Sampling is the process of measuring a small number of sites or people in order to obtain a perspective on all sites and people.
Why do geographers use sampling?
- Sampling is quicker
- Sampling is cheaper
- Often it is impossible to access whole population
A sample needs to be representative of the whole population. Representative means how closely the characteristics
of the sample match the characteristics of the population
An unrepresentative sample is biased. In a biased sample, some elements of the population are less likely to be included than others. In Geography fieldwork, times of day, week and year, the choice of locations to collect data, and the weather can all lead to bias.
There are three methods of sampling to help overcome bias. These are:
- systematic sampling
- random sampling
- stratified sampling
Systematic sampling
In a systematic sample, measurements are taken at regular intervals, e.g.
- every 5th person who walks past
- every 20 metres along a street
- every 50 metres along a beach
Random sampling
In a random sample, each member of the population is equally likely to be included in the sample.
For taking random samples of an area, use a random number table to select numbers. Use pairs of numbers as x and y co-ordinates. You could use metre rule interval markings (e.g. to take pebble samples on a beach) or grid references (e.g. to find random samples in a city).
Stratified sampling
This is when the population is split into could have sub groups. In a stratified sample, a proportionate number of measurements are taken is taken from each group.
For example, an urban ward may contain 8 deprived wards and 2 undeprived wards. A random sample may by chance miss all the undeprived areas. By contrast, with a stratified sample, you can make sure that 80% of your samples are taken in the deprived areas and 20% in the undeprived areas.