Populations, censuses and samples
Studying this course? Keep your progress in one place.
Key facts
- A population is the whole set of interest; a census observes every member; a sample observes some.
- A sampling frame lists every sampling unit in the population.
- A census is exact but slow and expensive; a sample is cheaper but has sampling variability.
- A statistic is calculated from the sample only; a parameter describes the population.
- A statistic is a random variable with its own sampling distribution.
Population, census and sample
A census measures every member of the population; a sample measures a selection.
A population is the whole set of items or people of interest, such as all households in a council area. A census observes or measures every member. A sample is a selection of members, used to find out about the whole.
The population need not be people: it could be all the fuses made in a day or all the bags of flour in a warehouse.
Population
- The whole set of interest
- e.g. all 4200 households
Census
- Every member observed
- Exact population values
Sample
- A selection of members
- Used to learn about the whole
A list of every household in a council area is an example of a:
Sampling units and frames
The frame lists the whole population; the sample is the part chosen from it.
A sampling unit is a single member of the population that can be selected, such as one household or one fuse. A sampling frame is a list of all the sampling units, such as an electoral register or the serial numbers to .
A good frame is complete and has no duplicates. A random sample is chosen by numbering the units and selecting numbers with a random number generator.
- 1
List the frame
Every sampling unit
- 2
Number the units
1 to N
- 3
Generate random numbers
Ignore repeats
- 4
Select those units
They form the sample
Is the sampling frame the same as the sample?
Census or sample?
A census is exact but slow and sometimes impossible; a sample is cheaper but varies.
A census gives the exact population values and has no sampling variability, but it is expensive and slow, and impossible if testing destroys the item or the population is infinite. A sample is cheaper and quicker, but its results are subject to sampling variability and may be biased.
In an exam, always refer to the context: testing fuses until they blow destroys them, so a census would leave none to sell.
| Census | Sample | |
|---|---|---|
| Cost and time | Expensive and slow | Cheaper and quicker |
| Accuracy | Exact population values | Sampling variability, may be biased |
| Destructive testing | Impossible | Possible |
Census
- Cost and time:
- Expensive and slow
- Accuracy:
- Exact population values
- Destructive testing:
- Impossible
Sample
- Cost and time:
- Cheaper and quicker
- Accuracy:
- Sampling variability, may be biased
- Destructive testing:
- Possible
Why is a sample used to test how long light bulbs last?
Statistics and parameters
A statistic uses only sample values; anything needing an unknown parameter is not a statistic.
A parameter is a fixed number describing a population, such as , or the proportion . Parameters are usually unknown.
A statistic is any quantity calculated only from the values in a random sample. The sample mean and the sample maximum are statistics. The expression is not a statistic because is unknown.
Parameter
- Describes the population
- Fixed, usually unknown
- , ,
Statistic
- Calculated from the sample only
- A random variable
- , sample maximum
Which of these is a statistic? ( is unknown.)
Sampling distributions
It is the distribution of a statistic over all possible samples of a given size.
Different random samples give different values of a statistic, so a statistic is a random variable. Its distribution over all possible samples of a given size is its sampling distribution.
To build one, list every possible sample, find the statistic for each, then total the probabilities of equal values. Draw two discs with replacement from , , : the larger value has , and the sample mean has the distribution shown.
In the disc example, find for the larger value .
Try an exam question
A factory makes fuses each day and tests them by running them until they blow. (a) Give two reasons why a sample is used rather than a census. (b) Suggest a suitable sampling frame.
[4 marks]
- [1]Testing destroys the fuses, so a census would leave none to sell.
- [1]A sample is cheaper and quicker.
- [1]A sampling frame is a list of all the fuses made that day.
- [1]For example, the serial numbers to .
That's the notes covered.
Carry on to the next subtopic.
Try a question
The register of households is an example of
1 mark, marked against the mark scheme
The register of households is an example of
1 mark, marked against the mark scheme