In context, name the group that is left out and say how that could change the result.
To estimate how students travel to college, asking the first 30 arrivals at the entrance is opportunity sampling. It may over-represent those who live nearby. A simple random sample from the college register avoids this, but needs the register and more time.
- 1
Frame
Is there a frame, and can every member be chosen?
- 2
Bias
Who is missing, and why does that matter?
- 3
Size
Is the sample large enough to be useful?
- 4
Practical
Is it practical in cost and time?
A café owner surveys the first 20 customers on a Monday morning about opening hours. Give one reason this may be biased.
Try an exam question
A café owner wants to know customers' views on opening hours. She asks the first 20 customers who arrive on a Monday morning. Name this sampling method, give one reason it may be biased, and describe how to take a simple random sample from a list of 400 loyalty-card customers.
[4 marks]
- [1]Opportunity sampling.
- [1]Biased because evening or weekend customers cannot be chosen, so their views are missing.
- [1]Number the customers 1 to 400 and generate random numbers in that range.
- [1]Reject repeats and out-of-range numbers until the sample size is reached.
That's the notes covered.
Carry on to the next subtopic.
Try a question
Which sampling technique is the researcher using?
1 mark, marked against the mark scheme
Which sampling technique is the researcher using?
1 mark, marked against the mark scheme