Sampling involves taking measurements of individuals, selected from the population of organisms which is being investigated. However sometimes, the effects of chance takeover, and the sample lacks representitiveness. There are several reasons why they might not be, including:
- Sampling bias - the selection process may be biased. The investigators may be making unrepresentative choices, either deliberately or unwittingly.
- Chance - even if sampling bias is avoided the individuals chosen may, by pure chance, not be representative.
The best way to prevent sampling bias is to eliminate, as far as possible, any human involvement in choosing the samples.
This can be achieved by carrying out random sampling. This can be done by:
- divide the study area into a grid of numbered lines
- using random numbers from a table or generated by a computer, obtain a series of coordinates.
- Tae samples at the intersection of each pair of coordinates.
Comments
No comments have yet been made