Sampling and bias
In a nutshell
Sampling is where data is collected about a group in order to make predictions about another much larger group. If a sample is biased, this means that it does not properly represent the larger group and the predictions are likely to be incorrect.
How does sampling work?
When a population is large, it is difficult to survey it entirely so samples of smaller groups are taken instead.
Definitions
Population | The large group that you want to find out information about. The group could be anything from humans to plants to animals.
|
Sample | Data collected about a smaller group taken from the population.
|
Random sample | Every member of the population has an equal chance of selection. |
In order to be able to draw accurate conclusions from a sample, it must be representative of the whole population.
Characteristics of a representative sample
1.
| It must be a random sample. |
2.
| It must be big enough: the larger the sample, the more reliable it is. |
Identifying bias in a sample
Bias happens when a sample is not representative, meaning either the sample is not random or is too small.
Definitions
Biased sample
| A sample that does not accurately represent the wider population.
|
Tip: To decide whether or not a sample is random, think about when, where and how the sample is taken and who is in it!
Example 1
Kate wants to find out information about her village. She takes a sample of 100 women in the village - this sample is not random as it excludes the men.
Example 2
Oliver wants to find out information about how many foxes there are in his garden. He counts the number he sees during the day - this sample is not random as it was only taken during the day where there is a much smaller chance of seeing a fox.
Excluding a certain group from a sample can lead to bias due to different things , such as age or gender.