Sampling
Samples & Populations
- Sampling is used to obtain the participants who will take part in the research
- A sample is taken from the population that is relevant to the research topic
- A sample may be taken from a target population
- Often a sample is taken from a population which is more generalised than a target population:
- a researcher who wishes to investigate the effects of being a single teen parent will require their sample to be drawn from that specific population
- a researcher who is running an experiment on the duration of STM does not need such a specific target population as the one outlined above (generally anyone from the age of 18-60 would suffice)
- The researcher draws the sample from the target population to take part in the experiment and then generalises the findings across the target population
- Researchers use different sampling techniques depending on varying factors within their study, as each sampling technique has both strengths and limitations:
Method & evaluation of different sampling types
| Sampling Type | How it is selected | Strengths | Limitations |
| Stratified |
Small-scale reproduction of the target population, dividing and categorising the population by characteristics important to the research (strata), e.g. age, gender, education etc. Then the population is sampled within each category proportionate to the overall amount - if the whole population has 18% of people who are male aged 30-40, then the representative sample will have 18% of people who are male aged 30-40. |
The sample is representative of the target population. Easy to generalise as the sample is representative. Selection is unbiased as it is based on the sub groups in society. |
Time consuming, knowing the subgroups and dividing the population into categories, then selecting participants to match these, can be very time consuming. The researcher requires knowledge of the subgroups and categories of the population, which may not be available. |
| Random |
Selecting people, in a way that everyone has a fair chance of being selected, this could be done by having all names in a hat and pulling them out or by using a computer generator. |
Unbiased selection, meaning it is more likely to be a representative sample. As the results are fairly representative, it means the results are able to be generalised to the target population. |
Time consuming and impractical, it is not always a possibility to be able to have all of the information on a target population, or, for all of the target population to want to take part in the study . It may be non-representative as all of one gender could be selected randomly, which would not be a true example of the target population. |
| Volunteer (also known as self selected) | People volunteer (choose) to take part, selecting themselves as participants of a study often by replying to adverts. |
Willingness of participants as they are choosing to take part so participants are less likely to want to jeopardise the study and its results. Not much effort is required to obtain the sample e.g. put up a poster, put out an ad on social media etc. |
Volunteer bias, which means the results will not be able to be generalised as volunteer participants often have personality traits in common e.g. sociable, out-going etc. Volunteers are often eager to please, which causes demand characteristics, which means that they may behave how they think the researcher wants them to |
| Systematic | Selecting every nth person from a list to make a sample. The researcher has to calculate how many people they need to know which nth they need |
Unbiased selection, meaning it is more likely to be a representative sample. As the sample is likely to be reasonably representative it means that the results can be generalised to the target population. |
Not always truly unbiased as the selection process can interact with a hidden periodic trait, if every 10th person is a 19 year old shop worker, then they are the only people in the sample. It may be non representative as all of one gender could be selected systematically, which would not be a true example of the target population. |
| Opportunity | Selecting those who are most convenient, willing and available to take part. This could be asking people who are passing in the street to take part. |
Natural experiments use opportunity sampling as the researcher has no control over who is being studied. Quick and easy way to get information as it is using people who are readily available to use. |
Cannot generalise as the sample is like to be unrepresentative, as it is likely to miss whole sections of the population, students could be at school for example or people who are at work can not be in the sample. It is a self selected sample as participant have the option to agree or decline to join in at the time of the study |
