Sampling

What is a target population?

• A group of people who share a given set of characteristics which a researcher wishes to draw conclusions from.

• A target population is too large so a subset of the population is investigated.

• Allows the findings to be generalised to the target population.

• Then we have a bias sample. Sampling size important > sampling error likely to occur. The larger the sample the more likely to provide an accurate estimate about the nature of the population from which its been drawn (less likely to be bias).

What is a sample?

• Allows the findings to be generalised to the target population.

• A group of people who share a given set of characteristics which a researcher wishes to draw conclusions from.

• A target population is too large so a subset of the population is investigated

• Then we have a bias sample. Sampling size important > sampling error likely to occur. The larger the sample the more likely to provide an accurate estimate about the nature of the population from which its been drawn (less likely to be bias).

Why does the sample need to be representative of the target population?

• A target population is too large so a subset of the population is investigated.

• Allows the findings to be generalised to the target population.

• Then we have a bias sample. Sampling size important > sampling error likely to occur. The larger the sample the more likely to provide an accurate estimate about the nature of the population from which its been drawn (less likely to be bias).

• a group of people who share a given set of characteristics which a researcher wishes to draw conclusions from.

What happens if the sample isn't representative? importance of Sampling size?

• Then we have a bias sample.

• A target population is too large so a subset of the population is investigated.

• Sampling error likely to occur. The larger the sample the more likely to provide an accurate estimate about the nature of the population from which its been drawn (less likely to be bias).

• Sampling size is a separate sample away from the target population, used as a controlled variable.

Opputunity sampling

• Bias = High (not everyone in the target population has chance to be selected- only those available at the time of selection)

• Generalisability- medium as the sample chose won't fully represent the target population as not everyone had the opportunity to be selected.

• Ppts selected using those people who are most easily available. E.g a friend. Easiest method to use.

• Bias- Low as all members of the Target population have an equal chance of being selected, researchers have to have access to details of ALL members of the target population.

• Generalisability- low (this sample NOT representative of the target population)

Systematic sampling

• Accessing the target population and putting them into an order (e.g alphabetical order). The researcher then selects every nth participant from the list of participants who are available e.g selecting every 5th person.

• Bias- Low as all members of the Target population have an equal chance of being selected, researchers have to have access to details of ALL members of the target population.

• Ppts are selected by either adding all data to a computer + allowing computer programme randomly selecting the sample. OR each Ppts allocated a number which is put in a hat etc.

• Generalisability- medium as the sample chose won't fully represent the target population as not everyone had the opportunity to be selected.

• Bias- medium as not all participants have an equal chance of being selected.

Random sampling

• Bias- medium as not all participants have an equal chance of being selected.

• Ppts are selected by either adding all data to a computer + allowing computer programme randomly selecting the sample. OR each Ppts allocated a number which is put in a hat etc.

• Generalisability- low (this sample NOT representative of the target population)

• Bias- Low as all members of the Target population have an equal chance of being selected, researchers have to have access to details of ALL members of the target population.

• Generalisability- high due to random selection from the entire target population. Sample will represent the key characteristics etc.